Research Handbook on Intellectual Property and Technology Transfer 9781788116633, 9781788116626

Citation preview

RESEARCH HANDBOOK ON INTELLECTUAL PROPERTY AND TECHNOLOGY TRANSFER

RESEARCH HANDBOOKS IN INTELLECTUAL PROPERTY Series Editor: Jeremy Phillips, Intellectual Property Consultant, Olswang, Research Director, Intellectual Property Institute and co-founder, IPKat weblog Under the general editorship and direction of Jeremy Phillips comes this important new Research Handbook series of high quality, original reference works that cover the broad pillars of intellectual property law: trademark law, patent law and copyright law – as well as less developed areas, such as geographical indications, and the increasing intersection of intellectual property with other fields. Taking an international and comparative approach, these Research Handbooks, each edited by leading scholars in the respective field, will comprise specially commissioned contributions from a select cast of authors, bringing together renowned figures with up-and-coming younger authors. Each will offer a wide-ranging examination of current issues in intellectual property that is unrivalled in its blend of critical, innovative thinking and substantive analysis, and in its synthesis of contemporary research. Each Research Handbook will stand alone as an invaluable source of reference for all scholars of intellectual property, as well as for practising lawyers who wish to engage with the discussion of ideas within the field. Whether used as an information resource on key topics, or as a platform for advanced study, these Research Handbooks will become definitive scholarly reference works in intellectual property law. Titles in the series include: Research Handbook on the History of Copyright Law Edited by Isabella Alexander and H. Tomás Gómez-Arostegui Research Handbook on Intellectual Property and Climate Change Edited by Joshua D. Sarnoff Research Handbook on Intellectual Property Exhaustion and Parallel Imports Edited by Irene Calboli and Edward Lee Research Handbook on Intellectual Property in Media and Entertainment Edited by Megan Richardson and Sam Ricketson Research Handbook on Intellectual Property and the Life Sciences Edited by Duncan Matthews and Herbert Zech Research Handbook on Copyright Law Second Edition Edited by Paul Torremans Research Handbook on Intellectual Property and Creative Industries Edited by Abbe E.L. Brown and Charlotte Waelde Research Handbook on Patent Law and Theory Second Edition Edited by Toshiko Takenaka Research Handbook on Intellectual Property and Digital Technologies Edited by Tanya Aplin Research Handbook on Intellectual Property and Technology Transfer Edited by Jacob H. Rooksby

Research Handbook on Intellectual Property and Technology Transfer Edited by

Jacob H. Rooksby Dean and Professor of Law, Gonzaga University School of Law, Spokane, Washington, USA

RESEARCH HANDBOOKS IN INTELLECTUAL PROPERTY

Cheltenham, UK • Northampton, MA, USA

© The Editor and Contributors Severally 2020

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Control Number: 2019952345

This book is available electronically in the Law subject collection DOI 10.4337/9781788116633

ISBN 978 1 78811 662 6 (cased) ISBN 978 1 78811 663 3 (eBook)

To Foster—A curious mind is the most beautiful thing

Contents

ix x xi xiii xiv

List of figures List of tables List of contributors Acknowledgments List of abbreviations 1

PART I

Introduction to the Research Handbook on Intellectual Property and Technology Transfer Jacob H Rooksby

1

TECHNOLOGY TRANSFER COMES OF AGE: POLICIES, PRACTICES AND POLITICS

2

University technology transfer structure and intellectual property policies Jennifer Carter-Johnson

3

The politics of university technology transfer Jessica A Sebeok

41

4

Bayh-Dole beyond patents Shubha Ghosh

69

5

University as knowledge-based enterprise: organizational design and technology transfer Jarrett B Warshaw

92

Policy advocacy and organizational change at the Association of University Technology Managers (AUTM) Christopher S Hayter and Jacob H Rooksby

131

6

7

Conflicts of interest and academic research Jorge L Contreras and Marc Daniel Rinehart

8

Modern intellectual property valuation in the academic technology transfer setting Bryce Pilz

PART II 9

4

143

166

INTELLECTUAL PROPERTY AND TECHNOLOGY TRANSFER IN THE INNOVATIVE UNIVERSITY The innovation arms race on academic campuses Todd Sherer and Liza Vertinsky

vii

196

viii Research handbook on intellectual property and technology transfer 10

Tacit knowledge and university-industry technology transfer Peter Lee

214

11

Technology transfer and the public good Brian L Frye and Christopher J Ryan, Jr

236

12

US patent sales by universities and research institutes Brian J Love, Erik Oliver and Michael Costa

256

13

Intellectual property exchanges and auctions: non-traditional mechanisms for technology transfer Daniel R Cahoy

14

Currents and crosscurrents in litigation of university and nonprofit related patents: is there a coming wave of patent litigation involving those patents? Teo Firpo and Michael S Mireles

15

Is patent enforcement efficient? Mark A Lemley and Robin Feldman

16

Reviewing inter partes review five years in: the view from university technology transfer offices Cynthia Laury Dahl

283

309 324

339

17

Data governance and the emerging university Michael J Madison

364

18

“Free data?”: open science in the age of personal data protection Paolo Guarda

391

PART III

GLOBAL PERSPECTIVES ON INTELLECUTAL PROPERTY AND TECHNOLOGY TRANSFER

19

A European perspective on intellectual property and technology transfer James A Cunningham, Marco Romano and Melita Nicotra

412

20

The current state of university technology transfer in China Zhang Chu and Shi Xiaoxue

434

21

Make and share: intellectual property, higher education, technology transfer, and 3D printing in a global context Matthew Rimmer

Index

447

480

Figures

11.1

University utility patents granted 1980–2012

240

12.1

US patent sales by universities and research institutes (2012–2017), grouped by seller type

264

12.2

US university patent sales (2012–2017), by university ranking

264

12.3

US patent sales by universities and research institutes (2012–2017), grouped by buyer type

267

US patent sales by universities and research institutes (2012–2017), grouped by reason for sale

268

13.1

Basic forms of non-traditional transfer mechanisms

289

13.2

Building blocks of a successful exchange or market

298

14.1

Patent age at litigation

313

14.2

Cases by year

314

14.3

Length of litigation in ED Texas

317

12.4

ix

Tables

12.1

Overall statistics on US patent sales by universities and research institutes (2012–2017)

263

US patent sales by universities and research institutes (2012–2017), grouped by country of origin

265

US patent sales by universities and research institutes (2012–2017), grouped by country of destination

267

US patent sales by universities and research institutes (2012–2017), grouped by technology category

269

Litigation of US patents sold by universities and research institutes (2012–2017)

270

12A.1

Assignors (Sellers)

276

12A.2

Assignees (Buyers)

280

14.1

Patents by type of organization

312

19.1

Ownership of IPRs at universities in selected European countries

417

20.1

Major regulatory and policy documents pertaining to university technology transfer in China

436

Top ten university and public research organization patent applications since 1995

455

12.2 12.3 12.4 12.5

21.1

x

Contributors

Daniel R Cahoy is Professor of Business Law and Dean’s Faculty Fellow in Business Law at Penn State Smeal College of Business in University Park, Pennsylvania. Jennifer Carter-Johnson is Associate Professor of Law and the Co-Associate Director of the Intellectual Property, Information and Communications Law program at Michigan State University College of Law in East Lansing, Michigan. Zhang Chu is Dean and Professor of Law at Xi Hua University of Law and Intellectual Property in Chengdu, China and Professor of Law at China University of Political Science and Law, School of Civil and Commercial Economics and Law, in Beijing, China. Jorge L Contreras is Professor of Law and Adjunct Professor of Human Genetics at the University of Utah in Salt Lake City, Utah. Michael Costa is an Intellectual Asset Analyst at the law firm of Richardson Oliver Law Group LLP in Los Altos, California. James A Cunningham is Professor of Strategy Management at Northumbria University Newcastle Business School in Newcastle upon Tyne, United Kingdom. Cynthia Laury Dahl is Practice Professor of Law and Director of the Detkin Intellectual Property and Technology Legal Clinic at the University of Pennsylvania Law School in Philadelphia, Pennsylvania. Robin Feldman is Arthur J Goldberg Distinguished Professor of Law and Director of the Center for Innovation at University of California Hastings Law in San Francisco, California. Teo Firpo is Senior Researcher at Nesta in London, United Kingdom. Brian L Frye is Spears-Gilbert Associate Professor of Law at University of Kentucky College of Law in Lexington, Kentucky. Shubha Ghosh is Crandall Melvin Professor of Law and Director, Technology Commercialization Curricular Program and Syracuse Intellectual Property Law Institute at Syracuse University College of Law in Syracuse, New York. Paolo Guarda is Assistant Professor at the University of Trento Faculty of Law in Trento, Italy. Christopher S Hayter is Assistant Professor, School of Public Affairs, Watts College of Public Service and Community Solutions, and Senior Sustainability Scholar, Julie Ann Wrigley Global Institute of Sustainability at Arizona State University in Phoenix, Arizona. Peter Lee is Martin Luther King Jr. Professor of Law at UC Davis School of Law in Davis, California.

xi

xii Research handbook on intellectual property and technology transfer Mark A Lemley is William H Neukom Professor of Law and Director, Program in Law, Science, and Technology at Stanford Law School in Stanford, California. Brian J Love is Associate Professor of Law and Co-Director of the High Tech Law Institute at Santa Clara Law University School of Law in Santa Clara, California. Michael J Madison is Professor of Law at University of Pittsburgh School of Law in Pittsburgh, Pennsylvania. Michael S Mireles is Professor of Law at University of the Pacific McGeorge School of Law in Sacramento, California. Melita Nicotra is Assistant Professor of Management at the University of Catania Department of Economics and Business in Catania, Italy. Erik Oliver is a partner at the law firm of Richardson Oliver Law Group LLP in Los Altos, California. Bryce Pilz is Director of Licensing, Office of Technology Transfer at the University of Michigan in Ann Arbor, Michigan. Matthew Rimmer is Professor in Intellectual Property and Innovation Law on the Faculty of Law at Queensland University of Technology in Brisbane, Australia. Marc Daniel Rinehart is Director of the Conflict of Interest Office at the University of Utah in Salt Lake City, Utah. Marco Romano is Professor of Entrepreneurship and Business Planning at the University of Catania Department of Economics and Business in Catania, Italy. Jacob H Rooksby is Dean and Professor of Law at Gonzaga University School of Law in Spokane, Washington. Christopher J Ryan, Jr is Associate Professor of Law at Roger Williams University School of Law in Bristol, Rhode Island and an Affiliated Scholar at the American Bar Foundation in Chicago, Illinois. Jessica A Sebeok is Deputy Vice President for Federal Relations & Counsel for Policy at the Association of American Universities in Washington, District of Columbia. Todd Sherer is Associate Vice President for Research and Executive Director of the Office of Technology Transfer at Emory University in Atlanta, Georgia. Liza Vertinsky is Associate Professor of Law and Emory Global Health Institute Faculty Fellow at Emory University School of Law in Atlanta, Georgia. Jarrett B Warshaw is Assistant Professor of Higher Education in the Department of Educational Leadership and Research Methodology at Florida Atlantic University in Boca Raton, Florida. Shi Xiaoxue is a Doctor of Intellectual Property Law student at China University of Political Science and Law, School of Civil and Commercial Economics and Law, in Beijing, China.

Acknowledgments

This book is the culmination of hard work and attention from its contributors. I am honored that they chose to share their research and perspectives in these pages. Working with them in the publication process was a pleasure. This publication would not have occurred without the support and encouragement of several individuals. First, Chris Hayter served as a sounding board, advisor, and chapter co-author as the book took shape. Chris and I share an excitement for technology transfer and intellectual property, as well as the quest to lead an authentic life well lived. It is little wonder he is my go-to academic collaborator. Second, Kim Hai Pearson graciously agreed to take on enhanced responsibility at Gonzaga at a critical time. Her selflessness and dependability allowed me spare moments to write, edit, and finish the book, not to mention gave me peace of mind when I desperately needed it. Third, Stephanie Conlin heard my complaints and always kindly reminded me to take my foot off the gas if my pace started to approach ludicrous speed. Fourth, my wife, Susan, supported my academic and administrative work as we moved from one coast to the other for my career during the middle of this project. Her unfailing spirit and grounded support, in all aspects of my life, permits me the luxury to pursue my passions. Finally, my research assistant at Gonzaga, Emily Hazen, performed commendable work on the book over the course of her third year in law school. She expertly edited drafts, corresponded with contributors, and put up with my idiosyncratic calendar and work schedule as we moved this project to completion. Thank you, Emily, for all your behind-the-scenes contributions to this book. It is a better product because of your involvement. The above recognitions notwithstanding, any errors or oversights in the framing or editorial work associated with this book are completely my own. Jacob H Rooksby

xiii

Abbreviations

AAU APLU AUTM CAFC IP OECD TTO(s) USPTO

Association of American Universities Association of Public and Land-grant Universities Association of University Technology Managers United States Court of Appeals for the Federal Circuit Intellectual property Organisation for Economic Cooperation and Development Technology transfer office(s) United States Patent and Trademark Office

xiv

1.

Introduction to the Research Handbook on Intellectual Property and Technology Transfer Jacob H Rooksby

University researchers help improve lives by solving problems that impact health, happiness, and societal wellbeing. Universities harness the power of faculty innovations and disseminate them to the public, using intellectual property to create protections and provide rewards. This process of technology transfer has been vibrant in US universities since at least the Second World War and has grown in importance everywhere since then. The Bayh-Dole Act of 1980 helped spark formalized attention to technology transfer in the US and has inspired similar legislation in countries across the globe. Nearly forty years since the beginning of concerted attention to the topic, technology transfer has come of age. Institutions have grown in their sophistication and enhanced their abilities to be of service to society through innovation. Indeed, institutional governing boards and the public increasingly demand it. A focus on innovation and economic development figures prominently in institutional mission statements and the day-in, day-out functioning of the university. Entrepreneurial spirits are not uncommon, and faculty frequently think in terms of the value of their research and the potential contributions it can make to humankind. At the same time that higher education has adopted market-based approaches and incentives for the commercialization of technology, universities face ever more legal complexities and challenges. Changes to the patent law in the US in 2011 have led to a landscape where the stakes are higher, and the costs are greater, than ever before for universities seeking success in this realm. While universities face pressures to keep up, they also face criticism from all quarters: both that they are insufficiently savvy in the business and legal markets, and insufficiently sensitive to the ramifications of their participation in those markets. The public interest in reaping the benefit of university research reigns supreme, but the vexing question still remains of how universities can best further the public interest through their approaches to intellectual property and technology transfer. Through twenty-one chapters, this book speaks to the past and present of technology transfer and intellectual property. Part I examines the policies, practices, and politics that led to the current system of technology transfer and intellectual property management on academic campuses. Both established and emerging authors provide perspective on important topics such as institutional intellectual property policies and technology transfer structures; the politics of technology transfer; the enduring legacy and influence of the Bayh-Dole Act; organizational design and technology transfer; policy advocacy and organizational change at the Association of University Technology Managers; conflicts of interest and conflicts of interest policies; and intellectual property valuation in the academic technology transfer setting. Part II examines technology transfer and intellectual property in what I call the “innovative university,” as it now exists or soon will. The innovative university is at the frontier of knowledge production and dissemination. Its raison d’être is innovation. Functioning less like a bureaucracy, and more like a market-situated, public-oriented research enterprise, 1

2 Research handbook on intellectual property and technology transfer the innovative university mixes public and private resources and approaches in its efforts to further its mission. Intellectual property protections serve both as a source of opportunity and anxiety for universities attempting to achieve technology transfer in a complex, expensive, and highly-regulated innovation marketplace. Chapters in Part II consider such topics as the innovation arms race in higher education; tacit knowledge and university-industry technology transfer; technology transfer and the public good; patent sales by US universities; intellectual property exchanges and auctions; university involvement in patent litigation and the complexities of patent enforcement; university participation in inter partes review; data governance; and open science in the age of personal data protection. Part III concludes the book by offering global perspectives on many of these topics. Authors in that part contribute important information on intellectual property and technology transfer in Europe and China, as well as the treatment of 3D printing in higher education in a selection of countries. I hope you enjoy reflecting on chapters in this book as we collectively consider the promise of technology transfer and intellectual property to further the common good worldwide in the 21st Century.

PART I TECHNOLOGY TRANSFER COMES OF AGE: POLICIES, PRACTICES AND POLITICS

2.

University technology transfer structure and intellectual property policies Jennifer Carter-Johnson

I.

INTRODUCTION

University technology transfer is a broad topic. At base, universities transfer technology every day through their basic functions of teaching students. University researchers transfer technology to the public by publishing journal articles and giving talks. However important these types of interactions are, they are not the focus of this Chapter. Instead, this Chapter focuses on the entrepreneurial methods that universities implement to transfer new creations made by members of the university community to the public benefit – specifically perfecting intellectual property (“IP”) rights, licensing, and creating start-ups. Technology transfer also dovetails with university entrepreneurship more broadly. Universities may also transfer technology by creating incubator space and other mixed-use research areas. Such activities, taken in total, may strengthen the local or national economy.1 In fact, technology transfer is big business.2 From 1996 to 2015, technology transfer activities contributed almost six billion dollars to the United State gross domestic product. This staggering number is based on researchers collaborating with institutional technology transfer offices (“TTOs”). Specifically, researchers disclosed over 380,000 creations that resulted in over 80,000 issued patents. These patents were overwhelmingly licensed to small businesses and start-ups. In fact, over 11,000 start-ups have been created based on university technology during this timeframe. Additionally, these patents have supported over 4.3 million jobs and more than 200 new drugs and medical treatments.3 In 2016 alone, universities filed over 16,000 patent applications, received about 7,000 issued patents, executed 7,700 licenses, and helped create 1,100 new start-ups and 800 new products.4 Universities each approach technology transfer with different goals and perspectives. Therefore, it is difficult to truly rank them against each other. However, one study used four key objective indicators of TTO success to attempt to quantify the success of various univer-

1 For an in-depth overview of the literature of university entrepreneurship between 1981 and 2005, see Frank T. Rothaermel, Shanti D. Agung, & Lin Jiang, University entrepreneurship: a taxonomy of the literature, 16 Indus. & Corp. Change 691 (2007). 2 AUTM, Driving the Innovation Economy academic technology transfer in numbers, available at http://www.autm.net/AUTMMain/media/SurveyReportsPDF/AUTM-FY2016-Infographic-WEB.pdf (last visited Dec. 23, 2018). 3 University technology transfer is overwhelmingly patent driven except in the case of computer software. As described below in the description of intellectual property policies, universities often do not take ownership of copyrighted material that is not also patented. Therefore, the numbers collected by AUTM do not reflect the economic impact of copyrighted materials such as textbooks, music, and theatrical works to the economy. 4 AUTM, supra note 2.

4

University technology transfer structure and intellectual property policies 5 sities.5 The four key indicators of success considered were “patents issued, licenses issued, licensing income, and start-ups formed,” which were normalized based on research funding to each university. Based on this ranking, the top ten university technology transfer successes were (1) University of Utah, (2) Columbia University, (3) University of Florida, (4) Brigham Young University, (5) Stanford University, (6) University of Pennsylvania, (7) University of Washington, (8) Massachusetts Institute of Technology (MIT), (9) California Institute of Technology (Caltech), and (10) Carnegie Mellon University. This Chapter does not focus on these top ten universities. Instead, it attempts to give a broad overview of the diversity of ways that universities approach technology transfer. The Chapter begins with a brief overview of the history of technology transfer to situate the discussion of current technology transfer issues. TTO structure options are then discussed before turning to an overview of IP policies (“IP Policies”) that govern the technology transfer process at various universities. Finally, the Chapter wraps up with a discussion of the incentives of various groups within the technology transfer process.

II.

DEVELOPMENT OF TTOS

Diffusion of knowledge, innovations, and inventions from academia and into the broader public is widely understood to occur through diverse channels.6 Perhaps most importantly, academic researchers publish their ideas and innovations in journals accessible to other academics, industry and the broader public. Academic researchers commonly attend and speak at conferences attended by other researchers from academia and industry. Academic researchers commonly act as industry consultants. Students are trained and leave the university, carrying with them knowledge and skills from academia. These important ‘informal’ transfers of knowledge are core missions of the university and shape the more formal property rights transfers. Patent disclosures also inform the broader public and industry about innovations created by academic researchers, but patents allow the inventor (or an assignee such as a university employer) to retain property rights in those innovations. This ownership allows the inventor or the university to then license those rights—a more active and formal involvement than seen with other channels of knowledge diffusion. Technology transfer, when applied to universities, is often defined by some transfer of these property rights, such as a license, between the university and a third party.7 Technology trans-

5 Ross DeVol, Joe Lee, & Minoli Ratnatunga, Concept to Commercialization: The Best Universities for Technology Transfer, MIlken InstItute, April 2017, available at https://www.milkeninstitute.org/ publications/view/856 (last visited Oct. 17, 2019). 6 See, e.g., Wesley M. Cohen, Richard R. Nelson, & John P. Walsh, Links and Impacts: The Influence of Public Research on Industrial R&D, 48 MgMt sCI. 1 (2002); Jerry G. Thursby & Marie C. Thursby, Industry Perspectives on Licensing University Technologies: Sources and Problems, 15 Industry & hIgher ed. 289 (2001); Bhaven N. Sampat, Patenting and US Academic Research in the 20th Century: The World Before and After Bayh-Dole, 35 res. pol’y 772, 773 (2006); Ajay Agrawal & Rebecca Henderson, Putting Patents in Context: Exploring Knowledge Transfer from MIT, 48 MgMt sCI. 44 (2002). 7 The Association of University Technology Managers, a prominent industry interest group, defines technology transfer as “a term used to describe a formal transfer of rights to use and commercialize

6 Research handbook on intellectual property and technology transfer fer processes in universities has a long history. However, it is only in the late 20th century, with the passage of the Bayh-Dole Act, that large numbers of universities began formal, in-house programs of technology transfer. A.

History of University IP Rights

In the early 20th century there was considerable debate about the role of patenting by academic researchers. Many scholars feared that patenting by academic researchers would degrade the social norms surrounding academic research. In 1942, Robert Merton described an “ethos of science,”8 including four principal social norms of scientific research which he inferred from the activities and publications of scientists.9 One such norm, communism (today most often referred to as “communalism”), contains the idea that “substantive findings of science are a product of social collaboration and are assigned to the community.”10 Communalism is clearly at odds with the notion of an individual researcher or even an institution patenting an innovation.11 Merton also described the norm of disinterestedness, the idea that scientific institutions should avoid bias and act in the interest of science itself rather than self-interest or aggrandizement.12 Disinterestedness competes with the idea of monetary compensation for the works of science. The concerns of scientists over patenting was elucidated in a 1934 report, The Protection by Patents of Scientific Discoveries : Report of the Committee on Patents, Copyrights and Trade Marks of the American Association for the Advancement of Science (“AAAS Report”).13 The AAAS Report investigated many of the objections presented against patenting of scientific

new discoveries and innovations resulting from scientific research to another party.” Frequently Asked Questions, assoCIatIon of unIversIty teChnology Managers, https://autm.net/about-tech-transfer/ what-is-tech-transfer/tech-transfer-faq (last visited Dec. 20, 2018). 8 Robert K. Merton, A Note on Science and Democracy, 1 J. legal & pol. soC. 115, 116–17 (1942). 9 Id. at 118–26. 10 Id. at 121. 11 This social norm still resides among modern academic researchers, who often retain an emphasis on the sharing of materials and data despite the pressures of patent rights and competition between laboratories. See, e.g., Katherine Strandburg, User Innovator Community Norms: At the Boundary Between Academic and Industrial Research, 77 fordhaM l. rev. 2237 (2008); Nikolaus Franke & Sonali K. Shah, How Communities Support Innovative Activities: An Exploration of Assistance and Sharing Among End-Users, 32 res. pol’y 157 (2003); Mark A. Lemley, Ignoring Patents, 19 MICh. st. l. rev. (2008); John P. Walsh, Wesley M. Cohen, & Charlene Cho, Where Excludability Matters: Material Versus Intellectual Property in Academic Biomedical Research, 36 res. pol’y 1184, 1185–6 (2007). 12 Merton, supra note 8, at 124–5. (“There is competition in the realm of science, competition that is intensified by the emphasis on priority as a criterion of achievement, and under competitive conditions there may well be generated incentives for eclipsing rivals by illicit means. But such impulses can find scant opportunity for expression in the field of scientific research. Cultism, informal cliques, prolific but trivial publications—these and other techniques may be used for self-aggrandizement. But, in general, spurious claims appear to be negligible and ineffective. The translation of the norm of disinterestedness into practice is effectively supported by the ultimate accountability of scientists to their compeers. The dictates of socialized sentiment and of expediency largely coincide, a situation conducive to institutional stability.”) 13 American Association for the Advancement of Science, The Protection by Patents of Scientific Discoveries: Report of the Committee on Patents, Copyrights and Trade Marks, Joseph Rossman, chairman, oCCasIonal publICatIons aM. ass’n for advanCeMent of sCI. 1 (1934) [hereinafter AAAS Report].

University technology transfer structure and intellectual property policies 7 research results, including that the patents would encourage secrecy rather than publication and would constrain future research. The AAS Report also addressed the ideas that patenting debases scientific research and indicates researcher interest in financial profits. The concern that one researcher should not receive credit for work based in large part on the previous work of other scientists was also addressed. These objections were very much in line with the social norms of science that Robert Merton would later identify.14 Despite these scientific norms, patenting by academic scientists and universities did occur before and during this time period. For example, beginning in 1907, Frederick Cottrell, a researcher at the University of California-Berkeley, began to obtain what would eventually become a portfolio of six patents claiming his inventions in the area of electrostatic precipitation, which could remove particulate pollution from industrial smokestacks.15 Cottrell obtained patent rights because he believed that considerable investment would be needed to commercialize his inventions. Patent rights would provide incentives for investment by allowing companies that license the rights to operate with limited competition. However, Cottrell did not want to use his home institution, the University of California, to manage the patents and licenses. He feared that commercialism would have negative effects on the culture of the institution and eventually promote secrecy in the scientific work there. Therefore, he founded an independent entity, the Research Corporation, with the intention that it manage the patents and licenses and use the proceeds to support further scientific research. Over the next several decades the Research Corporation would manage Cottrell’s patent rights and award research grants to other scientists. During this period, other scientists also donated their patentable inventions to the Research Corporation. The actions of Cottrell and other scientists in the period were very much consistent with debate outlined by the AAAS Report mentioned above.16 Scientists did not wish to violate social norms but recognized that moving their innovations into public use often required incentivization in the form of patent rights. Having separate entities manage those rights allowed the appearance of financial interest to be kept at arm’s length while still commercializing the invention. The Research Corporation and the University of Wisconsin’s Wisconsin Alumni Research Foundation (WARF),17 established in 1925,18 are examples of this solution. The AAAS Report supported the use of these organizations for patent management and licensing. The AAAS Report concluded that mere publication of an innovation was often insufficient to move the innovation into broader public or industry use and that patent rights would provide a needed vehicle to do so.19 As stated by Archie Palmer in 1948, “[d]iscoveries

14

Merton, supra note 8, at 116–17. David C. Mowery & Bhaven N. Sampat, Patenting and Licensing University Inventions: Lessons from the History of the Research Corporation, 10 Indus. & Corp. Change 317, 320–34 (2001). 16 AAAS Report, supra note 13. 17 Wisconsin Alumni Research Foundation (WARF) is affiliated with the University of Wisconsin but is a legally separate entity. For a detailed description of the creation of WARF, see David C. Mowery, Richard R. Nelson, Bhaven N. Sampat, & Arvids A. Ziedonis, Ivory Tower and Industrial Innovation: University-Industry Technology Transfer Before and After the Bayh-Dole Act, 39–40 (2004) [hereinafter Ivory Tower and Industrial Innovation]. 18 See AUTM, AUTM U.S. Licensing Activity Survey: FY2009, 52 [hereinafter AUTM FY2009]. 19 AAAS Report, supra note 13. For a comprehensive analysis of this committee report, see Ivory Tower and Industrial Innovation, supra note 17, at 36–8. 15

8 Research handbook on intellectual property and technology transfer or inventions that are merely published, and are thus made available to everybody equally, are seldom adopted, despite their possibilities of commercial application.”20 Patenting by academic researchers and universities continued to gain acceptance through the 1930s. Recognizing the importance of the issue, several major research universities began to implement patent policies.21 For instance, by 1934 the Massachusetts Institute of Technology (MIT) had a formal patent policy describing the instances in which MIT would retain rights to an invention and the situations in which the faculty member or student would retain any potential patent rights.22 Generally, patent rights distribution depended on the source of funding for the work.23 In 1937, MIT entered into an Invention Administration Agreement with the Research Corporation for the management and licensing of MIT’s patents, thus perpetuating both trends of research universities having defined patent policies and using a separate entity to manage and license university patents.24 By the mid-1960s, nearly two hundred research universities and institutions had similar agreements with the Research Corporation.25 World War II, and the politics of the subsequent Cold War, brought considerable changes to university research in the United States. Prior to World War II, federal funding to university research was fairly modest.26 However, “big science” projects such as the Manhattan Project saw dramatic increases in federal funding.27 Through the 1950s and 1960s, federal funding of university research would increase from $138 million in 1953, to $1.6 billion in 1969.28 By 1980, federal funding of university research was nearly $4.1 billion, and the federal government funded approximately two-thirds of university research.29 This influx of federal funding was provided through numerous federal agencies, including the Department of Defense and the National Institutes of Health.30 IP rights, including patent rights, that were produced by federally funded work before 1980 were controlled by the rules of the individual funding agency31 and often simply inured to the funding agency.32 However, by the early 1970s, agencies such as the National Institutes of Health and the National Science

20

Archie M. Palmer, Survey of University Patent Policies: Preliminary Report, 5–6 (1948). Id. 22 Archie M. Palmer, University Patent Policies and Procedures, 16 J. pat. off. soC. 96, 107–9 (1934). 23 Id. 24 Mowery & Sampat, supra note 15, at 326. 25 Id. at 329–30. 26 Ivory Tower and Industrial Innovation, supra note 17, at 23; see generally Institute of Medicine & Nat’l Research Council, Large-Scale Biomedical Science, 234–7 (2003). 27 See Appendix Table 5-1, Total and federally financed higher education R&D expenditures, by type of R&D: FYs 1953–2016, nat’l sCI. board, available at https://www.nsf.gov/statistics/2018/ nsb20181/assets/968/tables/at05-01.pdf (last visited Dec. 28, 2018) [hereinafter Appendix Table 5-1]; see also D.C. Mowery, “The Bayh-Dole Act and High-Technology Entrepreneurship in US Universities: Chicken, Egg, or Something Else?” in 16 Advances Study Entrepreneurship, Innovation & Econ. Growth vol. 46 (G.D. Libecap ed., 2005). 28 Appendix Table 5-1, supra note 27. 29 Id. 30 Ivory Tower and Industrial Innovation, supra note 17, at 43, 45. 31 See, e.g., John E. Tyler III, Advancing University Innovation: More Must be Expected—More Must be Done, 10 MInn. J. l. & sCI. 143, 146 (2009) (“Prior to the passage of the Act, there were twenty-six different federal agency policies about using the results of federally funded research.”). 32 See, e.g., Bhaven N. Sampat, Patenting and US Academic Research in the 20th Century: the World Before and After Bayh-Dole, 35 res. pol’y 772, 776–8 (2006). 21

University technology transfer structure and intellectual property policies 9 Foundation made institutional patent agreements available to universities and other institutions that would streamline the process of allowing the university or institution to retain and license patent rights to innovations funded by that agency.33 Combining increased federal funding, increased resultant research and an increased ability to retain the patent rights to resulting innovations likely promoted the changes in university technology transfer seen in the late 1960s and 1970s.34 Before this period the majority of universities used a third party such as the Research Corporation for managing and licensing their patents.35 Increasing amounts of university patenting likely made it seem economical for many more individual research universities to provide their own patent management, either through an in-house TTO or through an affiliated entity such as the University of Wisconsin’s WARF.36 Stanford University, for instance, began a trial program managing their own patents in 1968.37 Stanford developed licensing revenue that year that far exceeded the revenue the Research Corporation had produced for them during the combined preceding thirteen years.38 Stanford would proceed to institute their own in-house TTO in 1970.39 B.

The Bayh-Dole Act

The Bayh-Dole Act of 198040 continued these trends by fully opening the door to university ownership of patents resulting from federally funded work. The Bayh-Dole Act’s purpose was to promote commercialization of innovations resulting from federally funded research by providing a consistent patent policy for research funding from federal agencies.41 In arguing for the need for such a law, it was noted that, of the approximately 30,000 patents owned by the federal government by 1980, only 5% of those had ever been licensed to industry.42 Some of the congressional arguments surrounding passage of the Bayh-Dole Act were similar to those observed earlier in the century in regards to patenting by scientists.43

33

Ivory Tower and Industrial Innovation, supra note 17, at 45; Sampat, supra note 32, at 778. Sampat, supra note 32, at 776. 35 See Sampat, supra note 32, at 773–4; Mowery & Sampat, supra note 15, at 329–30. 36 See generally, Sampat, supra note 32. 37 Ivory Tower and Industrial Innovation, supra note 17, at 45. 38 Id. 39 AUTM FY2009, supra note 18, at 50. 40 Bayh-Dole Act of Dec. 12, 1980, Pub. L. No. 96-517, 94 Stat. 3015 (1980) (codified at 35 U.S.C. §§ 200–212 (2012)). 41 One stated policy of the Bayh-Dole Act is “to promote the utilization of inventions arising from federally supported research or development.” 35 U.S.C.A. § 200 (West, Westlaw through P.L. 114-114 (excluding P.L. 114-92, 114-94, and 114-113) 2016). 42 National Research Council, Committee on Management of University Intellectual Property: Lessons from a Generation of Experience, Research, and Dialogue, Managing University Intellectual Property in the Public Interest, 16 (Stephen A. Merrill & Anne-Marie Mazza, eds. 2010) (“Prior to the passage of the Bayh-Dole Act, the United States government held over 30,000 patents, but of those only 5% were licensed out to private industry.”); see Daniel S. Greenberg, Science for Sale: The Perils, Rewards, and Delusions of Campus Capitalism ch. 10 (2007). 43 For a full description of the history leading up to the implementation of the Bayh-Dole Act, see Rebecca S. Eisenberg, Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research, 82 va. l. rev. 1663, 1671–95 (1996). 34

10 Research handbook on intellectual property and technology transfer The Bayh-Dole Act allows universities, businesses, and other research institutions receiving federal research funding to retain patent rights on innovations resulting from that research.44 By allowing universities the option to retain patent rights and licensing revenue, the Bayh-Dole Act incentivizes commercialization of federally funded inventions and provides an additional revenue stream for the universities. The Bayh-Dole Act also recognizes that researchers need to disclose patentable innovations to their institutions in order for patents to emerge. The Bayh-Dole Act attempts to incentivize researcher disclosure by requiring universities to share with inventors the royalties obtained from licensing.45 With this process, the Bayh-Dole Act also provides IP protection to licensees in order to promote investment in research and development required to create a commercial product from licensed technology.46 In exchange for this grant of ownership, the Bayh-Dole Act contains several requirements for these entities. First, the Bayh-Dole Act requires that the university (or other entity receiving federal funding for research) disclose patentable inventions to the funding agency. This disclosure must occur within two months of the institution’s patent administration office learning of the invention. Regulations require that the funded university or institution must have written agreements with research staff to disclose each invention to the institution’s patent administration office.47 Initially, many assumed that the language in the Bayh-Dole Act that provides universities the right to “elect to retain title to any subject invention” automatically vests ownership in the funded university.48 This assumption was tested and found false in the case of Stanford v. Roche.49 In Roche, the Supreme Court found that “[t]he Bayh–Dole Act’s provision stating that contractors may ‘elect to retain title’ confirms that the Act does not vest title.”50 The Roche decision was based on the actions of a Stanford University scientist who had signed both Stanford’s “Copyright and Patent Agreement” as well as a “Visitor’s Confidentiality Agreement” with Cetus (later sold to Roche).51 The agreement with Stanford required the scientist to assign any inventions to Stanford University.52 In contrast, the Supreme Court noted that the researcher’s agreement with Cetus stated “that he will assign and do[es] hereby assign to Cetus his right, title and interest in … the ideas, inventions, and improvements made as a consequence of [his] access to Cetus.”53 Therefore, the scientist had a contractual obligation to assign the invention to Stanford. However, the Roche agreement contained an actual assignment, leaving no remaining rights

44

35 U.S.C.A. § 202(a) (West, Westlaw through P.L. 114-114 (excluding P.L. 114-92, 114-94, and 114-113) 2016). 45 35 U.S.C.A. § 202(c)(7)(B) (West, Westlaw through P.L. 114-114 (excluding P.L. 114-92, 114-94, and 114-113) 2016). 46 Most inventions developed within academia require further refinement and adaptation before creating a commercial product. See Jerry G. Thursby, et al., Objectives, Characteristics and Outcomes of University Licensing: A Survey of Major U.S. Universities, 26 J. teCh. transfer 59, 62 (2001). 47 37 C.F.R. § 401.14(f)(2). 48 35 U.S.C.A. § 202(a) (West, Westlaw through P.L. 114-114 (excluding P.L. 114-92, 114-94, and 114-113) 2016). 49 Bd. of Tr. of Leland Stanford Junior Univ. v. Roche Molecular Sys., 563 U.S. 776 (2011). 50 Id. at 789. 51 Id. at 780–2. 52 Id. at 781. 53 Id. at 781 (internal quotations omitted; emphasis added).

University technology transfer structure and intellectual property policies 11 for the scientist to assign to Stanford.54 Thus, universities must have valid assignments from inventors in order to take title to the invention. The Bayh-Dole Act also requires that the funded institution not assign rights to a third party other than a patent management entity without prior approval from the funding agency.55 Therefore, universities theoretically are limited to licensing IP rights rather than transferring ownership of patents.56 Additionally, the Bayh-Dole Act requires that, if the invention is the subject of an exclusive license in the Unites States, then products embodying that invention must be substantially made within the United States unless this requirement is waived by the funding agency.57 Finally, as mentioned above, the funded institution must share some portion of any royalties from licensing with the inventor.58 In addition to these requirements, the Bayh-Dole Act also retains two additional rights for the federal government. The first is that it has “a nonexclusive, nontransferrable, irrevocable, paid-up license to practice or have practiced for or on behalf of the United States any subject invention throughout the world.”59 The second retained right is termed a “march-in” right.60 Using its march-in right, the federal government may force an institution to license an invention to a third party in order to meet a health or safety need, provided that the university had not already taken effective steps towards commercialization in compliance with the requirements of the Bayh-Dole Act. Passage of the Bayh-Dole Act coincided with a dramatic increase in academic technology transfer. Far more universities established their own TTOs, and today nearly every major research institution has a TTO or an affiliated patent management entity such as the University of Wisconsin’s WARF.61 In 2016, universities filed over 16,000 patent applications and received about 7,000 issued patents.62 Since its passage, the Bayh-Dole Act has been both lauded and derided. The Economist once termed the Bayh-Dole Act as “[p]ossibly the most inspired piece of legislation to be enacted in America over the past half-century.”63 However, the successes of the Bayh-Dole Act may be overstated. As described previously, the trends toward increased technology transfer from

54 Bd. of Tr. of the Leland Stanford Junior Univ. v. Roche Molecular Sys., Inc. 583 F.3d 832, 841–2 (Fed. Cir. 2009), aff’d by 563 U.S. 776 (2011). 55 35 U.S.C. § 202(c)(7)(A); 37 C.F.R. § 401.14(k)(1). For problems associated with such transfers, see also Jacob H. Rooksby, When Tigers Bare Teeth: A Qualitative Study of University Patent Enforcement, 46 akron l. rev. 169, 195 (2013) [hereinafter When Tigers Bare Teeth]; Jacob H. Rooksby, The Branding of the American Mind: How Universities Capture, Manage, and Monetize Intellectual Property and Why It Matters 161–3 (2016) [hereinafter The Branding of the American Mind] (discussing the fact that some universities use this section of the Bayh Dole Act to transfer patent rights to non-practicing entities for the sole purpose of asserting patent litigation claims). 56 This limitation on university IP transfer is limited to IP that is generated using federal funding, i.e., IP subject to the Bayh Dole Act. 57 35 U.S.C. § 204; 37 C.F.R. § 401.14(i). 58 35 U.S.C. § 202(c)(7)(B)–(C); 37 C.F.R. § 401.14(k)(2)–(3). 59 35 U.S.C. § 202(c)(4); 37 C.F.R. § 401.14(c)(3). 60 35 U.S.C. § 203; 37 C.F.R. §§ 401.6, 401.14(j). 61 See Gideon D. Markman et al., Entrepreneurship From the Ivory Tower: Do Incentive Systems Matter?, 29 J. of teCh. transfer 353, 353 (2004). 62 AUTM, supra note 2. 63 Innovation’s Golden Goose, eConoMIst (Dec. 12, 2002), available at http://www.economist.com/ node/1476653 (last visited Oct. 17, 2019).

12 Research handbook on intellectual property and technology transfer universities were already occurring in the decade before passage of the Bayh-Dole Act.64 Additionally, increased pressures to commercialize innovations may harm scientific norms and basic research at universities.65 Finally, the structure of the Bayh-Dole Act produces a situation in which tax payers pay twice for an invention—first to fund the research and later to buy the product.66 For good or evil, the Bayh-Dole Act is the basis on which the modern university technology transfer system sits.

III.

STRUCTURE OF TECHNOLOGY TRANSFER AT UNIVERSITIES

As noted above, the university technology transfer process begins well before the TTO becomes involved. At its base, university researchers are transferring technology to the public every day by educating students and creating new inventions and works of authorship that are published or performed for the public benefit. On the formal side of technology transfer, universities vary tremendously in their approach. Some universities do not have TTOs. Those universities that have invested in TTOs vary in organizational and operational structure. In addition, there are experimental models of technology transfer that few, if any, universities have adopted. A.

Overview of Technology Transfer Process

The technology transfer process begins in the university research laboratory.67 The large cast of the university research laboratory is typically led by a tenure-system faculty researcher, termed “Principal Investigator” (PI) in federal grants. However, PIs do not work alone. Other laboratory personnel and potential inventors include non-tenure track faculty, research associates, post-doctoral fellows, graduate students, undergraduate students, and technicians.68 Laboratory research results in inventions. The academic researchers, as the inventors, must disclose the inventions to the university’s TTO due to contractual obligations that were written in order to comply with the Bayh-Dole Act. Generally, the inventor is required to fill out a disclosure form providing basic information about the invention, relevant funding, and inventor identification.69 64

See Sampat, supra note 32, at 783–4. See Katherine J. Strandburg, “Curiosity-Driven Research and University Technology Transfer” in Univ. Entrepreneurship and Tech. Transfer: Process, Design, and Intellectual Property, 93 (Gary Libecap, ed. 2005) (examining potential alterations of basic research driven by university technology transfer). 66 Eisenberg, supra note 43, at 1666. 67 For a fuller description of the research laboratory, see Jennifer Carter-Johnson, Unveiling the Distinction Between the University and Its Academic Researchers: Lessons for Patent Infringement and University Technology Transfer, 12 vand. J. ent. & teCh. l. 473, 478–80 (2010). 68 See generally, Jennifer Carter-Johnson, Beyond Einstein and Edison: Claiming Space for Non-Faculty Inventors in Technology Transfer, 47 Ind. l. rev. 645 (2014). 69 There are two types of disclosure often mention in relation to university research, (1) disclosure by the inventor to the university, and (2) disclosure by the university to the relevant governmental funding agency. For the purposes of this Chapter, “disclosure” will refer to disclosure by the inventor of an invention to the university TTO. Carter-Johnson, supra note 67. 65

University technology transfer structure and intellectual property policies 13 Once the TTO receives the disclosure, it determines whether to patent and license the invention based upon its internal criteria—often including patentability opinions and marketability research. If the TTO determines that patent protection is appropriate, the TTO begins to prosecute the patent and attempts to license the invention or negotiate equity in a startup company. The TTO distributes revenues received from the patent back to the university to fund further research and support other university educational and administrative functions. Importantly, a portion of the licensing revenue must be distributed to the inventors as incentives to encourage invention and invention disclosure to the university.70 Again, this description is quite basic because variations exist between university technology transfer at practically all stages of the technology transfer process. In fact, the first question a university must ask is whether it needs a TTO at all. B.

Necessity of a TTO

The idea of having a TTO at a university sounds enticing. Technology transfer is a multi-billion-dollar industry. As such, there is the potential for a solid revenue stream in university technology transfer if it is the right fit for the university. A university thinking about creating a technology transfer process should first analyze its situation, consider the obligations of a TTO, and explore some basic alternatives to determine if a TTO is the right fit for its situation.71 In analyzing its situation, a university should first think about how well entrepreneurship and commercialization of technology fit into the university mission. 72 A successful TTO relies on more than the people who populate the office. Faculty, students, and staff must all work to create an atmosphere where successful commercialization happens.73 A teaching-focused university may find such an emphasis on developing technologies and entrepreneurship distracting from its central identity. A research-focused university, however, might find a technology transfer program to be a natural outgrowth of its creation of new technologies and drive to change people’s lives. Of course, many universities have dual missions, so these questions force an analysis of the balance between the goals. Additionally, a TTO can be incorporated into the teaching aspects of a university as students learn about the science, law and business involved in starting a technology company. Assuming a university finds the idea of technology transfer to align with its central mission, then it next must consider whether it produces enough research to support its own TTO.74 Although universities expect to initially subsidize the TTO office, most TTOs have the goal of becoming self-supporting based on income from IP.75 However, more than 50% of the TTOs

70

The Bayh-Dole Act requires non-profit organizations such as universities to share with the inventors some portion of the royalties obtained by the licensing of federally-funded inventions. 35 U.S.C. §202(c)(7)(b). 71 Terry A Young, “Establishing a Technology Transfer Office” in Intellectual Property Management in Health and Agricultural Innovation: A Handbook of Best Practices (Anatole Krattiger, Richart T. Mahoney, Lota Nelsen, et al., eds., 2007). 72 Id. at 545. 73 Brian Cummings, The Changing Landscape of Intellectual Property Management as a Revenue-Generating Asset for US Research Institutions, 21 geo. Mason l. rev. 1027, 1037 (2014). 74 Young, supra note 71, at 545–6. 75 Id. at 549.

14 Research handbook on intellectual property and technology transfer across the United States bring in less money than they expend. Of those that do bring in more gross revenue than expenditures, only 16% of TTOs become self-sustaining after distributions to inventors and the university administration.76 Additionally, based on university reported numbers, a university can expect one patent for each five million dollars invested in research.77 Since not every patent is licensed, license rates are even lower. With those numbers in mind, a university must determine how likely the TTO will become a loss in the university budget and whether the value gained from the TTO is worth that deficit. Finally, if the university decides to invest in a TTO, then it must be willing to commit for the long term.78 Again, early-stage TTOs require subsidies. TTOs must serve many different functions in order to be successful. TTOs help inventors identify commercially valuable creations and determine the types of IP protection necessary. TTOs must then find industry partners or help inventor-entrepreneurs to start new companies. Behind the scenes, TTOs must also keep the office running. Accounting, patent portfolio management, release of technology not pursued—each of these are essential for a healthy TTO.79 Additionally, TTOs are responsible for policing patent policies and resulting IP.80 In 2005, Yale University sued a former professor and Nobel prize winner for commercializing patented technology developed while he worked at Yale.81 Thus, a TTO whose primary goal may be to provide a service to the faculty must also be willing to force faculty to engage with the university’s technology transfer process. If, after introspection, a university wishes to have a technology transfer program but not a TTO, there are many alternatives to explore.82 A university may decide that any TTO is too much of an investment and may instead contract with an external organization—either non-profit or for-profit—to take invention disclosures and commercialize them. This costs the university little but also leaves the university with little control. Alternatively, the university could staff a small office to take and review invention disclosures. Promising inventions could then be passed along to an external organization for commercialization. Although costing more, this option allows the university to retain control over the technology that its researchers create. Of course, such a small office necessarily would have limited expertise which may be problematic if there are invention disclosures in many disciplines. A final alternative is to use a regional TTO. Universities can partner together or with non-profit or government research institutes to share TTO resources. 83 In all these cases, there is a danger of TTOs that are not part of the university having a difficult time understanding the university culture, taking part in its educational mission, and potentially, communicating with on-campus inventors.

76

Irene Abrams, Grace Leung, & Ashley J. Stevens, How are U.S. Technology Transfer Offices Tasked and Motivated—Is It All About the Money?, 17 res. MgMt rev. 18 (2009). 77 Young, supra note 71, at 546. 78 Id. 79 Id. at 553. 80 For an in-depth view of university-initiated patent litigation, see Jacob H. Rooksby, University Initiation of Patent Infringement Litigation, 10 J. Marshall rev. Ip l. 623 (2011); When Tigers Bare Teeth, supra note 55. 81 DeVol et al., supra note 5. 82 Young, supra note 71, at 546. 83 Id.

University technology transfer structure and intellectual property policies 15 C.

Organization of TTOs

Once a university has decided to establish a TTO, the planning begins. No two TTOs are the same. Even TTOs within statewide systems, such as the University of California, have differences between offices embedded at each campus.84 TTOs do not exist in a vacuum. Instead, TTOs work within a larger university push towards innovation. These TTOs may work alongside entrepreneurship mentors, laboratory space for incubators, industry research partnerships, and commercialization committees with industry experts.85 Much of a TTO’s role depends upon the structure, oversight, and financing of the office. From there the TTO’s operational model can be built. The internal structure of the TTO will dictate the way that the office works. TTOs may be internal or external to the institution. In addition, there are several ways to characterize organization and operational models. In all forms of organization, personal relationships between faculty, TTOs, industry, and entrepreneurs are at the core of the operation.86 TTOs may be organized as internal university departments or as external private foundations that take ownership of IP but work closely with the university and inventors.87 More than 80% of TTOs are organized as in-house departments within a university.88 Of those universities with external foundations, less than 5% are for-profit.89 Public universities more often set up external private foundations to avoid limitations on government entities.90 Perhaps one of the oldest, most well-known, private, non-profit foundations is WARF, the Wisconsin Alumni Research Foundation.91 The physical organization of TTOs is also important.92 Some universities have one central office to oversee all TTO functions. Other universities have satellite offices embedded within specific research groups. For example, Harvard University has an office at the medical school.93 Embedded offices add cost but allow for a TTO technology group to work more closely with the researchers for whom it is responsible. Relationship-building and ease of disclosure are two possible bonuses. Additionally, the work flow within the organization differs. In some TTOs, work is vertically integrated such that a disclosure is handled by one licensing associate as it moves through the process of patenting and licensing. Other TTOs have a more horizontal structure

84

Linara Axanova, U.S. Academic Technology Transfer Models: Traditional, Experimental and Hypothetical, les nouvelles 125 (2012). 85 DeVol, et al., supra note 5, at 18–24 (describing technology transfer efforts at the top ranked universities). 86 Peter Lee, Transcending the Tacit Dimension: Patent, Relationships, and Organizational Integration in Technology Transfer, 100 Cal. l. rev. 1503, 1521 (2012). 87 Young, supra note 71. 88 Abrams et al., supra note 76. 89 Daniele Battaglia, Paolo Landoni, & Francesco Rizzitelli, Organizational Structures for External Growth of University Technology Transfer Offices: An Explorative Analysis, 123 teCh. foreCastIng & soC. Change 45 (2017). 90 Abrams, et al., supra note 76. 91 WARF, https://www.warf.org/about-us/about-us.cmsx (last visited Dec. 26, 2018). 92 For an in-depth review of TTO organizational structure, see F. Brescia, G. Colombo, & P. Landoni, Organizational Structures of Knowledge Transfer Offices: an analysis of the world’s top-ranked universities, 41 J teCh. transfer 132 (2016). 93 Harvard OTD locations, harv., https://otd.harvard.edu/contact-us/ (last visited Dec. 26, 2018).

16 Research handbook on intellectual property and technology transfer such that different departments exist to handle patent prosecution, marketing technologies, and licensing.94 In one study, researchers linked organization forms to technology transfer outcomes. They concluded that the most effective structure is a semi-decentralized one in which semi-autonomous departments with different specialties are managed by a central group that holds the final decisions.95 Such an organization takes the best of the vertical and horizontal models by using specialists in each area while still allowing for central oversight of a disclosure. This organization also requires a high degree of communication and coordination. One important piece to this organizational structure is patent prosecution. The choice to handle patent prosecution in house or through external counsel is important. While paying external counsel for patent prosecution can be expensive, the benefits often outweigh the cost. For instance, if the university engages in research in a broad array of disciplines, no one patent attorney would be competent to prosecute patents in every discipline. Therefore, multiple in-house patent attorneys would be required. Additionally, industry partners may trust patents that are written by well-known law firms—leading to more commercialization deals. If a TTO chooses to forego having an attorney on staff, it should work closely with the university general counsel office for any legal issues that arise.96 A final organizational piece involves thinking about how TTOs operate within the larger university structure via reporting structures and budget. TTOs may report either through the academic channels of a university or through the administration. TTOs under academic control often report to the provost while administratively controlled TTOs usually report to a vice president.97 In either case, TTOs operate under set budget that combines university resources with any revenue from patent commercialization. Budgets impact how TTOs respond to disclosures. First, most TTOs have a budget that separates operating expenses from the money pool for pursuing patents.98 Using that patent money, TTOs have several approaches to patenting. TTOs with large budgets may choose to function more like a business, focusing on market potential rather than patent costs. Other TTOs may view their operations as a service to the faculty and patent everything. However, this strategy only works if there are a small number of disclosures relative to the budget. Still other TTOs follow a “pay as you go” approach and protect as many inventions as they can with available funds, then look for partners before spending money on patent protection. Finally, many TTOs will file provisional applications, then require a licensee or partner before converting to a PCT or US patent.99 D.

Operation of TTOs

The operation of TTOs will vary among universities depending on each one’s goals and missions. These missions may affect how the TTO works with faculty, whether the TTO partners

94

Axanova, supra note 84, at 128. Janet Bercovitz, Maryann Feldman, Irwin Feller, & Richard Burton, Organizational Structure as a Determinant of Academic Patent and Licensing Behavior: An Exploratory Study of Duke, Johns Hopkins, and Pennsylvania State Universities, 26 J. teCh. transfer 21 (2001). 96 Young, supra note 71, at 553–4. 97 Abrams, et al., supra note 76. 98 Id. 99 Axanova, supra note 84, at 129. 95

University technology transfer structure and intellectual property policies 17 with the local community, and how the TTO focuses on revenue. Revenue focus may include direct licensing revenues as well as equity in start-ups. Returning to the analysis of the university’s core missions discussed above, the operations of traditional TTOs are often a reflection of those core missions. Traditional missions include “(i) service to faculty, (ii) service to the public (i.e., by bringing new products to market), (iii) economic development (e.g., by supporting start-up companies licensing locally), (iv) revenue generation, and (v) compliance with the Bayh-Dole Act and institutional policies.”100 Each TTO skews toward missions of importance to its university.101 TTOs that follow a service model focus on patenting inventions and helping the faculty conduct research. Faculty research support includes negotiation of material transfer agreements (MTAs) between laboratories. MTAs memorialize the sharing of research tools between two groups and may include a cash royalty, an option to license future technology based on the tool, or pre-publication review of the results.102 Sponsored research agreements with industry partners provide another avenue for faculty research support. A sponsored research agreement is necessary when a researcher receives funding from a non-federal research sponsor and includes provisions regarding ownership of any resulting IP while maintaining compliance with the Bayh-Dole Act. The economic development model of TTOs tends to support the local economy with an emphasis on start-ups, graduate jobs, and mixed-use research areas. Finally, a revenue model of TTOs would find a focus on licensing and start-ups while maximizing returns.103 Most universities and TTOs serve a variety of core missions; thus, the emphasis of each TTO will be divided among those missions and prioritized uniquely. Of these missions, one survey found when questioned about motivations for TTOs, most top TTO executives list the goal of serving the faculty. The second most listed goal was to move research to the public. Far fewer TTOs listed revenue as a primary driver of operations.104 Although many TTOs deny that revenue generation is a primary driver of operations, the preferred revenue generation method is important in determining the operations of the TTO. There are three options for commercializing IP from universities—licensing, equity, or open-source.105 The first option, licensing IP to established companies, is the traditional revenue generator for TTOs. Licensing is the most straightforward way for a TTO to generate revenue from a patent. Negotiating a contract with an established partner allows the TTO to develop an on-going, long-term revenue stream that can be budgeted each year. As a second option, taking equity benefits both the university and the local economy, but it is not without drawbacks. In 2015, more than one thousand start-ups were created by TTOs. These start-ups tended to be in the same state as the source university.106 MIT pioneered using

100

Id. at 126. Id. 102 Rebecca S. Eisenberg & Arti Rai, Bayh-Dole Reform and the Progress of Biomedicine, 66 law & ConteMp. probs. 289, 294–5 (2003). 103 Axanova, supra note 84, 129–30. 104 Abrams et al., supra note 76. 105 For a more in-depth discussion, see Michael J. Bray & James N. Lee, University Revenues from Technology Transfer: Licensing Fees vs. Equity Positions, 15 J. bus. venturIng 385 (2000). 106 DeVol et al., supra note 5, at 13. 101

18 Research handbook on intellectual property and technology transfer equity in start-ups to generate revenue in the 1980s, averaging twenty-five companies per year during those times.107 While some have argued that using equity generally realizes more profit than licensing patents,108 equity is not the perfect answer. The timeline to revenue is much longer using equity than a license. It may take eight to twelve years for the start-up to develop enough revenue to go public or be acquired.109 Additionally, equity as a strategy does not work for every university. In order to have a strong start-up program, the TTO needs to have local support from both the university community and private entrepreneurs and local industry.110 Finally, sometimes the goal of transferring technology to the public benefit can happen with no revenue to the university. In particular, the software industry has used open source licenses extensively. Open source licenses allow for free sharing and adoption of software. Two of the main drawbacks to open source licensing include the lack of exclusivity and the possibility that derivatives of open source software also become open source by default due to the nature of some open source licenses. Surprisingly, in an industry that is known for its open source sharing of products, little time is given to the idea of open source license structures.111 E.

Experimental Structures of TTOs

The above discussion describes the general landscape of TTOs across the United States. However, there are many experimental structures that have been proposed. Some have been lauded while others have been widely criticized. None has been adopted widely. The Alfred Mann Foundation offered a novel collaboration with universities. The Alfred Mann Foundation offered one hundred million dollars to invest in university research in order to further develop the technology for commercialization. Under the program, the Alfred Mann Foundation would pick technology from the university that it thinks most promising, take control of the IP, and pay a portion of the revenue back to the university. Many universities have criticized the program due to questions over IP ownership and because the program allowed the foundation to cherry pick the best projects leaving the university with less revenue for its most promising projects.112 However, some universities did partner with the Alfred Mann Foundation.113 Collaborations between TTOs outside of joint research projects are rare but do occur. For example, the University of Pennsylvania and Arizona State University partnered their TTOs to allow each to take advantage of the expertise in the other.114 107

Bray & Lee, supra note 105, at 386. The exception to this statement would be in the case of direct licensing of a blockbuster drug. Bray & Lee, supra note 105, at 389. 109 Bray & Lee, supra note 105, at 389. 110 For a full discussion, see David Mowery, Bhaven Sampat, & Arvids Zeidonis, Learning to Patent, 48(1) MgMt sCI. 73 (2002). 111 The Branding of the American Mind, supra note 55. 112 Axanova, supra note 84. 113 Associated Press, Purdue partnered with the Alfred Mann Foundation in 2007 but ended the partnership in 2012. Purdue Ends $100M Deal with Research Foundation, Ind. bus. J. (Apr. 27, 2012) available at https://www.ibj.com/articles/34110-purdue-ends-100m-deal-with-research-foundation (last visited Oct. 17, 2019). 114 For Arizona State and Penn, 2 Tech-Transfer Offices Could Be Better Than One, Chron. hIgher eduC., available at https://www.chronicle.com/article/For-Arizona-StatePenn-2/42209 (last visited Dec. 22, 2018). 108

University technology transfer structure and intellectual property policies 19 Perhaps one of the most widely criticized proposals came from the Kauffman Foundation. According to this proposal, faculty would have control of their IP and be able to decide how to commercialize it, using the university TTO or another agency.115 Various criticisms of this proposal include questions about the ability of inventors to assess the best group for commercialization, increased bureaucracy as the commercialization group coordinates with the university as patent owner, and the potential for increased disputes among co-inventors. Finally, sole faculty ownership of patents on university inventions may run afoul of compliance with the Bayh-Dole Act. While the Bayh-Dole Act allows the return of patent rights to “the inventor,”116 the term inventor may not be synomomous with faculty member. Instead, inventor is defined as either an individual inventor or the collective individuals who are co-inventors.117 To have a policy which gives ownership to a faculty member ignores the non-faculty co-inventors on virtually every university patent118 who also have rights under the Bayh-Dole Act to any IP that the university chooses to return to the inventors.

IV.

STRUCTURE OF IP POLICIES AT UNIVERSITIES

In addition to adopting a structure for its TTO, a university also must devise a set of policies that delineate how it will operate. These policies are usually set out in the university’s IP Policy and are often incorporated into employment agreements and faculty or student handbooks. Typically, important subsections of the IP Policy include a general policy statement; definitions and obligations of university community members subject to the IP Policy; ownership of creations; revenue distribution; and dispute resolution. Like TTOs, the IP Policies vary widely between universities. This section discusses some of the common variations.119 A.

Stated IP Policy Goals

One of the initial pieces to many IP Policies is the statement of the university’s missions and goals for its technology transfer process. The main goals shared by many universities include (1) encouraging the development, production, and commercialization of creative or scholarly

115 Robert E. Litan, Lesa Mitchell, & E.J. Reedy, Commercializing University Innovations: A Better Way (May 2007), 14–15, available at https://www.brookings.edu/wp-content/uploads/2016/06/05 _innovations_litan.pdf (last visited Oct. 17, 2019). 116 35 U.S.C. § 202(d). 117 35 U.S.C. § 100(f). 118 Carter-Johnson, supra note 67. 119 For a more in-depth analysis of university technology transfer polices, see Jennifer Carter-Johnson, Survey of Technology Transfer Policies (manuscript in progress).

20 Research handbook on intellectual property and technology transfer works;120 (2) making the IP available in a manner that would provide a benefit to the public;121 and (3) clarifying, protecting, and managing IP covered by the IP Policy.122 Many IP Policies contain multiple goals.123 However, not every university IP Policy contains an explicitly stated goal.124 These goals mirror the aims of the IP system generally and the Bayh-Dole Act specifically. For instance, encouraging the development of creative works has its parallel in the US Constitution, which authorizes the patent and copyright system.125 Benefiting the public and managing ownership of university produced IP are goals of the Bayh-Dole Act. Other universities include goals that are more specific to their priorities. For example, the University of Michigan cites advancing economic development in Michigan.126 Other universities have goals of increasing revenue and generating further resources. The University of Texas System127 reports fostering relationships with industry and other academic institutions as a primary goal of its policy. Princeton University states a goal of its IP Policy is to exercise

120 See, e.g., Intellectual Property Policy, CarnegIe-Mellon u., available at https://www.cmu .edu/policies/administrative-and-governance/intellectual-property.html (last visited Dec. 23, 2018); Ownership of Research Data and Materials & Intellectual Property Management Implementation Policy, arIz. st. u., available at https://www.asu.edu/aad/manuals/rsp/rsp604.html (last visited Dec. 23, 2018); MSU Patent & Copyright Policy, MICh. st. u., available at http://www.technologies.msu .edu/researchers/patent-copyright-policy/msu-copyright-policy (last visited Dec. 23, 2018); University Patent & Invention Policy, Copyright Policy, Student Invention Policy, and Royalty Distribution Policy, nw. u., available at https://www.invo.northwestern.edu/invention-disclosure/policies-forms/index.html (last visited Dec. 23, 2018). 121 See, e.g., Patent Policy, Cal. Inst. of teCh., available at www.ogc.Caltech.edu/forms (last visited Dec. 23, 2018); Faculty Manual, Colo. st. u., available at https://facultycouncil.colostate.edu/faculty -manual-section-j/ (last visited Dec. 23, 2018); John Hopkins University IP Policy, Johns hopkIns u., available at https://ventures.jhu.edu/the-johns-hopkins-ip-policy/ (last visted Dec. 23, 2018); Guide to the Ownership, Distribution & Commercial Development of MIT Technology, MIt, available at http:// web.mit.edu/tlo/documents/MIT-TLO-ownership-guide.pdf (last visited Dec. 23, 2018). 122 See, e.g., Copyright, Inventions and Related Property Rights, Cornell u., available at https:// www.dfa.cornell.edu/sites/default/files/policy/vol1_5.pdf (last visited Dec. 23, 2018); Intellectual Property Policy, u. of arIz., available at http://policy.arizona.edu/printpdf/122 (last visited Dec. 23, 2018); Patents and Inventions, u. of utah, available at https://regulations.utah.edu/research/7-002.php (last visited Dec. 23, 2018). 123 See, e.g., Policies and Practices on Intellectual Property, u. of Ill. sys., available at http://otm .illinois.edu/sites/all/files/files/ippolicypaperformatted.pdf (last visited Dec. 23, 2018); Statement of Policy on Intellectual Property, nyu, available at https://www.nyu.edu/content/dam/nyu/compliance/ documents/IPPolicy.pdf (last visited Dec. 23, 2018). 124 See, e.g., Patent Rights and Technology Transfer, u. of pItt. pItt., available at https://www.cfo .pitt.edu/policies/policy/11/11-02-01.html (last visited Dec. 23, 2018); Patent, copyright – education materials, copyright ownership and management of software, intellectual property, Iowa st. u., available at https://www.policy.iastate.edu/policy/research-intellectual-property (last visited Dec. 23, 2018); OSU Policy on Intellectual Property, or. st. u., available at http://advantage.oregonstate.edu/policies -regulations (last visited Dec. 23, 2018). 125 u.s. Const., art. I, § 8, cl 8. “To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.” 126 University of Michigan Technology Transfer Policy, u. of MICh., available at http://spg.umich .edu/policy/303.04 (last visited Dec. 23, 2018). 127 Rule 90101: Intellectual Property, u. of tex. systeM, available at https://www.utsystem.edu/ sites/default/files/offices/board-regents/rules-regulations/90101.pdf (last visited Dec. 23, 2018).

University technology transfer structure and intellectual property policies 21 care for its fiduciary responsibilities.128 The University of Florida lists the important, but somewhat obvious, goal of “furthering the success of technology transfer at the University.”129 The Baylor University IP Policy contains a specific goal of promoting health.130 Harvard University’s goal of “appropriate protection of University name and insignia”131 may reflect Harvard’s popularity in pop culture and its reputation as an Ivy League university. Once the goals are defined, IP Policies must also define how to accomplish those goals. The first step in that process is to define who and what the IP Policy covers so that it is clear when a creation triggers the obligations of the policy. B.

IP Coverage: People and Obligations

Although the coverage of IP Policies is similar across universities, each IP Policy may define the covered groups in slightly different ways. Additionally, these IP Policies outline the rights and obligations of each university member in order to delineate the ownership of IP created under the IP Policy and the steps necessary for securing that IP to be owned by the university. While the nomenclature varies, students and employees are the most common broad categories of university community members covered under IP Policies. Students include everyone enrolled at the university, including both undergraduate and graduate students. Employees include faculty, staff, and very often, student on-campus workers. Thus, a student who created an invention would have to know whether she was working as a student or as an employee in order to fully understand her rights and obligations under the university IP policy. Furthermore, some policies include a catchall category for anyone who makes significant use of university resources to create IP. Additionally, many policies specifically state that IP created by covered individuals outside the scope of their employment or without the use of university resources is not covered by the policy. To the extent that a creation arises within the scope of university employment or using university resources, the IP Policies define the ownership of the resultant IP for each category of university member as described in section IV.C. It is important to note that IP Policies do not require that every creation at the university be assigned to the university. For those creations for which the university takes ownership under its IP Policy, the obligations imposed by the IP Policies show little variation. The two most common obligations are designed to give notice of creation to the university and allow it to take ownership of the IP. Specifically, these obligations are (1) mandatory disclosure132 and (2) assigning and con-

128

Rules & Procedures of the Faculty, prInCeton, available at https://dof.princeton.edu/book/ export/html/1141 (last visited Dec. 23, 2018). 129 University of Florida Intellectual Property Policy, u. of fla., available at http://generalcounsel .ufl.edu/media/generalcounselufledu/documents/Intellectual-Property-Policy.pdf (last visited Dec. 23, 2018). 130 Intellectual Property Policy of Baylor, baylor Coll. of Med., available at https://www.baylor .edu/content/services/document.php?id=42369 (last visited Dec. 27, 2018). 131 Statement of Policy in Regard to Intellectual Property (IP Policy), harv., available at https:// otd.harvard.edu/faculty-inventors/resources/policies-and-procedures/statement-of-policy-in-regard-to -intellectual-property/ (last visited Dec. 23, 2018). 132 See, e.g., arIz. st. u., supra note 120; baylor Coll. of Med., supra note 130; Intellectual Property Policy, b.u. Med CaMpus, available at https://www.bu.edu/academics/policies/intellectual -property-policy/ (last visited Dec. 23, 2018); harv., supra note 131.

22 Research handbook on intellectual property and technology transfer veying ownership of the IP to the university.133 Even after Stanford v. Roche134 clarified that universities risk losing IP to earlier assignments if they do not require assignment at the point of hire, many IP Policies still require assignment of the IP at the time of disclosure or patenting rather than including assignment in the IP Policy.135 However, some IP Policies have moved to language designed to assign rights to IP before creation.136 Other obligations that occur less frequently in IP Policies include assisting the university in securing IP rights137 and assisting the university in commercialization of IP.138 Even at universities in which the IP Policy does not contain these explicit provisions, creators who disclose often work closely with TTOs throughout the process of commercialization of the creations. C.

IP Coverage: Creations and Ownership

In general, university IP Policies cover inventions and copyrightable materials, though again, various universities use different wording. Some IP Policies also explicitly extend to IP beyond patents and copyrights. Once the IP subject to the IP Policy is defined, the IP Policies categorize when the university has ownership rights. 1. Creations subject to IP policies Some institutions have one IP Policy covering everything, while others have separate IP Policies for different categories of IP. The University of Alabama at Birmingham139 and Michigan State University140 are examples of universities with separate policies for copyrightable and patentable materials. Within the realm of university creations and IP, IP Policies govern the ownership of most creations made at a university.

133

See, e.g., arIz. st. u., supra note 120; Faculty Handbook IP Policy, ColuM., available at http:// techventures.columbia.edu/forms/faculty-handbook-ip-policy (last visited Dec. 23, 2018); Intellectual Property Policy, ga. teCh. res. Corp. u., available at http://www.policylibrary.gatech.edu/faculty -handbook/5.4-intellectual-property-policy (last visited Dec. 23, 2018); University of California Policy – Patent Policy; Copyright Policy, u. of Cal. sys., available at http://www.ucop.edu/research-policy -analysis-coordination/policies-guidance/intellectual-property-ex/index.html (last visited Dec. 23, 2018). 134 563 U.S. 776 (2011). 135 See, e.g., University Policies and Procedures Relating to Innovation, Intellectual Property and New Ventures, u. of va., available at https://lvg.virginia.edu/innovators/policies-procedures (last visited Dec. 23, 2018). 136 See, e.g., Patent and Copyright Agreement, Cal. Inst. of teCh., available at http://www.ogc .Caltech.edu/forms (last visited Dec. 23, 2018); MSU Patent Policy, MICh. st. u., available at http:// www.technologies.msu.edu/researchers/patent-copyright-policy/msu-patent-policy (last visited Dec. 22, 2018) [hereinafter MSU Patent]. Note that the MSU policy leaves out the word “hereby” bringing its effectiveness into question. Id. 137 See, e.g., CWRU Intellectual Property Policy, Case w. reserve u., available at https://case.edu/ research/faculty-staff/tto/ip-policy/ (last visited Dec. 23, 2018); Johns hopkIns u., supra note 121; u. of utah, supra note 122. 138 See, e.g., u. of va., supra note 135; Case w. reserve u., supra note 137. 139 Patent Policy (Rule 5.09), u. of ala. bIrMInghaM, available at http://www.uab.edu/policies/ content/Pages/UAB-RA-POL-0000115.aspx (last visited Dec. 23, 2018). 140 MSU Patent, supra note 136; Development of Copyrighted Materials—Faculty Handbook, MICh. st. u., available at https://technologies.msu.edu/researchers/patent-copyright-policy/msu-copyright -policy (last visited Dec. 23, 2018) [hereinafter Development of Copyrighted Materials].

University technology transfer structure and intellectual property policies 23 A few examples of the IP covered by IP Policies are illustrative of the breadth of coverage and the disparate language used by universities. Beginning with the relatively straightforward, the IP Policy of the California Institute of Technology (Caltech) covers inventions141 made by employees in the line of Caltech duty or with the use of Caltech facilities and copyrightable materials142 developed in the course of Caltech duties. The University of Arizona IP Policy covers “tangible research property and all forms of legally recognized IP including copyrights, patents, trade secrets, trademarks, and plant variety protection.”143 Similarly, the University of Southern California IP Policy covers “copyrights and copyrightable materials, patented and patentable inventions, tangible research property, trademarks, service marks, and trade secrets.”144 Other universities have IP Policies which define coverage in a more detailed form. The Ohio State University IP covers copyrightable materials including “plays, poems, novels, paintings, illustrations, sculptures, and musical compositions” along with inventions and “tangible research property.”145 The Yale University copyright policy covers works of authorship, books, articles, and other written works, as well as musical and dramatic works, pictures, films, videos, sculptures, computer software, and electronic chip designs.146 Similarly, the Princeton IP Policy covers copyrightable works including literary works, musical and dramatic works, pantomimes and choreographic works, and motion pictures and other audio-visual works.147 A few institutions define the scope of their policies based on federal law. The University of Wisconsin-Madison IP Policy148 covers “inventions” as defined by patent law.149 The IP Policy of the University of Massachusetts Medical School in Worcester covers “inventions (discoveries that may be protected under US or other countries’ patent laws); copyrightable works (creative works protectable under US or other countries’ copyright laws), and tangible research materials.”150 Student creations and IP are also subject to IP Policies.151 Such creations include graduate and undergraduate theses, homework assignments, and special and independent study pro-

141

Cal. Inst. of teCh., supra note 121. Copyright and Software Policy, Cal. Inst. of teCh., available at http://www.ogc.Caltech.edu/ forms (last visited Dec. 23, 2018). 143 u. of arIz., supra note 122. 144 University of Southern California Intellectual Property Policy, u. of s. Cal., available at https:// policy.usc.edu/files/2014/02/intellectual_property.pdf (last visited Dec. 23, 2018). 145 Intellectual Property University Policy, ohIo st. u., available at https://tco.osu.edu/v2/wp -content/uploads/IP-Policy.pdf (last visited Dec. 23, 2018). 146 Yale University Copyright Policy, yale, available at https://ocr.yale.edu/faculty/policies/yale -university-copyright-policy (last visited Dec. 23, 2018). 147 prInCeton, supra note 128. 148 Patents and Inventions, u. wIs. MadIson, available at https://www.wisconsin.edu/uw-policies/ uw-system-administrative-policies/inventions-and-patents/ (last visited Dec. 23, 2018). 149 Patent Act, 35 USC §§100-212 (2013). 150 University of Massachusetts Intellectual Property Policy Worcester, u. Mass. Med. sCh. worChester, available at https://www.umassp.edu/sites/umassp.edu/files/content/policies/board/ academic/Intellectual%20Property%20UMW.pdf (last visited Dec. 23, 2018). 151 Abigail Barrow, La Royce Batchelor, Alex Breger, Nathalie Duval-Couetil, Latanya Scott, Jeffrey Skinner, Phyl Speser, & Phil Weilerstein, Managing Student Intellectual Property Issues at Institutions of Higher Education: An AUTM Primer, 2 autM teChnology transfer praCtICe Manual (2014), available at https://autm.net/AUTM/media/TTP/ThirdEditionPDFs/V2/TTP_Manual_3rd_Edition _Volume2_StudentIP.pdf (last visited Oct. 17, 2019). 142

24 Research handbook on intellectual property and technology transfer jects; trade secrets; traditional academic works; the tangible and intangible results of research (data, lab notebooks, charts, biological materials, cell lines and samples), and databases. However, ownership of student creations often varies depending on the student’s university role at the time of creation. 2. Ownership of creations by universities Rules concerning the ownership of IP depend on the type of creation and IP as well as the status of the creator at the time of creation. As a rule, university IP Polices claim ownership of inventions but allow creators of copyrightable works to retain ownership of most copyrights. This division arises because most IP Policies allow creators to own rights to what are called “traditional works” (also referred to as traditional academic works or traditional works of scholarship), which are generally protected by copyright. However, there are recurring exceptions to these rules.152 At base, universities generally own all rights to IP developed as a result of university support or with the use of substantial university resources or facilities. For example, Cornell University’s policy on inventions provides that Cornell owns all inventions made by an inventor with a university appointment working in furtherance of her university responsibilities or using university resources or external grant funding. Cornell University’s copyright policy153 assigns copyrights to works of authorship, created with “substantial use of university resources,”154 to the university. However, the copyright policy allows creators to own traditional academic works. The definition of traditional academic works varies across IP Policies but generally includes scholarly publications and teaching materials. Other forms of traditional academic works mentioned in IP Policies include textbooks, works of art, music, lyrics, sound recordings, photographs, plays, choreography, fiction, and architectural works. The listing of traditional academic works is quite broad because disciplines across a university often have different venues of scholarly outputs. Instructional media or instructional works for teaching classes is another area of copyright in which faculty have traditionally retained ownership. Instructional media may include syllabi, course descriptions, lectures, and materials. Many IP Policies do give the university a non-exclusive license to use these resources.155 One exception to the faculty ownership of instructional materials has become the ownership of online courses created specifically as works made for hire for universities. Universities often carve out a separate policy for this “courseware.” For example, the Penn State IP Policy defines courseware as “a substantially computer-based package of content, assessment materials, and structure for interaction that permits a course to be taught without requiring physical access to a student.” Courseware that

152 For an in-depth overview and critique of the ownership of copyright in the university context, see Jacob H. Rooksby, A Fresh Look at Copyright on Campus, 81 Mo. l. rev. 769 (2016). 153 Copyright Policy, Cornell u., available at https://www.dfa.cornell.edu/sites/default/files/policy/ vol4_15.pdf (last visited Dec. 23, 2018). 154 Substantial use of university resources includes university resources such as “grants, contracts and awards made to the university not ordinarily used by or available to most or all members of the faculty.” Ordinarily available resources include “office space, personal office equipment, office computer workstations, libraries and other general-use information resources, and the means of network access to such resources.” Id. 155 Id.

University technology transfer structure and intellectual property policies 25 is created at the direction of the university, for the university’s ownership and use, belongs to Penn State, but authors own the copyright to courseware that they create without direction by the university.156 Once created, these on-line courses can be used multiple times by automated computer processes. Perhaps surprisingly, only a few IP Policies address what happens when a work or invention is both patentable and copyrightable. Generally, copyright ownership follows patent ownership in those cases. For instance, the Yale IP Policy grants copyright ownership to the author except in certain situations, including when a work is patentable.157 Similarly, NYU owns copyrightable works that are related to patents.158 Many universities’ IP Policies address ownership of computer software specifically. Many IP Policies grant ownership of the software to the university while others rely on the patentable nature of the software to determine ownership. For instance, the Columbia University patent policy grants the university ownership of computer software that is or may be patentable; otherwise there must be an independent basis for asserting ownership rights under the copyright policy.159 The above IP Policy provisions apply to employees rather than students. Students who are not acting as employees have rules separate from university employees. Generally, rights to student-created works vest in the students. For example, the USC IP Policy disclaims ownership of student works created for their education.160 At NYU, students own inventions that are created as part of uncompensated class work; however, student work created in the course of research involving substantial use of NYU resources, sponsored research, or university employment belongs to NYU.161 At Michigan State University, student inventions belong to Michigan State University if the student is employed by the university, made the invention using university funds, or made the invention using university resources not generally available to students.162 Finally, creators tend to own all IP created without substantial use of university facilities or created on their own time as opposed to within the course/scope of their employment (like works for hire). For example, the University of Washington IP Policy does not retain ownership of inventions developed on employee’s own time unless directly related to the business of the university or the university’s actual or anticipated research and development.163 Under the Woods Hole Oceanographic Institution IP Policy, the institution owns IP unless it is developed outside one’s regular duties on personal time.164

156 Intellectual Property, pa. st. u., available at https://policy.psu.edu/policies#Intellectual Property (last visited Dec. 23, 2018). 157 yale, supra note 146. 158 nyu, supra note 123. 159 ColuM., supra note 133. 160 u. of s. Cal., supra note 144. 161 nyu, supra note 123 162 MSU Patent, supra note 136; Development of Copyrighted Materials, supra note 140. 163 Patent, Invention, and Copyright Policy, u. wash., available at http://www.washington.edu/ admin/rules/policies/PO/EO36.html (last visited Dec. 23, 2018). 164 WHOI Intellectual Property Policy and Manual, woods hole oCeanographIC Inst., available at http://www.whoi.edu/fileserver.do?id=134466&pt=2&p=107670 (last visited Dec. 23, 2018).

26 Research handbook on intellectual property and technology transfer D.

Revenue Sharing

Once TTOs receive invention disclosures and finalize assignments of IP, the goal is to find a partner to commercialize the technology. Revenue from those partnerships—in the form of royalties, milestone payments, or proceeds sales of stock in the new company—are distributed amongst various groups at the university. The Bayh-Dole Act requires the university to share a part of the revenue with the inventor or inventors of the invention but does not specify a particular percentage. IP Policies govern the distribution of these royalties both between inventors and other areas of the university as well as amongst any co-inventors. 1. Revenue sharing across the institution When the TTO receives revenue from licensing IP, the money is shared amongst several groups within the university. In addition to the share to the inventor, revenue also may be shared with the general university; the TTO; and the inventor’s department, college/school, and lab. All IP Polices allocate a portion to the inventors as required by the Bayh-Dole Act, but each IP Policy splits the remaining portions differently. In addition to variance among the groups who receive revenue from IP licensing, the share amount each group receives also varies among universities. The University of Maryland system gives inventors 50% of the net revenue, leaving 25% each for the university and TTO.165 The University of Colorado IP Policy divides the university, TTO, and the inventor shares evenly at 25% each.166 In contrast, the University of California system provides for the university to receive 50% of the net revenue and the inventor to receive 35%.167 Additionally, many institutions have tiered systems in which the distribution of revenue changes depending on the amount of net revenue received. For example, some institutions give their investigators 100% of the net royalty revenues for the first $2,500,168 $5,000,169 $25,000,170 or $100,000,171 depending on the institution. Other institutions have flat distributions and do not change based on the amount of net revenues.172 While most IP Policy revenue distributions apply to net revenue from all IP, some institutions identify separate royalty sharing arrangements for copyright policies as compared to patent policies.173 These revenue sharing provisions indicate the value that research institutions place on the contributions of the inventors—both disclosure and help with commercialization. The IP

165

University of Maryland Intellectual Property Policy, u. of Md. sys., available at https://president .umd.edu/sites/president.umd.edu/files/files/documents/policies/IV-320A.pdf (last visited Dec. 23, 2018). 166 Intellectual Property Policy on Discoveries and Patents for Their Protection and Commercialization, u. of Colo. sys., available at https://www.cu.edu/sites/default/files/1013.pdf (last visited Dec. 23, 2018). 167 u. of Cal. sys., supra note 133. 168 See, e.g., ga. teCh. res. Corp., supra note 133. 169 See, e.g., Patent Policy, rutgers u., available at https://policies.rutgers.edu/sites/default/files/50 .3.1%20-%20current.pdf (last visited Dec. 23, 2018). 170 See, e.g., CarnegIe-Mellon u., supra note 120; Intellectual Property Policy, eMory u., https:// policies.emory.edu/7.6 (last visited Dec. 23, 2018). 171 See, e.g., Iowa st. u., supra note 124. 172 See, e.g., baylor Coll. of Med., supra note 130; harv., supra note 131; MIt, supra note 121. 173 See, e.g., u. of pItt. pItt., supra note 124; u. of utah, supra note 122.

University technology transfer structure and intellectual property policies 27 Policies give anywhere from 25% to 100% of the net royalty revenues back to the investigators. Additionally, while the Bayh-Dole Act requires investigators to receive a portion of royalties for inventions arising as a result of federally sponsored projects, IP Policies and their revenue sharing provisions apply regardless of funding sponsor. Revenue sharing amongst co-creators 2. Innovations from universities rarely result from the efforts of an individual inventor. Rather, multiple investigators from one or more laboratories work as a team to create new technologies. Therefore, the share of net revenue that goes to the inventor under a university IP Policy often must be subdivided among multiple inventors. Complicating that calculation is the fact that not every inventor contributes the same amount to any invention.174 Some, though not all,175 IP Policies do address this revenue distribution among co-inventors. Many IP Polices that identify a method for revenue distribution amongst co-inventors rely on the inventors to draft agreements among themselves as to how the revenues should be distributed.176 Other IP Policies default to equal shares amongst co-inventors.177 Most IP Polices contain a combination of options for revenue distribution. Specifically, these IP Polices require the division of revenues equally between co-inventors unless the inventors have made some other arrangement in a written agreement.178 Interestingly, two institutions flip this allocation.179 In the Harvard and Michigan State University IP Policies, the default revenue distribution has the co-investigators agree in writing as to the distribution of royalties, and only if no agreement can be negotiated does the university begin splitting revenues equally. While these two IP Policies reflect the same two mechanisms, the flip between primary and secondary mechanism is interesting as a reflection on the value some institutions place on encouraging their investigators to discuss and form an agreement compared to viewing all co-investigators as equal contributors to their joint inventions. A few IP Policies use some outlier methods for determining revenue sharing. Those methods include having a designated university office or authority to determine the distribution,180 requiring the co-inventors to designate percentages on their invention disclosure form,181 and distributing to co-inventors in proportion to their contribution toward the discovery or creation.182

174

In order to be a co-inventor a person must contribute to at least one patent claim. See, e.g., u. of ala. bIrMInghaM, supra note 139; eMory u., supra note 170; FSU-6.009 Inventions and Patents, fla. st. u., available at http://policies.fsu.edu/sites/g/files/upcbnu486/files/ regulations/adopted/FSU-Chapter-6.pdf (last visited Dec. 23, 2018). 176 See, e.g., Colo. st. u., supra note 121; ColuM., supra note 133; Policy on Technology and Literary and Artistic Works, vand. u., available at https://www.vanderbilt.edu/faculty-manual/part -iii-university-principles-and-policies/ch4-policy-on-technology-and-literary-and-artistic-works/ (last visited Dec. 23, 2018). 177 See, e.g., arIz. st. u., supra note 120. 178 See, e.g., MIt, supra note 121; nw. u., supra note 120; u. of arIz., supra note 122. 179 harv., supra note 131; MSU Patent, supra note 136. 180 See, e.g., Intellectual Property Management and Commercialization, tex. a&M researCh found., http://policies.tamus.edu/17-01.pdf (last visited Dec. 23, 2018) (if no written agreement exists). 181 See, e.g., ga. teCh. res. Corp. u., supra note 133. 182 See, e.g., pa. st. u., supra note 156. 175

28 Research handbook on intellectual property and technology transfer E.

Dispute Resolution Procedures

No matter which provision of the IP Policy is being enforced, the opportunity for conflict arises. An inventor or author may dispute whether the IP Policy requires her creation to be owned by the university or how to best commercialize the creation to either maximize revenue or the social benefit. Co-inventors may dispute the appropriate revenue distribution amongst themselves.183 In order to handle disputes, many IP Policies contain dispute resolution procedures. These dispute resolution procedures vary widely between university IP Policies. However, of the institutions that include such procedures, there is a common thread. Generally, the procedure consists of an initial decision followed by an appeals process. Initial dispute resolution proceedings are typically handled by a designated committee.184 These committees may be composed of faculty, members of the university administration, and perhaps a technical staff member or a student.185 However, not all universities designate committees. Rather than a designated committee, some universities may designate an administrative officer such as the vice president of technology transfer186 or the vice president of research to render the initial decision.187 Appeals from the initial decision are even more varied. IP Policies include appeals before the provost,188 the president of the university,189 or even the university board of trustees.190 The University of Pennsylvania policy states the vice provost for research considers issues de novo.191 Additionally, not all subject matter may be heard on appeal. For example, the University of Arizona explicitly exempts from review decisions regarding whether to pursue patent protection and decisions regarding payment of prosecution expenses as it affects net revenue. The University of North Carolina at Chapel Hill provides for a three-stage appeal.192 The TTO director decides initial disputes. The vice chancellor for research decides appeals. The university’s provost appoints a committee to assist in the decision on an appeal from the vice chancellor’s decision. This committee may include students “qualified to evaluate and 183 Such a dispute is particularly fraught with danger when the co-inventors have a power imbalance such as a principal investigator faculty member and his graduate student. Jennifer Carter-Johnson, supra note 67. 184 See, e.g., u. of arIz., supra note 122; baylor Coll. of Med., supra note 130; Policies Relating to Research Integrity, Intellectual Property, Accessibility, and Insurance at the University, u. of ChI., available at https://provost.uchicago.edu/handbook/research/research-policies (last visited Dec. 23, 2018). 185 See, e.g., CarnegIe-Mellon u., supra note 120. 186 See, e.g., Cornell u., supra note 122. 187 See, e.g., MIt, supra note 121. 188 See, e.g., Cal. Inst. of teCh., supra note 121. 189 See, e.g., arIz. st. u., supra note 120. 190 See, e.g., Duke University Policy on Intellectual Property Rights, Patent Agreement, Policy on Inventions, Patents, and Technology Transfer, duke u., available at https://policies.duke.edu/research/ institutional/property.php (last visited Dec. 23, 2018). 191 Patent and Tangible Research Property Policies and Procedures of the University of Pennsylvania, u. of pa., available at http://pci.upenn.edu/wp-content/uploads/2017/12/Patent-Policy-Most-Recent.pdf (last visited Dec. 23, 2018). 192 University of North Carolina at Chapel Hill, Patent & Invention Policy, u. of n.C. Chapel hIll, https://unc.policystat.com/policy/4466280/latest/ (last visited Dec. 23, 2018).

University technology transfer structure and intellectual property policies 29 resolve such dispute.” The provost makes the final and binding decision after the committee’s recommendation.

V.

PROBLEMS FOR INSTITUTIONS WITH TTOS

Technology transfer is a complicated business. As discussed in sections III and IV, universities must determine the appropriate structure for the TTOs and write the IP Policies that govern the process. Once those major hurdles are surpassed, still other issues arise. Institutions must determine how to incentivize disclosure from a possibly disinterested faculty member; to assign inventorship and revenue allocation among co-creators; to balance competing goals within the IP Policies; and to work with other institutions. A.

Incentivizing Disclosure

The one obligation that almost every IP Policy includes is the disclosure of creations to the TTO. The underlying basis of the university technology-transfer system relies on inventor researchers to disclose patentable and licensable innovations. Without disclosure, innovations are generally published in scientific journals and thereby dedicated to the public domain. However, faculty support of the technology-transfer process has lagged university investment in many universities where as many as 50% of patentable innovations are not disclosed by researchers to their university TTO.193 Other universities have developed a rich academic culture that has fully embraced technology transfer.194 Such disregard for the duty to disclose suggests an imbalance between incentives to disclose and other influences on researchers. Considering the existing disincentives to comply with the obligations to disclose, TTOs and universities should craft incentives specifically to combat those issues. The major incentive to disclose is revenue sharing of a licensed patent. Disincentives to comply with disclosure requirements are varied and include social norms, time-management issues, and a lack of education about the duty and the monetary incentives. The disclosure incentive is generally monetary—a slice of the licensing revenues, as is required by the

193

See Richard A. Jensen, Jerry G. Thursby, & Marie C. Thursby, Disclosure and Licensing of University Inventions: “The Best We Can Do with the S**t We Get to Work With”, 21 Int’l J. Indus. org. 1271, 1272 (2003); Jerry G. Thursby & Marie C. Thursby, “Pros and Cons of Faculty Participation in Licensing” in 16 University Entrepreneurship and Technology Transfer: Process, Design, and Intellectual Property (Advances in the Study of Entrepreneurship, Innovation, and Economic Growth) 187, 189 (Gary D. Libecap ed., 2005) [hereinafter Pros and Cons of Faculty Participation in Licensing]; Albert N. Link, Donald S. Siegel, & Barry Bozeman, An Empirical Analysis of the Propensity of Academics to Engage in Informal University Technology Transfer, 16 Indus. & Corp. Change 641, 642–3 (2007). 194 Such universities include Stanford University, Massachusetts Institute of Technology (MIT), California Institute of Technology, Cornell University, and Georgia Institute of Technology. Paul R. Sanberg, Morteza Gharib, Patrick T. Harker, Eric W. Kaler, Richard B. Marchase, Timothy D. Sands, Nasser Arshadi, & Sudeep Sarkar, Changing the Academic Culture: Valuing Patents and Commercialization Toward Tenure and Career Advancement, 111 proC. nat’l aCad. sCI. u s a. 6542 (2014).

30 Research handbook on intellectual property and technology transfer Bayh-Dole Act for federally-funded inventions.195 In maximizing the number of disclosures that they receive, TTOs should understand the disincentives to disclose so that they can craft appropriate incentives. Pressures inherent in the academic research environment often serve as disincentives to disclose creations. Social norms, lack of education, and a perception that time is better spent elsewhere all work together to offset the potential monetary incentive to disclose.196 Perhaps the most discussed disincentive to disclosure is the conflict of academic technology transfer with scientific social norms.197 Norms particular to academic researchers appear at the surface to be antithetical to the current academic technology transfer structure. Robert Merton described four basic scientific norms: communalism, universalism, disinterestedness, and organized skepticism.198 Of these, communalism and disinterestedness are most applicable to discussions of incentives to disclose. Communalism, the idea that there is a common ownership of scientific discoveries, directly conflicts with the private ownership of IP. Communalism would suggest that research results should be published and that these results should be freely usable to other researchers. Communalism further suggests that the proper reward for scientific discovery is recognition and esteem.199 Katherine Strandburg has refined the communalism concept, applying it to more specific situations within academic research and technology transfer such as sharing research tools and materials as examples of communalism.200 Paralleling this view is empirical data suggesting that academic researchers often ignore patent protection on research tools which can be re-created in their own laboratories.201 These problems based in communalism objections could be partially offset by educating researchers about IP. For instance, patent protection is not unlimited. Patents are available for “anything under the sun made by man,” but not laws of nature, natural phenomena, or abstract ideas.202 These restrictions mean that the basic building blocks of science are not subject to patent monopoly. Only specific technological advances can be patented. Furthermore, not every advance need be patented. TTOs should work with inventors to determine if an invention can be made available to the public without exclusive patent licenses. Similarly, copyright protects a particular expressive work, but that protection does not extend to the general ideas behind the work. Therefore, Harry Potter and the Philosopher’s Stone203 is copyrighted, but the idea of an orphaned child (muggle or magical) who grows up to save the world is not. By educating researchers on the limits of IP protection, TTOs may be able to offset some of the communalism-based objections to technology transfer.

195 See generally 35 U.S.C.A. §§ 200–204 (West, Westlaw through P.L. 114-114 (excluding P.L. 114-92, 114-94, and 114-113) 2016). 196 For a full discussion of the disincentives to disclose, see generally Carter-Johnson, supra note 67. 197 See, e.g., Pros and Cons of Faculty Participation in Licensing, supra note 193, at 189 (“[S]ome faculty may refuse to disclose for ‘philosophical’ reasons related to their notions of the proper role of academic scientists and engineers.”). 198 Robert K. Merton, The Sociology of Science: Theoretical and Empirical Investigations (1973). 199 Thursby, et al., supra note 46, at 62–3. 200 Strandburg, supra note 11, at 2248. 201 Id., at 2239. 202 Diamond v. Chakrabarty, 447 U.S. 303 (1980). 203 J.k. Rowling, harry potter and the phIlosopher’s stone (1997).

University technology transfer structure and intellectual property policies 31 Communalism would also suggest that disclosure and publication delays would be anathema to a university researcher. Yet timing for conference attendance and publications can be tricky to coordinate with patent requirements. Under US patent law, publication of research more than a year before application for a patent will bar grant of the patent.204 Other countries are considerably stricter, having a publication bar in which publication of the innovation before applying for the patent results in an immediate bar of patent grant. Therefore, academic researchers may fear that disclosure of innovations may result in requirements that they delay publication so that patent applications can be timely filed.205 Some evidence supports this trepidation on the part of academic researchers. One study found that almost 20% of disclosing researchers in the life sciences have experienced publication delays of greater than six months due to TTO patenting.206 Thus, academic researchers may strategically fail to disclose innovations in the face of publication delay fears.207 TTOs are in the best position to address these timing issues. For inventions with only a US market, publications and presentations merely start the one-year clock for filing a patent. TTOs should be able to work within that timeline to file the appropriate patent applications. Inventions with a worldwide market may be more difficult since any disclosure may bar patenting. In those cases, TTOs should prioritize the ability, either through onsite staff or outside counsel, to file a quick provisional patent application. In fact, filing a provisional patent application in either case would allow a TTO more time to analyze the invention and its particular market without appreciably delaying researcher disclosure. Disinterestedness is the idea that scientists should act in ways that are selfless, i.e., they should have no financial attachment to their research. Disinterestedness runs afoul of one of the primary incentives to disclose—revenue sharing. Disinterestedness immediately conflicts with monetary incentives for the creator. Strandburg’s work supports this preference for selflessness by identifying preferences of academic researchers for “performing scientific research and participating in the scientific discourse” and “learning the results of the collective research project.”208 Again, education may be the key to offsetting disinterestedness. There certainly are benefits to the creators and the university in protecting and licensing IP; however, the underlying impetus for the Bayh-Dole Act was to provide a structure for the access of new creations by the public. Before the Bayh-Dole Act and the patent/licensing structure of TTOs, new creations languished on shelves and in dusty tomes as there were no incentives for private industry to

204

35 U.S.C. 102. See, e.g., Intellectual Property Policy, wash. u., available at https://wustl.edu/about/compliance -policies/intellectual-property-research-policies/intellectual-property/ (last visited Dec. 23, 2018) (“The publication of research results must not be hampered by agreements made to commercialize intellectual property. However, a minimal and defined delay to protect intellectual property through patent applications may be included.”). 206 See David Blumenthal, et al., Withholding Research Results in Academic Life Science, 277 JaMa 1224 (1997). 207 Also see Carter-Johnson, supra note 67, at 483 (“[I]n spite of a government requirement to disclose government-funded inventions to the university for licensing and the university’s considerable interest in licensing such inventions, academic researchers routinely publish their inventions in scientific journals without university disclosure rather than spending the extra time required to also disclose the inventions to the university.”). 208 Strandburg, supra note 11, at 2249. 205

32 Research handbook on intellectual property and technology transfer invest in development. Additionally, an engaged faculty can help direct humanitarian licensing efforts as discussed below. Violation of these norms can also have consequences for the academic researcher. Tenure, publication, and grant approvals are all peer-reviewed processes. Each of these areas—all critically important to the career of an academic researcher—are opportunities for research society to punish those seen as deviating from the scientific norms. Aside from scientific norms, another disincentive to disclose may be the simple balancing of the time investment necessary for disclosure and the expectations of recoupment.209 Disclosure may initially be completed using a form document from the TTO, but assignments, patent preparation conferences, and marketing meetings take up more time as the invention moves through the technology transfer process.210 For many academic researchers, this time balance may weigh strongly in favor of failing to disclose, since working with the TTO takes time away from research and publication. While an invention disclosure possibly211 may result in monetary gain in the distant future, the immediate goals of a researcher to receive more funding and job stability, such as tenure, may lead an academic researcher at any level to decide that publication or further grant writing is a better use of her time than filing an invention disclosure.212 Such a decision is logical given the lack of import that technology transfer activities traditionally are given in tenure decisions.213 As a result, academic researchers sometimes publish their inventions in scientific journals without university disclosure rather than spending the extra time required to also disclose the inventions to the university.214 In order to overcome the time pressure on university researchers, TTOs should look for support from the broader university. For instance, tenure requirements typically include some mix of research, teaching and service.215 At the very least, disclosure and work with TTOs toward commercialization should be considered service toward tenure requirements and perhaps allow the researcher some service relief. Better would be to count patents as part of the research portfolio at tenure as some universities are beginning to do.216 Writing a patent

209

Carter-Johnson, supra note 67. Most disclosed inventions are at “embryonic” stages and require considerable additional research and development before being ready for commercialization. See Thursby, et al., supra note 46, at 62 (“71% of licensed inventions are viewed as requiring inventor cooperation for commercial success.”). 211 Many academic researchers do not perceive that any direct monetary benefit will be forthcoming. They are likely correct. See Carter-Johnson, supra note 67 (discussing how many disclosures make money). 212 For a more in-depth discussion of the motivations of academic researchers, see Carter-Johnson, supra note 67. 213 See, e.g., Donald S. Siegel, David A. Waldman, Leanne E. Atwater, & Albert N. Link, Commercial knowledge transfers from universities to firms: Improving the effectiveness of university–industry collaboration, 14 J. hIgh teCh. Manage res. 111 (2003); Saul Lach & Mark Schankerman, Incentives and invention in universities, 390 rand J. eCon. 403 (2008); Elisabeth Pain, Science careers, Playing well with industry, 319 sCI. 1548 (2008). 214 Daniel W. Elfenbein, Publications, Patents, and the Market for University Inventions, 63 J. eCon. behav. & org. 688, 689 (2007); Jensen, Thursby, & Thursby, supra note 193, at 1272. 215 For a discussion of tenure and promotion policies and recommendations on the use of patents therein, see Sanberg et al., supra note 194. 216 For example, Texas A&M, University of Maryland (system), University of Minnesota, Arizona State University, The University of Arizona, and The Ohio State University are a few universities that consider patenting and technology transfer activities as a significant factor for tenure. Id. 210

University technology transfer structure and intellectual property policies 33 disclosure that caused a delay in publication should be viewed as scholarly output.217 However, few institutions have consistent, unambiguous policies as to how patent and technology transfer efforts count toward tenure.218 The major incentive that TTOs use to incentivize disclosure is to share revenue among all co-inventors when the technology is brought to market. This monetary incentive is mandated by the Bayh-Dole Act. Universities have the freedom to set revenue sharing percentages, but studies have shown that the amount of revenue shared has a relatively modest impact on the rates of disclosure.219 In the face of the disincentives to disclose, the possibility of future revenue sharing from licensed patents begins to seem weaker. A more immediate payout in the form of extra laboratory support for each patentable invention disclosed could yield increases in disclosure if researchers felt there was immediate compensation for their time even if the disclosure was not pursued by the TTO. Tying the money to laboratory support could help pre-tenure faculty, as the disclosure could be framed similar to a grant application. In the end, universities and TTOs must find ways to balance the incentives and disincentives to disclosure in order to have a successful technology transfer program. It is likely that the approach will be different at each university and will evolve as the academic culture at the university evolves. An understanding of academic researcher concerns and a comprehensive university IP education program will lead to the best possible disclosure rates. B.

Determining Inventorship and Allocating Revenue

Once the university researchers have made the decision to disclose, the next hurdle in the technology-transfer process is to determine who should be a named inventor on the patent. The lack of education in the patent process may indeed be a disincentive for disclosure in the first place. Perhaps more importantly, lack of education combined with inherent biases in the research community may result in the omission of inventors from a patent application and problems with allocation of shares of revenue among co-inventors.

217

CS Renault, Academic capitalism and university incentives for faculty entrepreneurship, 31 J. teCh. transfer 227 (2006). 218 Judy Genshaft, Jonathon Wickert, Bernadette Gray-Little, Karen Hanson, Richard Marchase, Peter E. Schiffer, & R. Michael Tanner, Consideration of Technology Transfer in Tenure and Promotion (2016), Iowa u., available at http://lib.dr.iastate.edu/provost_pubs/1 (last viewed Dec. 20, 2018); Sanberg, et al., supra note 194. 219 See, e.g., Donald S. Siegel, David Waldman, & Albert Link, Assessing the Impact of Organizational Practices on the Relative Productivity of University Technology Transfer Offices: An Exploratory Study, 32 res. pol’y 27, 44–5 (2003); Joseph Friedman & Jonathan Silberman, University Technology Transfer: Do Incentives, Management and Location Matter?, 28 J. teCh. transfer 17, 29 (2003) (Showing a positive but weak correlation of license revenue share incentives to faculty researchers with the number of licenses executed, and a strong correlation with license income. This discrepancy may be due to a skewing of the data by one or more “blockbuster” inventions or could also be due to limits on TTO resources to execute more licenses.); Albert N. Link & Donald S. Siegel, Generating Science-Based Growth: An Economic Analysis of the Impact of Organizational Incentives on University-Industry Technology Transfer, 11 eur. J. fIn. 169, 179 (2005); Pros and Cons of Faculty Participation in Licensing, supra note 193, at 192; Saul Lach & Mark Schankerman, Incentives and Invention in Universities, 39 rand J. eCon. 403, 404 (2008) (showing that license revenue sharing with scientists strongly affects licensing outcomes).

34 Research handbook on intellectual property and technology transfer To understand the problems with invention determination in the university technology-transfer process, it is first important to understand inventorship and the underlying patent laws. As confirmed in Stanford v. Roche, ownership of a patent initially vests in the inventor. Therefore, for a university to perfect its ownership of a patent, it must have an assignment from every inventor. An inventor is determined by conception.220 The Federal Circuit has held that “[c]onception is the formation in the mind of the inventor, of a definite and permanent idea of the complete and operative invention, as it is hereafter to be applied in practice.”221 Conception may encompass multiple people, resulting in multiple inventors. Joint inventorship has been defined as “the product of a collaboration between two or more persons working together to solve the problem addressed.”222 However, determining joint inventorship is not easy. Counterintuitively, multiple researchers can be joint inventors on a patent even if “(1) they did not physically work together or at the same time, (2) each did not make the same type or amount of contribution, or (3) each did not make a contribution to the subject matter of every claim of the patent.”223 Each person only has to perform “part of the task which produces the invention.”224 Confounding this analysis, a patent usually contains multiple claims, each relating to a different aspect of the invention. The conception for each claim may have happened at different times and often is attributed to several inventors, each of which must have contributed conceptually to at least one of the claims in the patent.225 Additionally, since the conceived invention changes over time, initial conception is often not an indication of all the inventors.226 As difficult as it is for a patent attorney or court to determine inventorship, it can be even more vexing for a scientific researcher to determine who should be included as an inventor. Problems in the technology-transfer process may arise due to ambiguities in inventorship. Without an assignment from all inventors, a non-acknowledged inventor could sue for inclusion and license his rights to a competitor.227 Due to the definition of inventorship and the complexities of the modern university research environment, inventions often include conceptual and creative contributions by many people building on an initial idea. In determining inventorship, universities often rely on the input from researchers themselves. But again, these researchers lack patent law training, and their

220

Ethicon, Inc. v. U.S. Surgical Corp., 135 F.3d 1456, 1548 (Fed. Cir. 1998). Hybritech, Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1376 (Fed. Cir. 1986) (internal quotation marks omitted); Burroughs Wellcome Co. v. Barr Labs., Inc., 40 F.3d. 1223, 1227–8 (Fed. Cir. 1994) (“Conception is the touchstone of inventorship …”). 222 Burroughs Wellcome Co., 40 F.3d. at 1227. 223 35 U.S.C.A. § 116(a) (West, Westlaw through P.L. 114-114 (excluding P.L. 114-92, 114-94, and 114-113) 2016). 224 Ethicon, Inc., 135 F.3d at 1460. 225 See 35 U.S.C.A. § 116(a) (West, Westlaw through P.L. 114-114 (excluding P.L. 114-92, 114-94, and 114-113) 2016); Ethicon, Inc., 135 F.3d at 1548 (“[E]ach joint inventor must generally contribute to the conception of the invention.”). 226 For a full discussion of how to determine patent inventorship, see generally Christopher McDavid, I Want a Piece of That! How the Current Joint Inventorship Laws Deal with Minor Contributions to Inventions, 115 penn st. l. rev. 449 (2010). 227 Jennifer Carter-Johnson, Intellectual Property Revenue Sharing As a Problem for University Technology Transfer, 49 akron l. rev. 647 (2016). 221

University technology transfer structure and intellectual property policies 35 designation of joint inventors may be legally incorrect. If so, the result is improper patent prosecution and assignments. Problems associated with the failure to list an inventor often arise because faculty researchers deny that non-faculty members, particularly graduate and undergraduate students, are inventors. Some faculty researchers have blatantly declared themselves the sole inventor despite several graduate student and post-doctoral researcher co-authors on the papers at the base of the innovation. Other faculty have denied that any innovative work in the lab comes from non-faculty.228 Problems with inventorship may lead to costly legal disputes. For example, in Chou v. University of Chicago,229 Dr. Chou’s faculty advisor failed to name her as an inventor on a patent and fired her when she pressed her claim for inventorship. Dr. Chou then sued for correction of inventorship in order to have her name added to the patent as a joint inventor so that she could receive a portion of the licensing revenue. On appeal, the Federal Circuit held that her faculty advisor had a fiduciary duty to his student with regards to giving her credit as joint inventor on the patent application. These sorts of problems between faculty and non-faculty inventors can be viewed as power imbalances that affect the ability of the non-faculty inventor to negotiate inventorship credit.230 More so than in many employment situations, the faculty researcher holds a great deal of power over the future career prospects of post-doctoral fellows and students in their laboratories. Due to the apprentice-like structure of graduate science programs, the faculty member controls degree prospects of students and publishing abilities of both students and post-doctoral fellows. For the non-faculty researcher, this relationship makes bargaining for inventorship credit problematic as the faculty member has a great amount of power over the non-faculty researcher. Similarly, such power dynamics may impact negotiating revenue allocation among co-inventors. Once an invention is disclosed to a university, the inventorship issues are resolved, and a patent is issued to the university, the invention may be licensed for commercialization. Fortunate universities and inventors will receive revenue based on the patented inventions. As described in section IV, revenue from licensing the patent must be shared with the inventors in order to comply with the Bayh-Dole Act. However, neither the Bayh-Dole Act nor its implementing regulations define how such revenue must be shared with any given inventor, even though the Code of Federal Regulations specifically recognizes that co-inventors exist by including reference to federal employee co-inventors in the revenue sharing requirements.231 Since many university patents have multiple named joint inventors, this revenue sharing requirement necessitates that the university and/or inventors determine an allocation mechanism for revenue amongst multiple competing interests. As described in section IV, most IP

228 Corynne McSherry, Who Owns Academic Work? Battling for Control of Intellectual Property, 183 (2001) (quoting a faculty researcher describing inventorship in his laboratory: “I think there’s rarely more than one inventor … if you wake up and you have an idea, that’s the invention … The postdoctoral researchers contributed to the work [around the idea], but they didn’t do any really innovative work such as contributing new concepts, [or] coming up with something that, in my lab, I haven’t thought about.”). 229 Chou v. Univ. of Chi., 254 F.3d 1347 (Fed. Cir. 2001). 230 Carter-Johnson, supra note 67, at 676. 231 37 C.F.R. § 401.14(k)(2) (2016) states that “[t]he contractor will share royalties collected on a subject invention with the inventor, including Federal employee co-inventors” (emphasis added) but does not specify how those royalties should be divided.

36 Research handbook on intellectual property and technology transfer Policies rely on written agreements among co-inventors or default to an equal share for each co-inventor. To understand the best way to allocate revenue among joint inventors, it is important to understand the attitudes and interests of the various parties.232 Joint inventors may be any combination of laboratory members described above—faculty members, graduate students, post-doctoral fellows, to name a few. Additionally, the monetary incentives geared toward disclosure are more immediate when discussing revenue allocation. An individual inventor trying to determine her fair revenue allocation based upon her contribution would likely have immediate monetary incentive to assess her contribution. However, as in the disclosure context, the monetary incentives may not offset other pressures on an individual to minimize her contribution. For instance, equitable distribution may be a difficult argument if the one co-inventor has trouble believing his co-researcher is a joint inventor. As discussed above, university researchers are not educated as to the rules concerning inventorship. This lack of education may result in overestimation of some contributions to the invention with concomitant underestimation of other contributions. Due to a lack of education, these mis-estimations are likely exacerbated by claim amendments during prosecution that narrow the scope of the disclosed invention.233 Other pressures on negotiating revenue distributions often arise when the parties negotiating the agreement for revenue distribution have uneven negotiation power.234 For example, a graduate student who is a joint inventor with her PI is likely to face several pressures to minimize her revenue allocation similar to the battles fought for recognition of inventor status. For instance, it has been reported that faculty often undervalue the contributions of student researchers. Unlike inventorship, there are no legal rules requiring any particular revenue allocation for a named inventor. Fairness and equity are the best arguments for sharing revenue above a minimum amount.235 Negotiating a revenue allocation under those circumstances can be quite difficult when views of contribution are not matched. Making the revenue allocation agreement more difficult is the negotiation power imbalance that exists between the faculty and non-faculty inventors. Even an immediate monetary incentive is unlikely to overcome a graduate student’s reluctance to anger her PI when project assignments and future recommendations are more valuable in the long term. With these competing interests and power imbalances in mind, universities must determine how to equitably define a distribution of revenue amongst the joint inventors. In whichever way a university decides to define the revenue distribution, it must be prepared to enforce it fairly and according to its terms. The case of Charest v. Harvard236 serves as an example of a what may happen when co-inventors cannot agree and the university does not enforce its IP Policy as written. In June 2013, Dr. Mark Charest, a chemistry PhD student who graduated from Harvard in 2004, sued the university along with Andrew Myers, his PhD advisor. The lawsuit arose due to the

232

Carter-Johnson, supra note 227. Id. 234 Carter-Johnson, supra note 67, at 676. 235 Carter-Johnson, supra note 227. 236 Complaint, Charest v. President of Harvard Coll., 2016 WL 614368 (D. Mass. Feb. 16, 2016) (No. 1:13-cv-11556). 233

University technology transfer structure and intellectual property policies 37 royalties associated with a patent covering a new synthetic method for producing antibiotics, which became the basis for Charest’s dissertation and first-author paper in Science. As with all such research inventions, the method was assigned to the sponsoring university, in this case, Harvard. From there, the Harvard TTO patented the method and licensed it to Tetraphase Pharmaceuticals. The distribution of royalties from the Tetraphase license led to the dispute. As noted in section IV, Harvard’s IP Policy requires the university to distribute royalties equally among all of the inventors on a patent unless the inventors agree to a different distribution. Harvard’s TTO asked Charest and his former labmates to voluntarily accept a distribution of 50% to Myers, 15% to Charest, 15% to Dionicio Siegel, 15% to Christian Lerner, and 5% to Jason Brubaker (the five co-authors of the paper) rather than an equal split of 20% each. The four non-faculty co-authors did not believe this to be an equitable split and agreed amongst themselves to a distribution of 18.75% to Charest, 11.25% to Siegel, 10% to Lerner, and 10% to Brubaker. Myers refused to participate in the royalty negotiation and maintained that his 50% share was not open for discussion. Charest initially refused to accept the unequal distribution of the royalties. When he began discussions with the Harvard TTO, Charest claimed that Harvard threatened to directly cut Charest’s share of the royalties or to shift the distribution of licensing payments to a second patent on which Charest was not listed as an inventor. In addition, Myers pressured Charest to accept the royalty distribution, using advice such as “tread lightly,” “be careful,” and “think about [your] career.” In light of the pressure from Harvard and his PhD advisor, Charest signed an agreement to accept 18.75% of the royalties for the first patent. The second patent never materialized, and Charest asserted in his complaint his belief that it was a ruse fabricated to force his hand to volunteer to let Myers get a 50% cut of the royalties. As in Charest, problems may arise when universities ignore the stated policy. For instance, both Harvard and the PI believed that the PI should have 50% of the royalties, leaving the other four non-faculty joint inventors to share the remaining 50%. When Charest disagreed, preferring the default 20% division, Harvard stepped in and pushed for the agreement in spite of its policy to the contrary. Thus, university technology-transfer policies, while almost completely discretionary, need to be a statement of the true university policy. These policies should address the competing interests of the various parties to the technology-transfer process, recognizing that all inventors are not a monolithic group. C.

Social Responsibility v. Licensing for Income

Many IP Policies have stated goals that encompass both commercialization of university technology and enhancing the public good. In many ways, licensing is one of the major ways to transfer university inventions into the public sphere because universities do not produce and sell goods. In fact, the Bayh-Dole Act contains a requirement that universities preferentially license to small businesses as a method to grow the economy. However, sometimes the cost of patented technologies makes access out of reach. At that point, TTOs must determine how to walk the line between working with private companies to see the technology developed into products and making the product available to benefit the general public.

38 Research handbook on intellectual property and technology transfer The story of the HIV anti-retroviral drug Zerit® illustrates this problem.237 A critical part of the standard AIDS treatment, Zerit® was developed and patented at Yale University. Unfortunately, many in the developing world could not afford the medication. In 2001, the Yale community pressured the exclusive licensee, Bristol-Myers Squibb, to sell the drug in Africa at a nominal cost. In the years following, more universities have declared their support for such socially responsible licensing. Together with top research universities,238 the Association of American Medical Colleges released the 2007 Nine Points to Consider in University Licensing.239 Most pertinent for this discussion is Point Nine, which states “Consider including provisions that address unmet needs, such as those of neglected patient populations or geographic areas, giving particular attention to improved therapeutics, diagnostics and agricultural technologies for the developing world.” In discussing Point Nine, the document refers explicitly to the university’s “social compact with society.” Similarly, in 2011, the National Academies echoed this idea of a need for universities to promote the public good.240 The first finding in the National Academies report stated that “[t]he first goal of university technology transfer involving IP is the expeditious and wide dissemination of university-generated technology for the public good.” The report went on to endorse both the Bayh-Dole Act and its impact on technology transfer as well at the Nine Points document. The report went on to say that “[p]atenting and licensing practices should not be predicated on the goal of raising significant revenue for the institution” but rather “to encourage the widest dissemination of university-generated technology for the public good.” In this era of shrinking funding, raising funds in order to educate students and fund research while limiting tuition increases is also a public good. Additionally, it is unclear how well universities manage the balance between the pursuit of profit and the public good. The 2013 case of Association for Molecular Pathology v. Myriad Genetics241 arose in part because patients claimed that they could not afford access to breast cancer screening diagnostics based on a gene that had been patented by the University of Utah and licensed to Myriad.242 One problem is that technology licensed by universities is typically very early stage and requires the licensee to invest into more research and development before bringing the technology to market. It is difficult to negotiate a license in which the licensee agrees to invest resources and then give up exclusivity.

237

Jennifer Carter-Johnson, Jeffrey Carter-Johnson, & Jorge L. Contreras, “University Research and Licensing” in Bioinformatics Law: Legal Issues for Computational Biology in the Post-Genome Era, 107 (Jorge Contreras & A. James Cuticchia eds., 2013). 238 Signatories included California Institute of Technology, Cornell, Harvard, Massachusetts Institute of Technology, the University of California system, the Chicago and Urbana-Champaign campuses of the University of Illinois, University of Washington, Wisconsin Alumni Research Foundation and Yale. 239 Nine Points to Consider in Licensing University Technology, available at http://news.stanford .edu/news/2007/march7/gifs/whitepaper.pdf (last visited Dec. 23, 2018). 240 Managing Intellectual Property in the Public Interest, nat’l aCadeMIes of sCI., available at https://www.nap.edu/read/13001/chapter/2 (last visited Oct. 17, 2019). 241 Ass’n for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576 (2013). 242 For a fuller discussion of Myriad, see The Branding of the American Mind, supra note 55, at 122–30.

University technology transfer structure and intellectual property policies 39 Licensing structures exist to help deal with international access to technology. For instance, rather than granting a worldwide license for a technology or medicine, developing countries could be excluded from exclusive license grants, allowing for low cost licenses in those countries once the technology is developed. Alternatively, a worldwide license with required sublicenses to local producers in developing countries would allow the licensee to control the implementation of the technology. Additionally, retaining and using university private march-in rights under the Bayh-Dole Act if products are not made suitably accessible in developing countries would allow the government to bring pressure on private companies to work with local suppliers. Finally, universities could prohibit a licensee from filing patent applications in any country in which access concerns lie.243 Other options include contributing to socially conscious patent pools or partnering with private industry in order to decrease licensee risk of failure. D.

Industry Partnerships

Of course, partnering with another institution brings benefits and problems. Academic-industry partnerships bring in funding that allows research to expand in the era of shrinking pools of federal funding. The impact of these partnerships on the university mission, control over university hiring and spending and ownership of resultant IP are more troubling. The biggest benefit for universities in academic industry partnerships is funding.244 As traditional government funding decreased, universities have used industry dollars to update facilities and increase budgets. Additionally, industry has sponsored specific endowed chairs within university departments. Research funding from industry has also increased during this time period from 2.1 billion dollars in 1998 to 3.4 billion dollars in 2007 for much the same reason. But at what cost? Some scholars have argued that academic-industry partnerships divert universities from their traditional mission because the resulting “multiversity” no longer has its focus on open public science.245 However, it is unclear if there has ever been such a unity of mission. While “pure research” institutions may have been born in 1800s Germany and imported to graduate education in the United States at Johns Hopkins University, other universities have a long histories of working with industry. The University of North Carolina and the University of Kentucky have long worked with the tobacco industry, and the University of Minnesota historically developed iron ore processing methods for the mining industry. Additionally, the Massachusetts Institute of Technology has long been a general friend to industry. But even with such partnerships, these universities continued to perform basic research. Finally, ownership and dissemination of research data and IP is a primary concern in an academic-industry partnership. It is here that the TTO can have its biggest impact. Confidentiality agreements between researchers and industry sponsors and publication control by industry sponsors threaten the free flow of information that should be an academic norm.246 TTOs should ensure that sponsored research agreements allow for publication of any research funded. While pre-publication review may be allowed, no approval right to the sponsor should

243 244 245 246

Jennifer Carter-Johnson, et al., supra note 237, at 109–10. Carter-Johnson, supra note 67, at 120–4. For full discussion, see Carter-Johnson, supra note 67, at 126–34. Carter-Johnson, supra note 67, at 133–4.

40 Research handbook on intellectual property and technology transfer be granted. Additionally, ownership of the technologies from these partnerships flows from the sponsored research agreement. Thus, to the extent that the sponsored research agreement places IP ownership with the industry partner, the academic partner is constrained in its mission to benefit the public good. Such a constraint may conflict with the public’s interest in research that tax dollars also help to fund. Therefore, sponsored research should preferentially retain IP ownership by the university, offering an option or right of first refusal to the sponsor rather than full ownership.

VI.

CONCLUSION

Technology transfer is a complicated endeavor. Although the Bayh-Dole Act authorizes universities to take ownership of patents on technology by federally-funded grants, it does not delineate a protocol for doing so or for commercializing that technology. In leaving open the method of implementation, the Bayh-Dole Act allows universities to become laboratories, testing for themselves the best ways to implement a technology transfer policy. Universities must comply with the Bayh-Dole Act requirements. In order to do so, a university must examine its research program and its core values and mission. Only then will a university be able to establish a TTO that effectively serves its researchers and reflects its mission. A large, costly TTO with entrepreneurial collaborations and open incubator laboratory space may be a wonderful investment at some research universities. At other, smaller or less entrepreneurial institutions, a TTO might be better served in focusing on faculty research collaborations such as MTAs and sponsored research agreements along with the occasional invention disclosure. Furthermore, the university and TTO should craft a set of IP Polices that inform researchers of their rights and obligations in the technology transfer process. Again, these policies should reflect the mission of the university and compliance with the Bayh-Dole Act. Disclosure rules and obligations, coverage of people and creations, and revenue sharing rules should be clearly explained. Because even the clearest explanation may lead to conflict, dispute resolution policies should be included in every IP Policy. In drafting these IP Policies, the university must consider the full spectrum of incentives that drive university researchers. In considering these incentives, universities must look beyond monetary incentives of revenue sharing to social norms of research. By aligning IP Policy incentives with researcher incentives, universities can maximize invention disclosures and researcher engagement.

3.

The politics of university technology transfer Jessica A Sebeok

I.

INTRODUCTION

There is no single way of thinking, even within individual universities, about either the proper means or the desired ends of technology transfer, the process by which universities convey their patented innovations to the private sector for development into goods and services. Universities are remarkably heterogeneous—as a result of historical, geographical, structural, and many other phenomena—and their technology transfer operations and ambitions are similarly diverse.1 American universities “converted to the cause of technology transfer at different times and with different degrees of commitment,”2 and the decisions that led most US research universities to participate in technology transfer were not the result of a cohesive or coherent process. Rather, they stemmed from a “variety of political projects undertaken by an array of political actors holding a variety of political and economic philosophies.”3 Today, technology transfer practices, as well as levels and forms of institutional support for technology transfer, differ markedly from university to university. In addition, there is no settled consensus within the variegated US university system regarding the form that universities’ relationships with industry through technology transfer should take or, for that matter, the extent to which those relationships are appropriate in the first place, particularly when federal (i.e., taxpayer) funding of research enables those relationships. One key to understanding this phenomenon is to appreciate the political processes that surround technology transfer and the way that universities have engaged in the development of law and policy relating to technology transfer. Research universities have long sought to dictate their own destinies by engaging in federal legislative and regulatory processes. However, as this Chapter discusses, because universities have adopted different positions regarding the processes and even the propriety of technology transfer, universities sometimes struggle to coalesce around policy and political positions that fairly represent “research universities” as a whole. This perspective is illuminating insofar as it demonstrates that analyses

1

In fact, it is this heterogeneity—both within and across universities—that makes the US university system unique in the world. The American university system is distinct from those of other large industrial nations by virtue of its “large scale, the high level of autonomy enjoyed by individual universities and colleges, the dependence by these institutions on local sources of financial and political support, and the strong competition among universities and colleges for funds, prestige, faculty, and students.” David C. Mowery, et al., Ivory Tower and Industrial Innovation: University-Industry Technology Transfer Before and After the Bayh-Dole Act 13 (2004) [hereinafter Ivory Tower and Industrial Innovation]; see generally David F. Labaree, A Perfect Mess: The Unlikely Ascendancy of American Higher Education (2017). 2 Roger L. Geiger & Creso M. Sá, Tapping the Riches of Science: Universities and the Promise of Economic Growth 34 (2008). 3 Elizabeth Popp Berman, Creating the Market University: How Academic Science Became an Economic Engine 16 (2012).

41

42 Research handbook on intellectual property and technology transfer of technology transfer must take into account its multifaceted nature and reject accounts that oversimplify the university community’s multi-vocal role in these debates. To that end, this Chapter explores how the complexities and uncertainties inherent in research universities’ commitment to technology transfer manifest in the political work of higher education associations in Washington, D.C. to advocate for university technology transfer writ large.4 The higher education associations referenced in this Chapter include the Association of American Universities (“AAU”), the American Council on Education (“ACE”), the Association of American Medical Colleges (“AAMC”), the Association of Public and Land-grant Universities (“APLU”), the Association of University Technology Managers (“AUTM”), and the Council on Governmental Relations (“COGR”). Of these, three—the AAU, ACE, and APLU—are known as “presidential associations,” meaning that the presidents and chancellors of their member institutions are themselves the members of those associations. At the same time, other campus representatives—such as provosts and government relations professionals—are active participants in their respective associations through a variety of constituent groups (for example, the AAU’s “Chief Academic Officers” constituent group comprises provosts). Higher education associations have a symbiotic, multilateral relationship with their diverse member institutions and the campus constituencies within those institutions. There are frequent contacts and sustained interactions between higher education association staff and representatives of those member institutions—including presidents and chancellors, provosts, senior research officers, public affairs officers, and government relations professionals—through meetings, correspondence, conference calls, and the like. Policy positions and advocacy initiatives result from this iterative process of communication and interaction, and associations seek to mirror the concerns and interests of their member institutions while guiding them to a tenable collective position in the prevailing political environment.5 The strength of associations as policy and political advocates has increasingly come not only from their convening power but also from their vantage point in Washington, D.C., which allows them to survey the bigger picture and identify emerging high-level policy matters. Nevertheless, while university associations can and do influence policymaking, judicial decisions, and politics,6 the benefits of the university-association symbiosis are not always clear and, even when clear, can be difficult to realize. First, associations represent a diversity of universities, which, as noted above, have widely diverging opinions on the practicalities

4

It is important to note here that, at times, some of the aforementioned associations cannot or do not want to engage in particular association actions, such as submitting statements for the Congressional Record. In this Chapter, I have endeavored to be clear regarding which associations have joined in a given political action, although for reasons of professional discretion, I am not always at liberty to expose why a particular association declined to participate in a particular action. 5 At times, the higher education associations “may temper the positions their members take on group policy issues.” See James D. Savage, Funding Science in America: Congress, Universities, and the Politics of the Academic Pork Barrel 48 (1999). 6 See, e.g., Brody Mullins, Douglas Belkin, & Andrea Fuller, Colleges Flex Lobbying Muscle, wall st. J., Nov. 8, 2015; Rick Cohen, Universities Pay Plenty for Influence and Access through Lobbying, nonprofIt Q., July 16, 2014; Andrew P. Kelly, Don’t Blame For-Profit Colleges for the Higher-Ed Lobbying Epidemic, atlantIC, Jan. 5, 2012; Constance Ewing Cook, Lobbying for Higher Education: How Colleges and Universities Influence Federal Policy 26 (1998).

The politics of university technology transfer 43 and politics of technology transfer.7 Second, associations are voluntary confederations,8 and as long as universities continue to meet their associations’ extant membership criteria, the associations do not impose sanctions on members that decline to adopt or actively disregard an association’s policy or political position.9 Instead, associations can only seek to influence the postures and behaviors of their member institutions through modeling and “moral” suasion. Associations’ member universities often try to implement their associations’ recommendations, whether on the campus level or at the national level. But, again, the multifarious nature of institutions’ local conditions, missions, and other variables such as financial stresses sometimes militate against individual institutions’ ability or willingness to adopt the same position as their fellow association members. Although associations must, out of practical and political necessity, try to speak in a unitary voice for university technology transfer, the reality is that this harmonized voice often results from a series of negotiations and compromises. In short, the associations’ work both reflects and affects the diverse and sometimes contradictory manifestations of university technology transfer, as well as the larger institutional and societal contexts within which university technology transfer operations reside.

II.

THE POLITICAL HISTORY AND HISTORICAL POLITICS OF UNIVERSITY TECHNOLOGY TRANSFER

University research with commercial aims or applications has preoccupied universities and their associations for over a century. In 1919, Ray Lyman Wilbur, president of Stanford, ruminated in a presentation to his fellow members of the AAU10 on what he termed “remunerative extra-university activities” conducted by university faculty on behalf of industry: [C]ommercialization of the university staff is dangerous. Of course, in the great mass of cases men do not need any administrative advice or control whatever. They have a sense of loyalty to the university

7

Even the AAU, which has a far smaller membership than the other presidential and institutional associations, “was and is highly stratified in terms of its member institutions’ research excellence and their relative financial resources,” as well as in terms of their dispersion across the geopolitical spectrum (i.e., “red” states and “blue” states). Savage, supra note 5, at 53. 8 Some, presumably, would argue that membership in the AAU is not truly “voluntary” given competition among universities to join—and the attendant calibration of universities’ priorities to the AAU’s membership criteria—and given the stature that is said to accrue to those who are admitted to membership. Savage, supra note 5, at 51. 9 The story of political “earmarking” provides a good example of how this dynamic can play out. In 1983, all but two AAU presidents voted to pass a resolution that urged both universities and members of Congress to refrain from pursuing earmarks. But when asked during a congressional hearing why the AAU would not impose sanctions on or expel the small number of AAU institutions that continued to seek earmarks despite the resolution, former Vanderbilt University President Joe B. Wyatt testified that “[T]hese are voluntary organizations. They simply cannot control the behavior of their members or impose sanctions on them.” Savage, supra note 5, at 54. 10 For an explanation of AAU’s founding and early years, see Ann Speicher, The Association of American Universities: A Century of Service to Higher Education 1900–2000, available at https://www .aau.edu/association-american-universities-century-service-higher-education-1900-2000 (last visited Jan. 25, 2019).

44 Research handbook on intellectual property and technology transfer and to their science, but they are exposed oftentimes to a pressure that is very considerable to do outside things.11

Nevertheless, President Wilbur concluded that certain “commercial undertakings” were not incompatible with the goals of universities and of science more generally, even specifying where, when, and under what conditions those commercially oriented activities would be acceptable. President Wilbur’s address was followed by a discussion among his fellow AAU presidents, during which University of Chicago President Harry Pratt Judson was emphatic that “[w]e have not thought it advisable on the whole in our laboratories to have commercial analysis made under any conditions” while nevertheless conceding that commercial research in university laboratories could be deemed acceptable as long as the results of such research “are open to all, are not confined simply to the donor of the fund and would not, therefore, lead to a patent right.”12 President Charles Homer Haskins of Harvard then offered a pragmatic coda to the conversation: “It has been assumed by President Wilbur and President Judson that the professor is sufficiently paid as a professor, and that any outside work is done for the service of the public, or the professor’s own development, or some similar non-pecuniary reason.”13 This issue resurfaced at the following year’s AAU conference in a paper entitled, “Co-operation in Research with Private Institutes,” presented by John Johnston (a Yale chemistry professor) and F. B. Jewett (chief engineer of Western Electric Company). This paper argued for increased university-industry engagement, asserting that “there is no real distinction” between “investigations of interest to the industry” and “research in ‘pure’ science.”14 Thus in the early years of the 20th century, the presidents of some of the leading research universities viewed commercialization of scholarly research as a topic worthy of concerted discussion and potential collective action. Through the 1920s and 1930s, the federal government played a modest role in financially supporting university research. Industry or philanthropic foundations, such as the Rockefeller Foundation and the Carnegie Corporation, largely funded this research.15 Universities traditionally were partial to such private contributions because they guaranteed “independence from governmental intrusion.”16 Indeed, even though relationships between scientists and the federal government date at least back to the Civil War, historians have identified a longstanding strain of “deep suspicions held by scientists and experts before World War II of any permanent alliance with the federal government” for fear that such a relationship would subjugate their academic work to “political agendas.”17 Those suspicions were largely set aside as a result of the exigencies of World War II. The attendant reconceptualizing and restructuring of university-government dynamics had a dramatic impact on the scale and substance of US academic science, on the willingness of

11 The Association of American Universities, Journal of Proceedings and Addresses of the Twenty-First Annual Conference 56 (1919). 12 Id. at 63. 13 Id. at 67. 14 The Association of American Universities, Journal of Proceedings and Addresses of the Twenty-First Annual Conference 51 (1920). 15 Geiger & Sá, supra note 2, at 9. 16 Homer A. Neal, Tobin L. Smith, & Jennifer B. McCormick, Beyond Sputnik: U.S. Science Policy in the 21st Century 94 (2008). 17 Rebecca S. Lowen, Creating the Cold War University: The Transformation of Stanford 8 (1997).

The politics of university technology transfer 45 university scientists to partner with the federal government, and on perceptions of the larger place of academic research in the American polity. Concomitantly, science was generating more military-industrial scale projects that benefitted from federal financing in a way that pre-war projects did not need or could not successfully exploit.18 In this context, the federal government gave universities generous research contracts that accelerated university research, and this newly robust university-government partnership also created a normative foundation for federal support of university scientific research. In the postwar years, federal science agencies developed their own idiosyncratic patent policies regarding agency-funded extramural research conducted at universities. Some allowed agencies to waive their title to an invention if requested by the inventor; others required agencies to retain the title to inventions developed from their funding; and still others mandated the release of the inventions into the public domain. A number of these agencies generated Institutional Patent Agreements (“IPAs”) that, in certain limited circumstances, permitted universities to own and license patents stemming from these agencies’ funding. However, the agencies issued very few IPAs and to only a handful of universities.19 In addition, these IPAs “varied from university to university, and did not pursue any consistent principles or goals.”20 By the mid-1970s, federal agencies began to narrow the circumstances and terms and conditions under which they would grant IPAs, and the process for securing even these limited IPAs was time-consuming and uncertain. As a result, universities found it difficult to establish any commercialization plans.21 During this period, the broader contours of the government-university research relationship were also changing, including a flattening of the levels of R&D funding that universities had come to rely on during World War II and in the immediate postwar years. Although the “political coalition that funded basic science was the same one that funded the cold war,” this alliance began to fray because of fundamental differences over the Vietnam War. As a result, “academic science funding fell into disarray from which it did not recover until the 1980s.”22 Furthermore, the contentious politics of the Vietnam War era strained and even fractured relations between the federal government and universities. Military agency funding of university research served as a lightning rod for the anger of those on campuses who protested the Vietnam War. These protests then roused policymakers who supported the war to threaten to curtail all federal funds to universities that permitted these protests to happen on their campuses. This confluence of events—namely, administrative challenges associated with IPAs23 and diminishing federal funding for research amid rising political discord and economic malaise—

18

See Big Science: The Growth of Large-Scale Research (Peter Galison & Bruce Hevly eds., 1992). Jacob H. Rooksby, The Branding of the American Mind: How Universities Capture, Manage, and Monetize Intellectual Property and Why It Matters 131 (2016). 20 Elizabeth Popp Berman, Why Did Universities Start Patenting? Institution-building and the Road to the Bayh-Dole Act, 38 soC. stud. asCI. 835, 845 (2008). 21 See University and Small Business Patent Procedures Act: Hearings on S. 414 Before the Senate Comm. on the Judiciary, 96th Cong. (1979) (statement of Elmer B. Staats, Comptroller Gen. of the United States). 22 Sheila Slaughter & Gary Rhoades, Academic Capitalism and the New Economy: Market, State, and Higher Education 47 (2004). 23 Peter Lee, Patents and the University, 63 duke l.J. 1, 18–19 (2013); see also David C. Mowery, et al., Ivory Tower and Industrial Innovation: University-Industry Technology Transfer Before and 19

46 Research handbook on intellectual property and technology transfer generated a sense among universities and their representatives that they needed to engage actively on patent policy in the federal sphere. The blocking of IPAs mobilized university patent administrators to undertake their first attempt at concerted political activity. To be sure, universities were not the only or even primary forces that drove the introduction in 1980 of the legislation that became the Bayh-Dole Act, which created a uniform patent policy among the many federal agencies that fund research, enabling universities, nonprofit research institutions, and small businesses to retain patent and licensing rights to inventions developed by university researchers supported by federal research grants. During these years there was “no single interest group with enough power to pass major patent policy legislation of any sort; all kinds of groups had been making such efforts for 30 years.”24 But universities’ active expressions of dissatisfaction with the new status quo served as a focal point for, and helped to give momentum to, what became the Bayh-Dole Act. According to the now-dominant historical narrative,25 Senator Birch Bayh was approached jointly by Ralph Davis, a Purdue University patent administrator who had negotiated one of the first IPAs and helped to create the Society of University Patent Administrators (“SUPA”).26 As Senator Bayh recalled, they discussed the concern of Davis and his colleagues that “whenever federal dollars went into research … and any ideas that were developed from those dollars, those grants, the patents that were secured were owned by the government and no private individual or company could get access to them. This meant that unless the private sector could get involved, there was nobody to invest the risk capital.”27 Around the same time, Barry Leshowitz, who had taken a hiatus from his professorship in psychology at the University of Arizona to serve as an intern in the office of Senator Robert Dole, convinced Senator Dole that the federal scientific funding agencies were failing to commercialize the groundbreaking university innovations to which the agencies retained title.28 Representatives of other research universities, to a large extent rallied by the newly formalized SUPA, soon began to follow Purdue’s example and to lobby their members of Congress (and their institutions’ alumni in Congress) on behalf of a sweeping overhaul of federal patent policy that would allow universities and small businesses to own and patent their federally funded inventions. Senator Bayh later characterized this activity as “good old fashioned politics … The issue made sense practically but politically it took the muscle power from the folks

After the Bayh-Dole Act 30 (2004) (“[P]olicy shifts that threatened IPAs helped motivate support for the Bayh-Dole Act by universities that already enjoyed the benefits of patenting federally funded inventions.”). 24 Berman, supra note 20, at 854. 25 This history is recounted in numerous places. See, e.g., Ann Hammersla, Patricia Harshe Weeks, & Catherine Innes, Recollections: Celebrating the History of AUTM and the Legacy of Bayh-Dole (2004); Ashley J. Stevens, The Enactment of Bayh–Dole, 29 J. teCh. transfer 93 (2004). 26 In the late 1980s, presumably to reflect the broader role of university technology transfer administrators and in recognition of the fact that patents are only one aspect of university technology transfer activities, SUPA changed its name to the Association of University Technology Managers (“AUTM”). Note that the AUTM’s membership—unlike that of the presidential and institutional associations like the AAU and COGR—comprises individual university technology transfer professionals. 27 Gene Quinn, Exclusive Interview: Senator Birch Bayh on Bayh-Dole at 30, Ip watChdog, Nov. 7, 2010. 28 Stevens, supra note 25, at 94; see also Henry Etzkowitz, MIT and the Rise of Entrepreneurial Science 119 (2002).

The politics of university technology transfer 47 back home.”29 “The way I like to look at this,” Howard Bremer reminisced in an interview in 2001, is that “finally universities were speaking with a loud single voice in this arena.”30 Importantly, it was at this time that the AAU began to reimagine and reorganize itself as a professional federal lobbying apparatus with a leading responsibility for promoting university patenting and technology transfer. The AAU did not have a formal government affairs apparatus, much less a full-time professional staff, until it established a Washington, D.C., office in 1962.31 It took the AAU another seven years to create its Council on Federal Relations (“CFR”), comprising “campus administrators … who were given the task of paying particularly close attention to federal issues.”32 The AAU did not adopt a formal federal relations agenda until 1978, but, by the time AAU promulgated its second federal relations agenda in 1979, patent policy—and, by extension, technology transfer—was highlighted as one of the federal issues for which AAU would bear primary responsibility among higher education associations (including SUPA).33 Some commentators maintain that it was the university patent administrators who, now finally organized as the SUPA, were responsible for convincing the higher education presidential and institutional associations to take an active interest in the emerging patent legislation.34 However, the AAU, from the onset of its “professionalization,” conceived of itself as the leading voice in higher education association on patent policy and legislation and acted accordingly. In other words, technology transfer had come to be seen by university presidents and chancellors as an issue of sufficient consequence to universities that it was worthy of political action by the universities themselves and not just their constituent technology transfer officers. Despite this increasing institutional interest in patent policy and legislation, the AAU was initially hesitant to offer full-throated support for the Bayh-Dole Act legislation, even while a number of its member presidents testified and provided other indicia of approval for the bill.35 The AAU’s 1979 federal relations agenda recommended, for example, that the AAU support Bayh-Dole Act “but that support should not preclude … support for other initiatives

29

Quinn, supra note 27. Interview by Barry Teicher with Howard Bremer, former patent counsel to WARF, and Hector F. DeLuca, Emeritus Professor, University of Wisconsin-Madison, Madison, Wi. (Mar. 22, 2001), available at https://minds.wisconsin.edu/handle/1793/70606 (last visited Oct. 17, 2019). 31 Before that time, the AAU’s business “had been handled by officials of member campuses, who would pass on their duties and materials to the next campus volunteering for the duty.” See The Association of American Universities: A Century of Service to Higher Education 1900–2000 (2000). 32 Id. Universities and their presidential and institutional associations did lobby the federal government well before they became professionalized. But in the “decades after World War II, especially the 1950s and 1960s, it was possible for higher education to be the ‘worst lobby in Washington’ and still fare very well.” See Cook, supra note 6, at 26. 33 Association of American Universities, AAU Federal Relations Agenda (1979). To provide additional perspective on this agenda, the other issues for which the AAU arrogated primary responsibility to itself were as follows: basic research, health education, graduate education, research library resources, and accountability in the use of federal funds. 34 Berman, supra note 20, at 856. 35 See, as one example among many, the following 1978 statement by Arthur G. Hansen (President, Purdue University): “The university, where the invention originated, is in a better position to transfer technology than the government … The increased technology transfer that can result from this legislation will lead to new products, new competition, job creation, and economic growth so essential for a strong America. Senators Bayh and Dole are to be commended for their foresight in sponsoring this legislation.” Statement by Arthur G. Hansen, President of Purdue University (Sept. 13, 1978). 30

48 Research handbook on intellectual property and technology transfer that provide title-in-contractor.”36 Accounts vary as to whether the AAU took a cautious approach because it was “unsure that it was appropriate for universities to be systematically pursuing patents”37 or because it wanted to hedge its political bets by adopting a nuanced political position. The truth of the matter likely lies somewhere in between. Ultimately, however, the AAU, its association counterparts, and the representatives of its member institutions nonetheless participated actively in the negotiations leading up to the enactment of Bayh-Dole Act.38 The associations’ political strategy around the proposed Bayh-Dole Act legislation capitalized on the specter of America losing its global technological and economic dominance, successfully promoting university technology transfer as one solution to this perceived erosion of America’s competitive edge. The fact that Bayh-Dole Act was limited to universities, other nonprofits, and small businesses was not solely or even principally because universities lobbied vigorously on behalf of the bill. Rather, Senators Bayh and Dole and their staff recognized when they introduced the bill that including big corporations as beneficiaries would result in an instant outcry—that giving those companies federal research funding would create windfalls for those companies and strengthen those companies’ monopolistic inclinations at the taxpayers’ expense.39 Yet anxieties over whether or not the policy embodied in Bayh-Dole Act represented a government give-away eventually yielded to questions of whether or not the legislation’s provisions would effectively expedite the transfer of innovations from university labs into the channels of commerce.40 This economic framing ultimately predominated in the debates over the legislation.41 In this context, universities positioned themselves to act (and be seen) as key loci of innovation for the American economy. And universities hoped that Bayh-Dole Act would provide the practical and symbolic tools to allow them to fulfill this role by replacing the system of federal agency IPAs with a uniform policy and by giving a definitive congressional imprimatur to university-industry agreements to develop federally funded innovations.

36

Association of American Universities, supra note 33. berman, supra note 3, at 150 (citing a 2005 interview with Sheldon Steinbach, former vice president and general counsel of the American Council on Education). 38 Mowery, et al., supra note 23, at 90 (“A number of universities, including Harvard University, Stanford University, the University of California, and MIT, lobbied for passage of the bill and throughout the debates were active in ‘commenting and helping to develop the final language’ of the House and Senate versions of the bill. Not surprisingly, witnesses from active institutional patenters … testified in support of the bill, as did representatives from various university associations.”). 39 Berman, supra note 20, at 857. As Berman notes, for this very reason, previous attempts at patent legislation were defeated. See also Sean M. O’Connor, The Real Issue Behind Stanford v. Roche: Faulty Conceptions of University Assignment Policies Stemming from the 1947 Biddle Report, 19 MICh. teleCoMM. & teCh. l. rev. 379, 405 (2013) (“S. 414 avoided many of the controversies plaguing other government patent policy bills by covering only nonprofit, university, and small business contractors. The retention of federally funded patent rights by large business federal contractors had been a lightning rod for commentators concerned with unjust windfalls and monopolies for such businesses.”). 40 Berman, supra note 20, at 858. US Navy Admiral Hyman Rickover, frequently called the “Father of the Nuclear Navy,” was one of the most vocal advocates of this view and thus staunchest opponents of the legislation. Rickover prodded his powerful friends in Congress, such as Senator Russell Long of Louisiana, to oppose the bill on the grounds that if taxpayers paid for research, they should own the end product. Rickover also delivered congressional testimony to this effect. See University and Small Business Patent Procedures Act, supra note 21 (statement of Hyman Rickover, Admiral, US Navy). 41 See Berman, supra note 20, at 857 (citing, as examples, the testimony in the US Senate Committee on Commerce, Science and Transportation, 1980; US Senate Committee on the Judiciary, 1979). 37

The politics of university technology transfer 49 In the years after the passage of Bayh-Dole Act, university presidents and other university representatives even more purposefully promoted what has come to be described by scholars as, variously, the “triple helix” of university-industry-government relations, “academic capitalism,” and the “market university.”42 Nevertheless, the coalition of universities, other nonprofits, and small businesses that made enactment of Bayh-Dole Act possible was not consistently unified during political negotiations over subsequent legislation involving technology transfer. For example, universities expressed vehement opposition to the Small Business Innovation Development Act (1982), which required federal agencies with annual extramural research budgets over $100 million to dedicate 1.25 percent of their budgets to research conducted by small businesses, on the grounds that this bill would divert funding away from university research. Despite this initial opposition, however, universities began to take equity stakes in start-ups created by their faculty, and these start-ups often had funding provided under the auspices of the Small Business Innovation Development Act.43 Thus, over time, universities came around to supporting the Small Business Innovation Research (“SBIR”) (and the closely-related Small Business Technology Transfer (“STTR”)) program made possible by the Small Business Innovation Development Act, and to view SBIR and STTR as crucial elements of the university technology transfer ecosystem. As Robert Berdahl, former president of the AAU, explained this evolution in universities’ thinking in his 2009 testimony before the House Committee on Science and Technology: “In the early years of SBIR many of our campuses were critical of the program, viewing it as coming at the expense of funding that would have otherwise supported university basic research. In recent years, however, as our universities and faculty have become … much more interested in commercializing new technologies, our universities have come to support SBIR and STTR Programs.”44 The next and more critical inflection point—at least from the perspective of universities’ political engagement—was the sweeping overhaul of the nation’s intellectual property (“IP”) laws embodied in the bipartisan Leahy-Smith America Invents Act (“AIA”) (2011).45 The AIA, among many other things, moved the US patent system from a “first-to-invent” to a “first-inventor-to-file” system; created an administrative mechanism housed in the US Patent and Trademark Office (“USPTO”) for the post-grant review of patents intended to improve the overall quality of the pool of patents in existence by eliminating “bad” patents; and did away with USPTO Board of Patent Appeals interference proceedings.46 Signed into law by President 42

David E. Winickoff, Private Assets, Public Mission: The Politics of Technology Transfer and the New American University, 54 JurIMetrICs J. 2 (2013) (citing the work of Henry Etzkowitz, Sheila Slaughter, Gary Rhoades, and Elizabeth Popp Berman). Universities and their presidential and institutional associations viewed patents as necessary but not, by themselves, sufficient to the success of the “triple helix” model. In 1980, for example, former MIT president Paul Gray testified to Congress “on behalf of the AAU in favor of an R&D tax credit for industry, noting that ‘as we look about us today we say that American industry faces a severe crisis brought about by rising costs, particularly with respect to energy; diminishing resources; declining productivity; and growing international competition.’” Berman, supra note 3, at 155. 43 Slaughter & Rhoades, supra note 22, at 53. 44 The Role of the SBIR and STTR Programs in Stimulating Innovation at Small High-Tech Businesses: Hearing Before the House Comm. on Sci. and Tech., 111th Cong. (2009) (statement of Robert Berdahl, President, AAU). 45 Pub. L. No. 112-29, 125 Stat. 284 (2011) (codified in scattered sections of 35 U.S.C.). 46 Before the AIA, when two separate parties filed patent applications claiming the same invention, the USPTO’s Board of Patent Appeals conducted interference proceedings to decide which of those parties was the first to invent that invention.

50 Research handbook on intellectual property and technology transfer Obama in September 2011, its main provisions went into effect in September 2012 and in March 2013.47 Negotiations over the AIA, which took over six years,48 were often heated, with a complex concatenation of political interests representing both sides of the argument.49 Universities—both at the association level and the individual university level—participated strenuously in the political process, including by engaging in behind-the-scenes negotiations on Capitol Hill and by offering testimony to Congress on issues of particular interest to universities. For example, in March 2011, John Vaughn, the AAU’s Executive Vice President, testified before the House Judiciary Subcommittee on IP, Competition, and the Internet that the AIA would go “a long way toward reforming the U.S. patent system to more effectively advance U.S. innovative capacity.”50 Vaughn praised certain aspects of the House’s proposed bill, as well as the overall legislative initiative, but he also articulated areas of concern, such as the expansion of prior-user rights and the lowered threshold to commence an inter partes review. Consistent with the competitiveness narrative embraced by universities since the years leading up to passage of Bayh-Dole Act, Vaughn emphasized that this bill would, by addressing perceived weaknesses in the US patent system, “build a robust framework for 21st century U.S. economic competitiveness.”51 After the House and Senate had both passed their versions of the AIA in May 2011, the presidents of six higher education associations wrote to their respective members to update them on the progress of the negotiations around the AIA and to encourage them to support efforts to pass the legislation.52 This letter lauded core aspects of the legislation, including the proposed adoption of the first-inventor-to-file system, and argued that the bill overall would improve patent quality and reduce patent litigation costs for universities and other patent holders. It also identified two remaining provisions of concern in the House bill but reassured association members that during the House Judiciary Committee’s April markup of the bill, the Committee “adopted major improvements to both provisions and continues to work on further refinements.”53 What is particularly revealing about this letter from a political standpoint, however, is the fact that it explicitly acknowledged differences within the higher education world that complicated the associations’ efforts to establish a unitary university position on the AIA. While asserting that there was “widespread concurrence” in the university community that the legislation was “worthy of our support,” the letter conceded that there was “not unanimous agreement” for support of the bill or for the associations’ analysis of the House version of the 47

Pub. L. No. 112-29, 125 Stat. 284 (2011) (codified in scattered sections of 35 U.S.C.). Representative Lamar Smith introduced the first iteration of the legislation in 2005. 49 For the full legislative history of the AIA, which is replete with references to universities and their positions in the negotiations, see Joe Matal, A Guide to the Legislative History of the America Invents Act: Part I of II, 21 fed. CIr. b.J. 435 (2012) [hereinafter Part I of II]; Joe Matal, A Guide to the Legislative History of the America Invents Act: Part II of II, 21 fed. CIr. b.J. 539 (2012). 50 Hearing on H.R. 1249, the “America Invents Act” Before the House Comm. on the Judiciary, 112th Cong. (2011) (statement of John Vaughn, Executive Vice President, Association of American Universities). Note that this testimony was provided by Vaughn on behalf of the AAU, as well as the American Council on Education (“ACE”), the Association of American Medical Colleges (“AAMC”), the APLU, the Council on Governmental Relations (“COGR”), and the AUTM. 51 Id. 52 AAU, et al., Memorandum to Association Constituencies on Patent Reform: Support H.R. 1249, the America Invents Act. 53 Id. 48

The politics of university technology transfer 51 bill. “Understandably,” the letter admitted, “we have had disagreements among universities throughout the six-year patent reform effort.”54 In addition, the letter communicated what could be described as political realism or political fatigue—or both—on the part of the higher education associations: “the nature of Washington is such that there comes a point when deliberation on any given issue must be brought to a conclusion in order to protect the progress that has already been made.”55 Universities ultimately achieved what they viewed as meaningful successes in the AIA negotiations, but, not long after, universities and their associations saw some of these victories as Pyrrhic. For one thing, universities did not anticipate the extent to which their own patents would be subject to post-grant validity proceedings at the USPTO or the burdens these actions would place on them.56 By June 2015, at least sixteen US universities experienced fifty-two inter partes review challenges contesting the validity of their patents.57 According to Robert Hardy, Director for Contracts and IP Management at the Council on Governmental Relations (“COGR”) and a participant in higher education association advocacy efforts to pass the AIA, “on reflection, perhaps we were naïve in not foreseeing that the new AIA IPR process might lend itself to abuses with its relative ease of petitioning for review.”58 Whether these specific challenges to university patents are abuses of the new system, the universities failed to predict that their own patents would be subject to IPR proceedings, mostly in conjunction with litigation over those same patents. Furthermore, the higher education associations and individual universities that advocated for the AIA were disappointed when they realized that the grace period under the AIA (a period of abeyance that would strike the right balance between universities’ drive to patent and university inventors’ scholarly mandate to publish) was not what they thought they had

54

Id. For a brief description of opposition to the AIA from within the higher education community, see Goldie Blumenstyk, Several Universities Oppose Pending “Patent Reform” Legislation, Chron. hIgher eduC., June 21, 2011. Per Jacob Rooksby, these institutions argued that the bill “favored large companies at the expense of higher education institutions and individual inventors, whose efforts to cost-effectively obtain, maintain, and enforce patents would, they believe, be jeopardized by the bill if it became law.” Rooksby, supra note 19, at 167–8. See also Lee, supra note 23, at 69 (“[I]n 2005 testimony regarding a previous patent reform bill, managing director of WARF Carl Gulbrandsen argued that ‘[t]he first-inventor-to-file proposal would be a hardship for a vast majority of universities.’”). 55 AAU, et al., supra note 52. 56 See Major Differences between IPR, PGR, and CBM, uspto, available at https://www.uspto .gov/sites/default/files/ip/boards/bpai/aia_trial_comparison_chart.pptx (last visited Oct. 17, 2019) (overviewing the post-AIA USPTO post-grant review, inter partes review, and covered business method proceedings). 57 Rooksby, supra note 19, at 169–70. 58 Email from Robert Hardy, Director, Contracts & IP Management, COGR, to Jessica A. Sebeok, Deputy Vice President for Federal Relations and Counsel for Policy, AAU (Aug. 16, 2018) (on file with author). As John Vaughn has explained, former Yale President “Rick Levin thought this was one of the most important goals of patent reform and the university community early argued accordingly … it may be that the current procedure is seriously flawed, but I suspect a good deal of the umbrage from the university community stems from the view, common no doubt among most patent holders, that ‘my’ patents are sacrosanct and should therefore be immune from any challenge, legitimate or otherwise.”). Email from John Vaughn, Senior Fellow, AAU, to Jessica A. Sebeok, Deputy Vice President for Federal Relations and Counsel for Policy, AAU (Aug. 9, 2018) (on file with author).

52 Research handbook on intellectual property and technology transfer negotiated.59 The associations and universities trusted that they had achieved, during the long process of negotiations that led to the AIA, a general consensus that the existing grace period for university publications should be preserved in the AIA’s first-to-file system, and that this was reflected in the revised section 102(b); various statements by the Congressional sponsors of the AIA, they believed, supported this view.60 Hence the higher education associations were dismayed when the subsequent USPTO Examination Guidelines interpreting the AIA substantially narrowed the grace period to encompass only disclosures of identical subject matter with “trivial changes or obvious variations.”61 In addition, universities found themselves the beneficiaries of an expanded “micro-entity” patent application status (with accordingly reduced filing fees) for which they had not pushed.62 The AIA had been fully implemented for a little under two years, and its effects were only beginning to take hold when Congress—with the vocal approval and encouragement of the Obama administration63—began to consider further reforms to the patent litigation system.

59

According to John Vaughn, “[t]he university community had learned how to operate successfully under the first-to-invent system (FTI) and its effective support of early publication, and UC, MIT, and several other patent-active universities opposed moving to a first-inventor-to-file system … Modifying the FTI grace period to provide in an [first-inventor-to-file] system the early publication protections it provided in the FTI system proved exceedingly difficult for two quite different reasons: 1) it proved difficult to provide the prior protections without encroaching on other legitimate transactions, and 2) it provided an opportunity for individuals and groups who always hated the grace period to launch new attacks on it.” Email from John Vaughn, supra note 58. 60 See, e.g., Part I of II, supra note 49, at 479, (“Senators Leahy and Hatch commented on § 102(b) (1)(B)’s first-to-disclose grace period, stating that it would ‘fully protect’ inventors who publicly disclose their inventions against ‘disclosures by others that are made after their disclosure.’”). 61 In October 2012, six higher education associations (AAU, ACE, AAMC, APLU, COGR, and AUTM) submitted comments in response to the Federal Register notice (77 fed. reg. 43759) (July 26, 2012) on the USPTO proposed Examination Guidelines, expressing dismay that they had made certain compromises during AIA negotiations—such as agreeing to the move from a first-to-invent (FTI) to first-inventor-to-file (FITF) system—based in part on an understanding that the grace period statutory language in the AIA would be interpreted more broadly. See Comments of the Higher Education Associations Regarding “Examination Guidelines for Implementing the First Inventor to File Provisions of the Leahy-Smith America Invents Act” (Oct. 2012), https://www.uspto.gov/sites/default/files/ patents/law/comments/hea_20121005.pdf (last visited Oct. 17, 2019). Several individual academic and research institutions, including the Massachusetts Institute of Technology (MIT), the Purdue Research Foundation, the Research Foundation for SUNY, the University of California, the University of Maryland, Vanderbilt University, the Washington State University’s Office of IP Administration/WSU Research Foundation, and the Wisconsin Alumni Research Foundation (WARF), also filed comments. See Comments on Changes To Implement the First Inventor To File Provisions of the Leahy-Smith America Invents Act, uspto, https://www.uspto.gov/patent/laws-and-regulations/comments-public/ comments-changes-implement-first-inventor-file (last visited Aug. 16, 2018). With regard to expectations vis-à-vis the grace period, Robert Hardy of COGR has opined that “[n]ot all the stakeholders involved in the AIA negotiations were friendly to the previous grace period, and we suspected that their views might have influenced USPTO.” Email from Robert Hardy, supra note 58. 62 Robert Hardy has stated that the universities and their associations had heard that USPTO “strongly believed micro-entity status should be reserved for truly small entities,” and therefore the associations were “surprised and not happy” with this addition to the AIA provisions. Email from Robert Hardy, supra note 58. In any event, this provision’s usefulness to universities has been tempered by uncertainty over the identity of the “applicant” for purposes of micro-entity certification. 63 FACT SHEET—Executive Actions: Answering the President’s Call to Strengthen Our Patent System and Foster Innovation, whIte house (Feb. 20, 2014).

The politics of university technology transfer 53 This time around, the stated legislative priority was curbing the predatory behaviors of patent assertion entities (or, more provocatively, “patent trolls”) that acquired and aggregated patents solely to profit from litigating them. Congress saw such patent assertion entities as a particular irritant to the high-tech industry.64 In November 2013, Senator Patrick Leahy introduced the Patent Transparency and Improvements Act of 2013 (S. 1720).65 In December 2013, Representative Robert Goodlatte introduced the Innovation Act (H.R. 3309).66 Higher education associations expressed approval of the Innovation Act’s goal of reducing the abusive litigation practices of patent assertion entities but opposed the bill as a whole because it included provisions on fee shifting and joinder, expansion of covered business method patents, and prescriptive statutory instructions to courts on pleading and discovery, all of which, according to the associations, “would entail greater cost than benefit by undermining the ability of universities and their licensees to enforce their patent rights.”67 The associations were more optimistic about the Patent Transparency and Improvements Act, which sought to counter patent assertion activity while omitting some of the offending provisions found in the Innovation Act. However, they were concerned that it, too, would shift the balance of interests in favor of patent defendants in a way that could compromise universities’ ability to enforce their patents and thus carry out technology transfer. Neither bill was successful during the 113th Congress (2013–14). H.R. 3309 passed the House of Representatives by a substantial 325 to 91 vote and was referred to the Senate Judiciary Committee, but, at the request of then-Senate Majority Leader Harry Reid, Senator Leahy took the patent bills off of the Judiciary Committee’s agenda, citing “university opposition in his statement killing the bill.”68 Even though neither Representative Goodlatte’s bill nor Senator Leahy’s bill passed during the 113th Congress, these bills set the stage for a number of lively and frequently antagonistic legislative battles in the 114th and 115th Congresses, political contests in which the higher edu-

64 Some contend that the real concern for the high-tech industry was not patent trolls, per se, but rather the incompatibility of the current patent system with the commercial needs and practices of that industry. See, e.g., Charles Duhigg & Steve Lohr, The Patent, Used As a Sword, n.y. tIMes, Oct. 7, 2012; Steven Levy, The Patent Problem, wIred, Nov. 13, 2012. As noted above, the patent system works in different ways for the life sciences and high-tech industries. See, e.g., Rana Foroohar, Big Tech vs Big Pharma: The Battle Over US Patent Protection, fIn. tIMes, Oct. 16, 2017. 65 S. 1720, 113th Cong. (2013). At the time, Senator Leahy was Chairman of the Senate Judiciary Committee and, in December 2013, the full Senate Judiciary Committee held a related hearing. See Protecting Small Businesses and Promoting Innovation by Limiting Patent Troll Abuse: Hearing Before the Senate Comm. on the Judiciary, 113th Cong. (2013). Nobody from the associations or any universities testified at this hearing. Id. 66 H.R. 3309, 113th Cong. (2013). 67 See Statement from the Higher Education Community on S. 1720, the “Patent Transparency and Improvements Act of 2013” (Dec. 11, 2013), available at https://www.aau.edu/key-issues/ associations-comment-senate-bill-curb-abusive-patent-litigation (last visited Oct. 17, 2019); see also University Comments on Legislative Proposals to Curb Abusive Patent Practices (Feb. 19, 2014), available at https://www.aau.edu/key-issues/university-comments-legislative-proposals-curb-abusive-patent -practices (last visited Oct. 17, 2019). 68 Joe Mullin, How the Patent Trolls Won in Congress, ars teChnICa, May 23, 2014. Many observers, however, attribute Senator Reid’s request to his alleged captivity to the interests of “trial lawyers” (i.e., plaintiffs’ lawyers) and “big pharma.” See, e.g., Timothy B. Lee, Harry Reid’s Retirement Is Good News for Patent Reform, vox, Mar. 27, 2015; Steve Lohr, With Patent Litigation Surging, Creators Turn to Washington for Help, n.y. tIMes, Apr. 29, 2015; Brian Fung, Who’s Behind the Last-minute Push to Thwart Patent Reform?, wash. post, May 21, 2014.

54 Research handbook on intellectual property and technology transfer cation associations and individual universities actively participated. Representative Goodlatte introduced a second iteration of his Innovation Act (now H.R. 9)69 early on in the 114th Congress, in February 2015. Universities anticipated this renewed effort, and, correspondingly, five higher education associations wrote to both House and Senate Judiciary Committee leaders to state that while they were willing to work with Congress to “develop targeted and measured reforms that address harmful patent enforcement practices,” they would continue to “strongly oppose legislation that would weaken the overall patent system and thereby diminish innovation and job creation in the United States.”70 Several weeks after the introduction of HR 9, 145 US universities wrote to these same congressional leaders, arguing that two of the proposals in HR 9—namely, mandatory fee-shifting and involuntary joinder—were “particularly troubling” to universities because these provisions would make legitimate patent enforcement too risky, thereby undermining university technology transfer and, by extension, American innovation and entrepreneurship.71 In quick succession after the introduction of the Innovation Act (HR 9), two additional patent litigation reform bills were introduced—the Targeting Rogue and Opaque Letters (“TROL”) Act (H.R. 2045)72 in the House of Representatives and, more importantly, the Protecting American Talent and Entrepreneurship (“PATENT”) Act (S. 1137)73 in the Senate. The leading patent reform bills in the 113th and 114th Congresses were bipartisan, but the politics of patent law tend to confound partisan orthodoxies and create “strange bedfel-

69

H.R. 9, 114th Cong. (2015). Letter from Coalition to Representative Bob Goodlatte, Chairman, House Judiciary Committee, Representative John Conyers, Ranking Member, House Judiciary Committee, Senator Patrick Leahy, Chairman, Senate Judiciary Committee, & Senator Chuck Grassley, Ranking Member, Senate Judiciary Committee (Dec. 10, 2014). This letter was written in conjunction with several organizations that represent the life sciences industry, as well as medical device manufacturers). In addition, the presidents of the Big Ten universities sent a letter in Jan. 20, 2015, to Representative Sean Duffy, in which they asserted that “a strong defensible patent is crucial to ensuring that those who want to commercialize the discoveries emerging from university research can access the investment dollars they need to move their discoveries into the marketplace,” and the presidents cautioned that the provisions of the Innovation Act/ HR 3309 would have “the effect of making patenting licensing negotiations more complex and likely discourage at least some of our members from licensing their inventions at all.” Letter from Big Ten University Presidents to Representative Sean Duffy, US House of Representatives (Jan. 20, 2015). 71 Letter from 145 US Universities to Representative Bob Goodlatte, Chairman, House Judiciary Committee, Representative John Conyers, Ranking Member, House Judiciary Committee, Senator Patrick Leahy, Chairman, Senate Judiciary Committee, & Senator Chuck Grassley, Ranking Member, Senate Judiciary Committee (Feb. 24, 2015). In March 2015, the Senate Judiciary Committee held a hearing entitled, “The Impact of Abusive Patent Litigation Practices on the American Economy,” at which Dr. Michael Crum, Vice President for Economic Development and Business Engagement at Iowa State University, testified on behalf of six higher education associations and councils (AAU, APLU, ACE, COGR, AAMC, and AUTM). The Impact of Abusive Patent Litigation Practices on the American Economy: Hearing Before the Senate Comm. on the Judiciary, 114th Cong. (2015) (statement of Dr. Michael Crum, Vice President for Economic Development and Business Engagement, Iowa State University). Notably, Dr. Crum emphasized that the purpose of university technology transfer “consistent with the overall university mission of education, research, and service—is to enable the commercial sector to generate products and processes that benefit society, not to enable the higher education sector to generate revenue. In fact, most university technology transfer operations do not receive enough royalties to offset their total operating costs.” Id. 72 H.R. 2045, 114th Cong. (2015). 73 S. 1137, 114th Cong. (2015). 70

The politics of university technology transfer 55 lows.”74 In this context, the higher education associations—which do not identify as left- or right-leaning—quietly welcomed powerful conservative opposition to the patent bills. The TROL Act took a narrow approach to combating the activities of patent assertion entities, and AAU and APLU issued a statement expressing approval of the limited scope of this bill.75 The higher education associations also saw the PATENT Act as being less systematically weighted in favor of patent defendants and as taking a more judicious approach to fee-shifting and joinder than the Innovation Act. As these various patent bills worked their way through the House and Senate, the higher education associations maintained a decided substantive and political preference for the PATENT Act over the Innovation Act.76 Yet the higher education associations were never able to reach the point of offering full support for the PATENT Act. Given differences of opinion within their respective constituencies—including certain member institutions’ fear of alienating some of the bill’s powerful congressional sponsors—they only went so far as to agree not to publicly oppose the bill. But it was “irreconcilable differences” between and among the many other stakeholders and the PATENT Act’s sponsors that resulted in an indefinite postponement of a vote in 2015.77 One significant factor in the postponement of the vote was several conservative groups’ threat to “score” any vote

74

Rooksby, supra note 19, at 174; see also Jay P. Kesan & Andres A. Gallo, The Political Economy of the Patent System, 87 n.C. l. rev. 1341, 1413 (2009) (observing that “the support for any reform to the patent system will depend on the specific factors that define the structure of each economic sector. Furthermore, we found that in each sector, there are different preferences, depending on the economic power and particular stake in the patent system.”). 75 AAU-APLU Joint Statement on the Targeting Rogue and Opaque Letters (TROL) Act (Apr. 16, 2015). 76 When the Innovation Act was marked up by the House Judiciary Committee in June 2015, the associations delivered a statement expressing their “continued serious concerns” about the bill, as well as the “process” by which the legislation developed. The process concern related to the higher education community’s sense that their perspectives were not fairly considered by the authors of the bill (“the university community was not consulted during the crafting of either H.R. 9 or the Manager’s Amendment, which has led to a bill with provisions specific to universities that don’t correspond to how universities and the technology transfer process function.”). For these reasons, the associations said they were compelled to “oppose the legislation.” Higher Education Association Statement on Scheduled House Judiciary Committee Markup of Innovation Act (H.R. 9) (June 10, 2015). The AAU and APLU simultaneously signed a joint statement by the Innovation Alliance (IA), Medical Device Manufacturers Association (MDMA), National Venture Capital Association (NVCA), and the Alliance of US Startups and Inventors for Jobs (USIJ), reiterating opposition to the Innovation Act (“Although we welcome some of the changes made in the latest version, these changes do not go nearly far enough. Overall, H.R. 9 would still dramatically weaken IP rights, harm US competitiveness and undermine a patent that has been critical to incentivizing innovation and job creation in our country for more than 200 years.”). Joint Statement from AAU, APLU, IA, MDMA, NVCA and USIJ on House Judiciary Committee Markup of H.R. 9 (June 10, 2015). The AAU and APLU again joined these groups in a June 25, 2015, statement (this time with the addition of the Biotechnology Industry Organization (BIO), the National Small Business Association (NSBA), the Pharmaceutical Research and Manufacturers of America (PhRMA), and the Small Business Technology Council (SBTC). See Joint Statement from the AAU, APLU, BIO, IA, MDMA, NSBA, NVCA, PhRMA, SBTC and USIJ on H.R. 9 (June 25, 2015) (“The bill needs significant work and should not be considered for floor action in the House of Representatives in its current form.”). 77 Pauline Pelletier & Eric Steffe, The Politics of Patent Law: Why Patent Reform Failed in 2015 and Prospects for 2016, landslIde Mag. 1 (ABA Sec. of Intell. Prop. Law) (2016).

56 Research handbook on intellectual property and technology transfer on those bills in the run-up to the 2016 election cycle based on their view that the bills would undermine constitutionally endowed property rights.78 Finally, a piece of patent reform legislation introduced in the 114th Congress proved to complicate matters relating to patent reform politics for universities and their higher education associations. In March 2015, Senators Chris Coons, Richard Durbin, and Mazie Hirono submitted their Support Technology and Research for Our Nation’s Growth (“STRONG”) Patents Act (S. 632).79 The sponsors of the STRONG Patents Act, like those of the Innovation Act and the PATENT Act, aimed to crack down on abusive patent demand letters. Unlike those other patent reform bills, however, sponsors put forth the STRONG Patents Act as a “pro-inventor” alternative to those other bills insofar as it included a “finding of Congress” that the Constitution enshrines patent property rights and that, accordingly, any incursions on those rights are deserving of the same remedies as other forms of property. Additionally, the STRONG Patents Act proposed to reform post-AIA Patent Trial and Appeal Board (“PTAB”) procedures, including by harmonizing the PTAB’s claim construction standard with that of the federal courts and by prioritizing federal court determinations of a patent’s validity. Both the AAU and APLU published statements in support of the STRONG Patents Act, describing the bill as one that would strengthen the technology transfer process by not weakening patent holders’ good faith efforts to enforce their patent rights.80 In an attempt to achieve political consensus among clamoring stakeholders, the sponsors of the Innovation Act and the PATENTS Act did agree to some cross-pollination between their bills and the STRONG Patents Act—for instance, both the Innovation Act and the PATENT Act incorporate the STRONG Patent Act’s provision on claim construction harmonization. But notwithstanding these substantive congruences, and even though these bills for the most part addressed very different aspects of the patent litigation system, the sponsors and proponents of the Innovation Act and the PATENT Act regarded support for the STRONG Patents Act as politically incompatible with support for their bills. In the end, the 114th Congress did not pass any patent litigation reform bill, and the appetite for such reform waned in the 115th Congress, having been overtaken by the federal election cycle, the possibility of congressional action on trade secrets, and various judicial developments.81 That is not to say, however, that congressional interest in patent reform has subsided entirely. In June 2017, Senators Coons, Durbin, and Hirono—with the addition of a new Republican cosponsor, Senator Tom Cotton—(re)introduced the STRONGER Patents Act (S. 1390),82 and in March 2018, Representative Steve Stivers and Representative Bill Foster intro-

78 Some political groups—conservative and liberal alike—scour politicians’ voting records and give them rankings based on how closely those politicians’ votes reflect the groups’ leanings and priorities. See, e.g., Kellan Howell, ACU to Use Patent Reform Vote in Next Congressional Scorecard, wash. tIMes, June 10, 2015; Conservative Groups, Academics Oppose Innovation Act Bill Pressure GOP to Block Legislation, wash. tIMes, Apr. 28, 2015. 79 H.R. 632, 114th Cong. (2015). 80 AAU Expresses Support for Strong Patents Act of 2015 (Mar. 2, 2015); see APLU Statement on Introduction of STRONG Patents Act of 2015 (Mar. 3, 2015). 81 Pelletier & Steffe, supra note 77. 82 S. 1390, 115th Cong. (2017).

The politics of university technology transfer 57 duced a bipartisan House companion to the STRONGER Patents Act (H.R. 5340),83 which, as of August 2018, has more than twenty cosponsors from both sides of the aisle.84 In the executive branch, too, there has been a distinct change in priorities and rhetorical tone relating to overarching patent reform since the transition from the Obama to the Trump administration.85 Again, the Obama administration was openly and vocally supportive of congressional patent reform efforts, such as the Innovation Act and PATENT Act.86 The Trump administration has not declared a distinct, much less definitive, position on patent reform; rather, it has concentrated its attention on expansive endeavors, such as the President’s Management Agenda, in which IP is but one component in service of larger economic goals.87 Since 2017, the USPTO and the National Institute of Standards and Technology (“NIST”), both of which have actively sought university input on patent and technology transfer policy, have been the center of patent-related activity in the executive branch.88 Indeed, the higher education associations welcomed the opportunity in 2018 to submit wide-ranging comments in response to NIST’s Return on Investment Initiative Request for Information regarding federal technology transfer and the public’s ability to access federally funded R&D through collaborations, licensing, and other mechanisms.89 These association comments are instructive in that they they display the full scope of the associations’—and, thus, their members’—current collective concerns and goals for university technology transfer. In brief, the associations expressed overall confidence in the technology transfer system made possible, they believe, by Bayh-Dole Act while also listing a number of specific areas in which regulatory action could improve university technology transfer. The associations’ hope is that NIST will have the wherewithal to implement positive changes that strengthen the technology transfer ecosystem while discouraging legislative action, which can present more unpredictable practical and political risks than regulatory action.

83

H.R. 5340, 115th Cong. (2018). The AAU, APLU, COGR and AUTM published a joint statement approving this second iteration of STRONG Patents; see University Associations Express Support for STRONGER Patents Act (June 22, 2017). 85 See Shearman & Sterling, Predicting Patent Policy Under the Trump Administration (Jan. 19, 2017); Shearman & Sterling, Updated Predictions on Patent Policy Under the Trump Administration (Mar. 10, 2017). 86 In addition to the White House’s own pronouncements on the issue, see David Kravets, History Will Remember Obama As the Great Slayer of Patent Trolls, wIred, Mar. 20, 2014. 87 For example, Cross-Agency Priority (CAP) Goal 14 on the President’s Management Agenda is to “improve transfer of federally-funded technologies from lab-to-market.” The Trump Administration’s FY 2019 Research & Development Budget Priorities, issued in August 2017, highlighted “innovative early-stage research.” See Memorandum for the Heads of Executive Departments and Agencies, FY 2019 Administration Research and Development Budget Priorities (Aug. 17, 2017). 88 See, e.g., Oversight of the U.S. Patent and Trademark Office: Hearing Before the Senate Comm. on the Judiciary, 115th Cong. (2018) (statement of Andrei Iancu, Director, USPTO); National Press Release, Institutes of Standards and Technology (NIST), NIST Seeks Public Input to Help Increase Return on Investment from Federal Research (May 1, 2018); and Bulletin, American Institute of Physics, Sweeping Review of Technology Transfer Laws Among Top Priorities of New NIST Director (Jan. 4, 2018). 89 See Higher Education Associations Submit NIST RFI Response, AAU, available at https://www .aau.edu/sites/default/files/AAU-Files/Key-Issues/Intellectual-Property/Higher-Ed-Associations-RFI -Response-Federal-Technology-Transfer-Authorities-and-Processes.pdf (last visited Jan. 25, 2019). 84

58 Research handbook on intellectual property and technology transfer

III.

DRUG PRICING: A CASE STUDY IN THE POLITICS OF UNIVERSITY TECHNOLOGY TRANSFER

Universities have played a major role in the development of drugs and medical technologies, and they have steadily become an even more important source of potential medications and medical technologies for pharmaceutical companies.90 Over the years, many have charged that the cumulative effect of this conduit from universities to life sciences companies is that taxpayers pay twice—often at substantial prices downstream—for medicines developed out of research that taxpayers paid for in the first place.91 In this particular context, universities often find themselves caught between competing concerns: on the one hand, a commitment pursuant to Bayh-Dole Act to patent and license their federally funded medical innovations to the private sector and, on the other hand, a commitment to provide affordable healthcare to patients at university hospitals and medical centers. One repeatedly proposed approach to the problem of high drug pricing domestically and abroad is a more expansive use of Bayh-Dole Act “march-in” rights by the federal agencies that fund university research.92 March-in rights under Bayh-Dole Act give the government, in limited circumstances, the right to require the owner of a patented invention developed with federal funding to grant additional licenses to that invention.93 Some proponents of the

90

Vladimir Drozdoff & Daryl Fairbairn, Licensing Biotech Intellectual Property in University-industry Partnerships, 5 Cold sprIng harb. perspeCt. Med. (2015); see Amy Kapczynski et al., Addressing Global Health Inequities: An Open Licensing Approach for University Innovations, 20 berkeley teCh. l. J. 1031, 1042 (2009) (explaining that over the past two decades, “early ‘upstream’ inventions that explain disease pathways and mechanisms and identify potential drug targets are increasingly likely to be patented.”). 91 Those who promote the idea that taxpayers are being charged twice for pharmaceuticals tend to underplay the cost of drug development, especially relative to the federal government’s typically far more modest initial investment in the underlying research. The average cost per approved drug, as reported by the Pharmaceutical Research and Manufacturers of America (PhRMA) industry group, is disputed, but there is little disagreement that “[b]ringing a new drug to market is expensive.” Lisa Larrimore Ouellette, How Many Patents Does It Take to Make a Drug – Follow-On Pharmaceutical Patents and University Licensing, 17 MICh. teleCoMM. & teCh. l. rev. 299, 302 (2010); A 2016 study under the auspices of Tufts University Center for the Study of Drug Development estimated the cost of R&D for new drugs and biologics to be $2870 million (2013 dollars). J.A. DiMasi, et al., Innovation in the Pharmaceutical Industry: New Estimates of R&D Costs, 47 J. health eCon. 20, 20 (2016); see also Sabarni K. Chatterjee & Mark L. Rohrbaugh, NIH Inventions Translate Into Drugs and Biologics With High Public Health Impact, 32 nature bIoteCh. 52 (2014) (estimating that pharmaceutical companies spend $100 to develop drugs for every $1 the government spent in research leading to medical inventions). In short, although an R01 grant or two might lead to the initial conception of an invention, federal grant funding usually constitutes a small fraction of the $2 billion or more of investment necessary to bring a tested and safe drug or medical technology to patients. 92 See, e.g., Michael Henry Davis & Peter S. Arno, Why Don’t We Enforce Existing Drug Price Controls? The Unrecognized and Unenforced Reasonable Pricing Requirements Imposed Upon Patents Deriving in Whole or in Part from Federally-Funded Research, 75 tul. l. rev. 631 (2001). Another proposed approach is employing 28 U.S.C. § 1498, which gives the US government the “right to use patented inventions without permission, while paying the patent holder ‘reasonable and entire compensation.’” 93 35 U.S.C. § 203; see generally John R. Thomas, March-In Rights Under the Bayh-Dole Act (Crs report No. R44597) (Aug. 2016); Wendy H. Schacht, Federal R&D, Drug Discovery, and Pricing: Insights from the NIH-University-Industry Relationship (CRS Report RL32324) (2006).

The politics of university technology transfer 59 increased use of march-in rights assert that such rights are an obvious mechanism to control and lower the cost of pharmaceuticals (and thus protect the interest of US taxpayers).94 To date, the NIH has received at least six petitions from private entities asking it to use its march-in rights with respect to particular drugs, but the NIH has denied each one.95 In response to every petition filed on the grounds that a given drug was priced at an unaffordably high level, the NIH reiterated its position, which is based on its reading of the statutory language of Bayh-Dole Act—namely, that “concerns over drug pricing were not, by themselves” enough to trigger the application of march-in rights.96 Several members of Congress, such as Representative Lloyd Doggett and Senator Bernie Sanders, have urged the NIH to adopt a more expansive view of appropriate circumstances for exercising march-in rights; however, the NIH has pushed back, stating that it considers petitions for application of the march-in statute on a case-by-case basis and “is prepared to use its authority if presented with a case where the statutory criteria are met regarding the commercialization of an NIH-funded, patented invention, and where march-in could in fact alleviate health or safety needs or address a situation where effective steps are not being taken to achieve practical application of the inventions.”97 The oft-stated position of the NIH on march-in rights, however, has not dampened enthusiasm for expanding march-in rights to combat high drug prices. In 2017, for example, the report of the Senate Armed Services Committee on the National Defense Authorization Act of 2018 (S. 1519)98 included a requirement, originally proposed by Senator Angus King, that the Department of Defense employ its rights under sections 209(d)(1) or 203 of title 35, US Code, to “authorize third parties to use inventions that benefited from Department of Defense funding whenever the price of a drug, vaccine, or other medical technology is higher in the United States than the median price charged in the seven largest economies that have a per capita income at least half the per capita income of the United States.”99 The higher education association unsuccessfully lobbied to have this language altered, removed, or negated in the

94

See Thomas, supra note 93; Amy R. Schofield, Currents in Contemporary Ethics: The Demise of Bayh-Dole Protections Against the Pharmaceutical Industry’s Abuses of Government-Funded Inventions, 32 J. law, Med. & ethICs 777 (2007). 95 Most but not all of these march-in petitions concerned drug pricing. For example, the petition regarding Fabrazyme, the first drug for the treatment of the deadly Fabry disease, asked the NIH to march in on the grounds that Fabrazyme was not being produced in sufficient quantities to meet patient need. 96 Thomas, supra note 93, at 9. 97 Letter from Sylvia M. Burwell, Secretary of the Department of Health and Human Services, to Representative Lloyd Doggett, US House of Representatives (Mar. 2, 2016), at 2; see also Ed Silverman, Obama Administration Rejects Congressional Push to Override Drug Patents, stat news, Mar. 8, 2016. 98 S. 1519, 115th Cong. (2017). 99 sen. rep. no. 115-125 (2017). Senator Thom Tillis, Chairman of the Senate Armed Services Personnel Committee, and Senator Jim Inhofe, Chairman of the Senate Armed Services Subcommittee on Readiness and Management Support, objected to inclusion of this language in the committee report (not, it should be said, at the behest of universities, although the higher education associations did communicate concerns to the Senate Armed Services Committee). Senators Tillis and Inhofe protested on the grounds that, in the words of Senator Tillis, “[t]he idea of regulating the price of a commercialized invention was never contemplated by Congress when passing the Bayh-Dole Act”; thus such a mandate “could chill medical innovation by raising the risk of a Federal partnership that is unacceptable for many private entities” (Statements of Senators Tillis & Inhofe on file with author.) See also Jennifer Plitsch & Alexander Hastings, Senate Committee Directs DoD to Reduce Drug Prices, InsIde gov’t Cont. (Covington & Burling LLP), July 26, 2017.

60 Research handbook on intellectual property and technology transfer subsequent year’s NDAA congressional report and thus have to rely on the fact that congressional report language, unlike statutory language, is not an authoritative directive; in any event, the position of the Department of Defense is that it has no existing capacity to manage such a process.100 Higher education associations, as well as some scholarly commentators,101 share the NIH’s view that employing march-in rights to lower drug prices would require an interpretation that is not supported by the plain language of the Bayh-Dole Act statute. In this regard, Ted Pohler of Johns Hopkins University said on behalf of his own institution and the AAU at a May 2004 NIH public hearing on the Norvir march-in petition: “To be sure, there are serious challenges surrounding the accessibility and affordability of pharmaceuticals,” but “to use march-in rights to address drug pricing is a misapplication of the statute.”102 By this analysis, which is supported by the legislative history of Bayh-Dole Act and later statements by the Act’s authors, Congress approved the march-in provision to counteract concerns that market-dominant companies would license competitive university inventions solely to prevent their development and entry into the marketplace.103 Universities and their associations thus were heartened when, in 2018, NIST released its return on investment draft green paper, which cites the associations’ comments on 2018 NIST’s request for information, calling for regulatory action to prevent the use of march-in rights to control the pricing of goods and services.104 Universities and their associations expect the debates over drug pricing—including the debatable necessity of using patents as a lever to control drug pricing—to gain momentum in the 116th Congress, particularly given the shift to Democratic control of the House of Representatives in the wake of the 2018 midterm elections. But these debates present additional political challenges for universities and their associations. Two pieces of legislation— Representative Doggett’s Medicare Negotiation and Competitive Licensing Act105 and Senator

100

Email from Julia A. Smith, former Assistant Vice President of Federal Relations at AAU, to the author relating Smith’s conversation with Senate Armed Services Committee staff (May 10, 2018) (email on file with author). 101 See, e.g., Carolyn L. Treasure et al., Do March-In Rights Ensure Access to Medical Products Arising from Federally Funded Research? A Qualitative Study, 93 MIllbank Q. 761, 762 (2015) (concluding that “as currently specified in the statute … march-in rights are unlikely to serve as a counterweight to power the prices of medical products arising from federally funded research.”). Beyond the limitations created by the statutory language itself, the fact that there is “not usually a one-to-one correspondence” between a given drug and patent is another reason why march-in rights likely would be of limited utility in controlling drug pricing. In circumstances where “some but not all of a drug’s patents were developed under a funding agreement with the federal government, a license to practice a subset of a drug’s patents is not likely to be the same as a license to the drug itself.” Rachel Sachs, March-In Rights Alone Won’t Solve Our Drug Pricing Problems (Jan. 12, 2016), http://blog.petrieflom.law.harvard.edu/ 2016/01/12/march-in-rights-alone-wont-solve-our-drug-pricing-problems/ (last visited Oct. 17, 2019). 102 Id. 103 See National Institutes of Health Public Meeting on Norvir/Ritonavir March-in Request (May 25, 2004) (statement of Senator Birch Bayh); Joseph Allen, NIH Director Collins Stands Up to the March in Mob, Ip watChdog, June 27, 2016; Joseph Allen, When Government Tried March In Rights To Control Health Care Costs, Ip watChdog, May 2, 2016; Joseph Allen, NIH Pressured to Misuse Bayh-Dole to Control Drug Prices, Ip watChdog, Mar. 30, 2016. 104 See Unleashing American Innovation, NIST, https://www.nist.gov/tpo/return-investment-roi -initiative (last visited Jan. 25, 2019). 105 H.R. 6505, 115th Cong. (2018).

The politics of university technology transfer 61 Sanders’ Prescription Drug Price Relief Act106—aim to tackle the problem of high prescription drug prices not by invoking more expansive applications of Bayh-Dole Act march-in rights but by broadly threatening compulsory licensing in all instances of drug manufacturers being unwilling to negotiate “reasonable” prices for their pharmaceuticals (i.e., not just when the patent arises from federally funded university research). While universities and their associations remain concerned that using the IP of a drug as leverage to control drug pricing could frustrate or stifle universities’ ability to move their patented research discoveries from university labs into use by medical practitioners, universities are also acutely aware of the need to hold down drug costs for patients treated in their own campus medical centers and for the faculty, staff, and students who are on university-funded insurance benefit plans. In other words, universities and their associations must square upstream concerns about the viability of their technology transfer operations with downstream concerns about the welfare of the patients they treat on their campuses. Moreover, many medical researchers are themselves physicians who, for professional ethics reasons, favor broader access to therapies. Accordingly, this is a prime example of how university heterogeneity—in this case, a balancing of internal priorities—complicates political activity relating to university technology transfer.

IV.

THE FUTURE POLITICS AND POLITICAL FUTURE OF TECHNOLOGY TRANSFER

In discussions of the minutiae of university technology transfer debates, including arguments over whether or not technology transfer comports with universities’ broader public mission, it is easy to lose sight of the larger university context in which technology transfer takes place. One key fact to remember is that most universities conduct their technology transfer operations amid worsening financial strains (exacerbated for state universities by an overall steady decline in state support). In what Rooksby has accurately described as the “new normal” of higher education funding, universities are obliged to identify supplementary financial sources not just for technology transfer but for all of their operations.107 Indeed, in December 2017 Moody’s downgraded its higher education sector financial outlook from stable to negative.108 Universities are hesitant to divert funds from student tuition to pay for their research projects. And research grants and patents, “which tend to fluctuate unnervingly over time,” cannot themselves “compensate for high salaries, low teaching loads, and high graduate-student support costs.”109

106

S. 115th Cong., https://bit.ly/2TRetzu (last visited Jan. 25, 2018). Jacob Rooksby, Innovation and Litigation: Tensions Between Universities and Patents and How to Fix Them, 15 yale J.l. & teCh. 312, 354 (2013). 108 Moody’s Investors Service Outlook (Dec. 2017) (“The annual change in aggregate operating revenue for four-year colleges and universities will soften to about 3.5% and will not keep pace with expense growth, which we expect to be almost 4.0%. A growing number of universities will have even weaker revenue growth, pressuring operating performance. Public universities will face more revenue strain than private universities. The negative outlook also incorporates uncertainty at the federal level over potential policy changes.”). 109 Labaree, supra note 1, at 267. 107

62 Research handbook on intellectual property and technology transfer Bayh-Dole Act, unfortunately, was and is an unfunded mandate. Senators Bayh and Dole believed that the expense of doing technology transfer would be folded into funds that are included in federal research grants for administrative needs. But the senators’ expectations were frustrated in the early 1990s when such administrative funds were capped, resulting in technology transfer becoming a net cost for most universities. Most universities spend a miniscule fraction of their overall research budgets on the transfer of that research and do not realize meaningful revenues from their technology transfer functions.110 One study showed that “more than half of university technology transfer programs bring in less money than the costs of their operations, while only 16 percent generate enough funds to fully cover their operating costs after distribution of revenues to their faculty inventors.”111 Any residual revenues are plowed back into more research. One interpretation of this imbalanced cost-revenue dynamic is that it “simply verifies the institutional mission of the research enterprise: get science into the public’s hands.”112 By this logic, if university technology transfer were truly only about revenue generation, universities would have stopped engaging in technology transfer decades ago.113 Another structural and fiscal challenge associated with technology transfer is universities’ need to find and secure new sources of funding to bridge the “valleys of death” that lie between early-stage university innovations requiring further research and the later stages of investing and implementing fully fledged products in the marketplace.114 The term “valley of death” entered the university commercialization lexicon after Congressman Vernon Ehlers, the first research physicist to be elected to Congress, used the phrase in his 1998 report, “Unlocking Our Future: Toward a New National Science Policy” for the House Committee on Science.115 Scholars, policy experts, and technology transfer practitioners ascribe the valley of death problem to various causes, and a variety of solutions to it have been proposed over the years. The associations themselves referenced the valley of death in their 2018 comments in response to the NIST’s Return on Investment Initiative Request for Information, recommending several ways—such as funding robust “proof of concept” programs for university research with com-

110

See, e.g., Richard Pérez-Peña, Patenting Their Discoveries Does Not Pay Off for Most Universities, a Study Says, n.y. tIMes, November 20, 2013. 111 James K. Woodell & Tobin L. Smith, Technology Transfer for All the Right Reasons, 18 teCh. & InnovatIon 295, 297 (2017). In Jacob Rooksby’s words, “[t]o hit a home run requires quite a few at-bats,” referencing a 2013 study showing that “as many as 84 percent of [technology transfer offices] actually lose money each year.” Rooksby, supra note 19, at 141 (citing Walter Valdivia, University Start-Ups: Critical for Improving Technology Transfer 9 (2013)). 112 See also Vicki Loise & Ashley J. Stevens, The Bayh-Dole Act Turns 30, 2 sCI. translatIonal Med. 1, 4 (2010). 113 See Irene Abrams, Grace Leung, & Ashley J. Stevens, How Are U.S. Academic Licensing Offices Tasked and Motivated, 17 res. MgMt. rev. 1 (2009) (finding that technology transfer offices are driven more by assisting faculty and translating the results of research than maximizing financial gain). 114 Loise & Stevens, supra note 112, at 4. The “valley of death” is the “metaphor often used to describe the gap between academic-based innovations”—which are usually early-stage and high-risk and have uncertain prospects—“and their commercial application in the marketplace.” Karen Elizabeth Gulbrandsen, Bridging the Valley of Death: The Rhetoric of Technology Transfer (2003) (unpublished PhD dissertation, Iowa State University). 115 Unlocking Our Future: Toward a New National Science Policy, U.S. House of Representatives Comm. on Science, 105th Cong. (Sept. 1998).

The politics of university technology transfer 63 mercial potential—in which the federal government could help university innovations traverse the valley of death.116 In the context of these micro and macro funding challenges, universities are appraised negatively by many for their supposedly dominant focus on capturing revenues from technology transfer.117 These criticisms come not just from scholars, journalists, and the general public, but also from the very industries to whom universities license and with whom universities partner. Some in industry allege that universities do not license their IP in a manner that reflects the goals of Bayh-Dole Act and lament that universities try “too hard to extract revenue from IP rather than promoting … the motivating purpose of the Bayh-Dole bargain.”118 One recent example illustrates industry’s frustration with university technology transfer practices: In the early 2000s, several lucrative licensed Columbia University patents were on the cusp of expiration, so Columbia secured a new patent based on claims similar to those used in the expiring patents and then insisted that licensees of the precursor patent pay additional royalties until 2019.119 This maneuver did not sit well with those companies, some of which sued Columbia for “double-patenting” and questioned whether the university was sacrificing good faith to profit-seeking. But Columbia’s defenders have argued that the university was “unfairly maligned for doing what all patent holders are legally entitled to do,” emphasizing that universities are held to a double standard—“one that allows corporations to be tough but demands that universities be wimps.”120 What is perhaps not sufficiently recognized is the fact that many universities are keenly aware of the reputational costs and benefits of their technology transfer activities. “For any given transaction with industry,” one can argue that “universities stand to reap limited benefits from the terms of the agreement, but risk substantial losses in reputation and managerial burden if accused of an ethical breach.”121 Certainly, those who criticize university-industry relationships “generally fail to appreciate how heavily the balance scales are weighted in favor of academic values.”122 Some evidence suggests that university researchers are largely focused on the extent to which patenting activity can serve as a marker of and thus advance their scholarly reputations as scientific innovators.123 Additional studies indicate that the “financial incentive from patent royalties plays a small role in academic scientists’ motivations” relative to their departments’ view

116 See Higher Education Associations Submit NIST RFI Response, AAU, at 16–17, https://www .aau.edu/sites/default/files/AAU-Files/Key-Issues/Intellectual-Property/Higher-Ed-Associations-RFI -Response-Federal-Technology-Transfer-Authorities-and-Processes.pdf (last visited Jan. 25, 2019). 117 For instance, “[f]antasies about potential riches may have led to a surfeit of patents at many institutions. One estimate puts the percentage of unused (unlicensed) patents held by TTOs at 95 percent of their cumulative portfolios.” Mowery, et al., supra note 23, at 146. 118 Winickoff, supra note 42, at 28; see also Walter Valdivia, University Start-Ups: Critical for Improving Technology Transfer 10 (2013) (referencing “industry’s regular complaint about universities being too aggressive negotiating patent license” and “certain practices that from time to time have exposed TTO as profit maximizing entities—for instance, contracts with ‘creative’ clauses such reach-through that ignited the OncoMouse controversy.”). 119 Bernard Wysocki, Jr., Columbia’s Pursuit of Patent Riches Angers Companies, wall st. J., Dec. 21, 2004. 120 Id. 121 Geiger & Sá, supra note 2, at 18. 122 Id. 123 Lisa Larrimore Ouellette & Ian Ayres, A Market Test for Bayh-Dole Patents, 102 Cornell l. rev. 271, 284 (2017).

64 Research handbook on intellectual property and technology transfer of entrepreneurial work (some, but not all, universities include patent success in tenure and promotion criteria).124 In 2007, in an “act of public accountability”125 with ramifications for political advocacy in support of university technology transfer, ten leading research universities together with the AAMC issued a white paper titled In the Public Interest: Nine Points to Consider in Licensing (the “Nine Points”).126 The Nine Points articulated a set of core principles and delineated shared guidelines designed to advance the public interest when universities license their patented innovations to private entities, and it has been endorsed by the major higher education associations and over 100 research universities and other nonprofit organizations.127 The Nine Points also emphasized that universities should always recall that they are bound by their public service mission to use patents to develop technologies that benefit society. A few years after the release of the Nine Points document, the National Research Council (“NRC”) of the National Academies of Sciences, Engineering, & Medicine published a thorough report, Managing University Intellectual Property in the Public Interest (the “NRC Report”), that analyzed and assessed university IP management practices.128 The NRC Report praised Bayh-Dole Act and described the successes of university technology transfer in the wake of the Act. The NRC Report also endorsed several of the principles set out in the Nine Points, urged universities to be transparent and visible about their commitment to promoting and supporting the public interest through IP management, and advised universities to take pains to ensure that their policies and practices align with this public interest. The AAU and APLU subsequently—and in the midst of congressional consideration of the patent reform legislation discussed above—each created high-level working groups, chaired by member presidents, on technology transfer and IP.129 These association working groups and

124

Id. at 285 (citing Alice Lam, What Motivates Academic Scientists to Engage in Research Commercialization: “Gold,” “Ribbon,” or ‘Puzzle”?, 40 res. pol’y 1354, 1366 (2011)). 125 Winickoff, supra note 42, at 30. 126 In the Public Interest: Nine Points to Consider in Licensing University Technology (Mar. 2017). 127 See id. In addition, the AUTM has issued both a Statement of Principles and Strategies for the Equitable Dissemination of Medical Technologies and has adopted guidelines on licensing transgenic mice and other research tools. See AUTM Statement of Principles and Strategies for the Equitable Dissemination of Medical Technologies, AUTM, available at http://www.iu.edu/~ufc/docs/addDocs/ AY12/AUTMStatement.pdf (last visited Mar. 21, 2019); Michael B. Dilling & Terese L. Rakow, Licensing Transgenic Mice and Other Research Tools: A Practical Guide, autM teCh. transfer praCtICe Manual (2010). These both constitute “an emotionally charged area of medical practice.” Loise & Stevens, supra note 112, at 4. The AUTM also provides its members with a “Global Health Toolkit” that includes examples of contractual clauses that are taken from successfully executed agreements at US universities and that are related to global health issues. AUTM Global Health Toolkit, AUTM, https://www.autm.net/AUTMMain/media/Advocacy/Documents/AUTMGHClauseToolkit3-17-12.pdf (last visited Mar. 21, 2019). The AUTM’s Better World project, which collects and collates examples of university technology transfer that have in some way influenced or improved society, is another example of the AUTM’s efforts to reorient views of technology transfer. The Better World Project, AUTM, http:// www.betterworldproject.org/ (last visited Mar. 20, 2019). 128 National Academies of Sciences, Engineering, & Medicine, Managing University Intellectual Property in the Public Interest (2011). 129 See AAU Technology Transfer Working Group Statement on Managing University Technology Transfer in the Public Interest (Mar. 2015); APLU Task Force on Managing University Intellectual Property Statement to APLU Members of Recommendations on Managing University Intellectual Property (Mar. 2015).

The politics of university technology transfer 65 their attendant recommendations served interlocking functions—namely, to influence campus policies in ways that both functionally and “optically” redound to their member universities’ benefit in the political arena. Following extensive review, and inspired and guided by the Nine Points and the NRC Report, both of these working groups reaffirmed that the chief goal of university technology transfer should be to serve the public interest. The AAU and APLU encouraged university leaders to direct their technology transfer operations towards promoting innovation and economic prosperity both locally and nationally. Recommendations of the AAU and APLU working groups included the following: (1) design and implement a clear mission statement for university management of IP in the express interest of the public good; (2) implement restrictions for university engagement with patent assertion entities (“patent trolls”) that acquire IP rights with no real intention of commercializing the technologies and instead rely only on threats of infringement litigation to generate revenue; and (3) broaden metrics and develop new evaluation criteria for university technology transfer offices to include measures that consider economic and societal impact and not simply revenue generation or patent counts. In a 2010 memo to the Office of Science and Technology Policy (“OSTP”), for example, five higher education associations articulated their concerns that there tends to be an inordinate emphasis on quantitative measures, such as the number of patents, licenses, and revenue.130 “Indeed,” the memo noted, “the statistics on university licensing revenues contained in the annual AUTM Licensing Activity Survey have too often been used as metrics by the media and others … to determine the ‘success’ of university technology transfer and commercialization efforts.”131 The memo further asserted that universities and others must reach agreement on a more capacious set of metrics that “accurately and appropriately reflect the range of university contributions to local, regional, and national economies.”132 The associations reiterated these points in their response to NIST’s 2018 Request for Information regarding its Return on Investment Initiative, signaling both that universities themselves take an expansive view of the impact value of technology transfer activities, and that those activities should not be judged by narrow criteria, particularly criteria imposed by government agencies or Congress.133 Exactly what broader metrics should be designated and instituted, however, is the subject of continuing examination and debate inside and outside academia.134 There is not even

130

Higher Education Association Memorandum regarding Commercialization of University Research, AAU, May 10, 2010, at 7. 131 Id. 132 Id. 133 See Higher Education Associations Submit NIST RFI Response, available at https://www .aau.edu/sites/default/files/AAU-Files/Key-Issues/Intellectual-Property/Higher-Ed-Associations-RFI -Response-Federal-Technology-Transfer-Authorities-and-Processes.pdf (last visited Feb. 1, 2019) (contending that metrics that look primarily at impact rather than income are hard to measure but that such alternative, broader measures are necessary because university returns on federal research investments “take many other forms than direct technology commercialization.”). 134 See the November 2017 APLU Report of the Technology Transfer Evolution Working Group of APLU’s Commission on Innovation, Competitiveness, & Economic Prosperity for an extended discussion of how to redefine “expectations by university leaders and governing boards about the purposes and success indicators for university engagement in innovation and technology transfer.” Technology Transfer Evolution: Driving Economic Prosperity, Report of the Technology Transfer Evolution Working Group of APLU’s Commission on Innovation, Competitiveness & Economic Prosperity (“CICEP”), APLU (Nov. 2017).

66 Research handbook on intellectual property and technology transfer a consensus on “just what ‘public’ means and to whom.”135 While an increasing number of scholars conceptualize innovation “as an outgrowth of extensive knowledge networks, social networks, and intersecting communities in an ecosystem that both generates and applies new knowledge,”136 it is not clear how to measure these attributes of university technology transfer. It is difficult to quantify returns on federal research investments and patenting activity, such as higher-than-average local wages and lower-than-average regional unemployment;137 the education and training of students who continue in academia or go on to work in industry; faculty consulting; and the publication of research results upon which industries have based new patents and products. At least in the life sciences, however, one recurring recommendation is that the return on investment for research be gauged “in terms of extended human life expectancy, reduced burden on disease and long-term economic impact.”138 For example, there is some evidence that drugs discovered by public-sector research institutions like universities “have a disproportionately large therapeutic effect,” suggesting that public-sector research positively impacts public health even more than was previously appreciated.139 However, “the latency of these outcomes makes them challenging to use as criteria for guiding current investments in research.”140 Thus, it is important that any metrics “be used carefully so as not to ‘punish’ those in the budget process that mainly investigate important clinical questions that do not lend themselves to private sector development.”141 Moreover, as the National Academies of Sciences, Engineering, and Medicine underscored in its 2014 Furthering America’s Research Enterprise report, metrics should be calibrated to individual universities’ local conditions and ambitions. The federal government should not establish one set of criteria for gauging the productivity of technology transfer at all US universities.142 Revising and implementing new technology transfer and IP policies and practices to mirror and also influence evolving norms and behaviors is not a simple undertaking at most universities, particularly inasmuch as these policies and practices usually are drafted by committees based on “numerous compromises.”143 This challenge is compounded by the aforementioned heterogeneity of universities—there is no one-size-fits-all solution even among the leading research institutions.144 Numerous factors play a part in shaping a university’s technology 135

Geiger & Sá, supra note 2, at 17. Ekaterina Galkina Cleary, et al., Contribution of NIH Funding to New Drug Approvals 2010–2016, proC. nat’l aCad. sCI. unIted states aMerICa 1 (2018). 137 Some scholars have observed that “[p]atent activity correlates positively with regional economic health, as high rates of patent creation are geographically associated with higher than average wages, lower regional unemployment and more startup company activity.” Michael J. Kalutkiewicz & Richard L. Ehman, Patents As Proxies: NIH Hubs of Innovation, 32 nature bIoteCh. 536, 536 (2014). 138 Kalutkiewicz & Ehman, supra note 137, at 537. 139 Ashley Stevens, et al., The Role of Public-Sector Research in the Discovery of Drugs and Vaccines, 364 new eng. J. Med. 535 (2011). 140 Kalutkiewicz & Ehman, supra note 137, at 537. 141 Id. 142 National Academies of Sciences, Engineering, & Medicine, supra note 128, at 45. 143 Robert C. Miller & Bernard J. Le Boeuf, Developing University-Industry Relations 47 (2009). 144 Nor should we lose sight here of heterogeneity in the larger patent system, with respect to how patents function in different industries and how patents influence research investments in those industries and, by extension, how those industries set and promote their patent policy and political priorities. See Mark A. Lemley & Dan L. Burk, Policy Levers in Patent Law, 89 va. l. rev. 1575, 1577 (2003) (observing that technology “displays highly diverse characteristics across different sectors” and that the patent system functions differently in these industries); Dan L. Burk & Mark A. Lemley, Is Patent Law 136

The politics of university technology transfer 67 transfer culture, practices, and priorities: a university’s geography (including whether it resides in a “red” or “blue” state);145 its place in the Carnegie Classification of Institutions of Higher Education;146 its historical and current success in obtaining federal research grants and patents;147 the number and expertise of its technology transfer personnel;148 and its IP policies.149 All that said, individual universities add to the innovation ecology in their own distinctly valuable ways. Thus, applying an “evaluation framework that does not take adequate account of the diverse channels of university influence or the differences among universities” would in fact “diminish the heterogeneity that historically has been a strength of the U.S. system of higher education.”150

V.

CONCLUSION

As noted above, since the beginning of the 20th century, individual universities and their associations have worked to influence the congressional and other federal political processes that impact the laws and regulations that govern institutional university technology transfer practices. Institutional policies, and how those policies are instantiated in practice, simultaneously shape thinking in Washington, D.C., about how laws and regulations should superintend uni-

Technology-Specific, 17 berkeley teCh. l.J. 1155 (2002) (concluding that what appears to be a unitary patent system in fact varies by industry); Heidi L. Williams, How do patents affect research investments? (Jan. 17, 2017); Kesan & Gallo, supra note 74. 145 See Waverly Ding, Amol M. Joshi, & Arvids A. Ziedonis, University Technology Transfer, kauffMan fndn. “‘Star’ university scientists often play a critical role in the geographical location decisions of startups in biotechnology. Moreover, academic scientists’ pursuit of commercial activities (academic entrepreneurship) is often a locally clustered phenomenon, with the social influence through collaboration and collegial ties playing a strong role in the transition process. Geographic proximity also affects the operation of market and non-market channels of university technology transfer” (internal citations removed). See also Walter W. Powell, Jason Owen-Smith, & Jeannette A. Colyvas, Innovation and Emulation: Lessons From American Universities in Selling Private Rights to Public Knowledge, 45 MInerva 121, 139 (2007) (“The experience of Stanford and the Boston region demonstrates that the elements that make technology transfer possible are highly contingent. Fruitful university-industry relations are difficult to imitate. They require knowledge sharing, in which information-rich communities develop and enforce each other.”). 146 The Carnegie Classification of Institutions of Higher Education, http://carnegieclassifications .iu.edu/ (last visited Oct. 17, 2019). The Carnegie Classification was created in 1970 by the Carnegie Commission of Higher Education (an education policy and research center) as a framework for classifying US colleges and universities according to a range of criteria that allow for the grouping of institutions of higher education into roughly comparable categories. All AAU member universities, for example, fall within the highest-level research category (commonly referred to as “R1” universities). 147 In 2016, 60% of all federal government R&D funding for universities went to the 60 AAU member institutions. aau by the nuMbers, aau (2018). And a limited subset of universities receive the lion’s share of US utility patents. See Top 100 Worldwide Universities Granted U.S. Utility Patents in 2017, nat’l aCad. Inventors, 2018, http://www.academyofinventors.org/pdf/top-100-universities-2017.pdf (last visited Oct. 17, 2019). 148 Shiri M. Breznitz & Neela Ram, Enhancing Economic Growth? University Technology Commercialization 5 (2012). 149 Id. at 6 (“For instance, one university might wholly own inventions while at another ownership is determined by the source of funding.”). 150 National Academies of Sciences, Engineering, & Medicine, supra note 128, at 45.

68 Research handbook on intellectual property and technology transfer versity technology transfer. It is impossible to understand contemporary university technology transfer without a sense of this political Ouroboros. Today’s universities participate in the volatile and ever-changing political system in manifold ways. The policy and political positions they take with respect to technology transfer reflect their long and complicated histories, their heterogeneous characteristics, and their changing roles and responsibilities in the face of the protean, competing requirements of the society they serve. Since their founding, American research universities have been pressed into service not only to interpret, explicate and solve a multitude of epistemological conundrums through scholarship and research, but also, increasingly, “to address real and pressing social needs such as health care, child care, economic development, and social mobility.”151 As Clark Kerr, who served as the path-forging president of the University of California system from 1958 to 1967, famously said: “the university is so many things to so many people that it must, of necessity, be partially at war with itself.”152 The complexity of what universities are called upon to do is put into perspective by the rather startling fact that higher education is the only industry regulated by every major federal agency.153 Universities are rightfully held to high standards, of course, particularly insofar as they are the stewards of public monies. But any discussions of the politics of university technology transfer must consider the multifaceted nature of the university enterprise and the challenges that universities and their associations confront as they navigate the complexities of working at the intersection of almost every facet of modern life, including that of academia and commerce.

151

Adam Daniel & Chad Wellmon, The University Run Amok!, Chron.

of

hIgher eduC., July 29,

2018. 152

Clark Kerr, The Uses of the University 7 (5th ed. 2001). See Hearing on The Rising Costs of Higher Education and Tax Policy Before the House Comm. on Finance, 113th Cong. (2014) (statement of Terry W. Hartle, Senior Vice President, American Council on Education). 153

4.

Bayh-Dole beyond patents Shubha Ghosh

I.

INTRODUCTION

The 1980 Bayh-Dole Act was a watershed moment for intellectual property (“IP”) management in United States universities. Permitting recipients of government funding to patent inventions that were the fruit of funded research reversed the course of commercialization efforts through industry-university collaboration. Pre-1980 hostility to university patenting was grounded in a concern with imposing a double burden. University research was funded through state and federal tax. Allowing patenting and the resulting increase in prices for commodities based on the patented invention would impose a second tax on consumers. Legislative findings, however, that the lack of patenting inhibited commercialization led to the enactment of Bayh-Dole Act, and a new regime of university engagement with commerce that persists today.1 The previous paragraph began with a reference to “IP management” and shifted to a focus on patents. The Bayh-Dole Act applies only to patents; the legislation was enacted as amendments to the Patent Act. This singular focus on patent reflects a concern with commercialization “technologies,” or the products of scientific research. However, as scholars like Jacob Rooksby have pointed out, Bayh-Dole Act is part of a sea-change in how universities seek to commercialize different aspects of their activities, whether book publishing, athletic events or teaching.2 Whether Bayh-Dole Act is a primary factor or merely a symptom of this sea-change is an interesting question; however, it may be an impossible question to answer. The overarching story may be one of universities needing sources of revenue beyond endowment income and tuition to further their missions. Patenting and technology commercialization may be one avenue; selling copyrighted teaching materials and trademarked athletic merchandising may be another. Despite the distinct legal treatment of patents under Bayh-Dole Act, it is important for universities to develop a comprehensive and integrated approach to IP management. This Chapter lays the foundation for such an approach. There are several reasons to consider application of Bayh-Dole Act beyond patents and to consider university IP management as a synthetic category. First, jurisdictions outside the United States have considered adopting similar legislation to Bayh-Dole Act. Proposed legislation consistently included all types of IP (patent, copyright and trademarks) and has not been limited to patents. In the past decade, legislatures in Brazil

1 See Wendy H. Schacht, The Bayh-Dole Act: Selected Issues in Patent Policy and the Commercialization of Technology, Cong. res. serv., Mar. 16, 2012, at 13, http://www.autm.net/Bayh _Dole_Act_Report.htm (last visited Oct. 17, 2019). 2 See Jacob H. Rooksby, Intellectual Property and the University: An Introduction, 54 duQ. l. rev. 1 (2016); Jacob Rooksby, The Branding of the American Mind (2016) [hereinafter Branding of the American Mind].

69

70 Research handbook on intellectual property and technology transfer and India have proposed comprehensive university IP management bills without the passage of new legislation.3 University IP management continues to be a topic of discussion in these jurisdictions as well as in others. A comprehensive approach to university IP management—i.e., an expanded application of Bayh-Dole Act beyond patents—is desirable to guide cross-national legislative debates. Second, even if the policy focus was solely on the United States, overlap across IP regimes requires a comprehensive treatment of university IP management. For example, software is subject to patent, copyright and trade secret protections. University management that ignores the relationships among areas of IP by turning solely to software patents may implement policies that are inconsistent and inadequate to meet commercialization efforts and the university mission. This overlap may also create problems for institutions within universities. Bayh-Dole Act mandated, without much guidance, that universities establish Technology Transfer Offices (“TTOs”). Offices were needed to identify patentable inventions through a disclosure process. Please note, TTO, Office of Technology Licensing (“OTL”), and Office of Technology Transfer (OTT) are synonyms. An open question is how TTOs should manage non-patent IP. Typically, governance of these other types of IP may be left to the general counsel’s office or other branches of the university, such as the athletic department or the university press. In some cases, management of non-patent IP may be left to individual departments or units operating under broad university policies (set forth by the university senate or the general counsel’s office). An example of this last approach is the implementation of online courses and the treatment of faculty generated content. Although there are benefits to a decentralized, ad hoc approach to IP management, comprehensive thought, at some level, is desirable for effective commercialization strategies consistent with the university mission. Finally, Bayh-Dole Act beyond patents may also guide efforts to structure university governance to align commercialization goals with the university mission. Professor Rooksby, and others, have demonstrated the tension between commercial gain and the university mission to promote the public interest. A university’s pursuit of short term profits through commercializing IP may come in conflict with the goals of basic research and broad-based learning and teaching. An overemphasis on branding efforts, especially through aggressive trademark enforcement, may be pitted against commitments to free speech and artistic expression. Furthermore, collaboration with industry may create financial benefits for some constituencies within the university at the expense of others, whose work provides a less amenable source for market returns. Even within the domain of commercialization, an emphasis on patents and new technologies may come at the expense of the humanities and arts, which may have some potential for university revenues, even if at a smaller scale. The hard question I turn to at the end of this Chapter is how to pursue a more comprehensive and integrated understanding of IP within the organization of university governance. Even if IP management as an integrated topic is possible, its substance may not be implementable. Nonetheless, the topic is worth raising and the debate worth having within some domain. This investigation of Bayh-Dole Act beyond patents is set forth in four parts. The first is to examine IP policies within specific universities to identify the ways in which different types of IP cohere, or more often fail to cohere, into an integrated policy. The universities examined in this Chapter are Syracuse University, Stanford University, University of Michigan at Ann 3 See Shamnad Basheer & Shouvik Guha, Outsourcing Bayh-Dole to India: Lost in Transplantation?, 23 ColuM. J. asIan l. 269, 271 (2010).

Bayh-Dole beyond patents 71 Arbor (“Michigan”) and University of Central Florida (“UCF”). Drawing on public statements of IP, the analysis identifies how universities deal with the overlap across IP regimes. The subsequent section turns to cases involving universities and IP as a way to single out the policy issues arising from university IP management. This portion of the investigation engages with Professor Rooksby’s work on the branding of the modern university and his thesis of a conflict between university IP and the public mission of universities. After an examination of the case law, the investigation turns to studies by the Association of University Technology Management (“AUTM”) on university IP management. This part of the Chapter identifies the policy recommendations of AUTM, as a leader in university IP policies. The Chapter concludes with recommendations for future research and conclusions on how understanding Bayh-Dole Act beyond patents can aid in better understanding university governance with respect to IP management.

II.

IP POLICIES IN SELECTED SCHOOLS

Universities have wide discretion in shaping IP policies. Bayh-Dole Act and IP statutes provide some limitations on this discretion. This section describes how four universities— Syracuse University, Stanford University, Michigan and UCF—have exercised this discretion. The choices of these four universities reflect my experience in studying and analyzing specific IP policies. Syracuse University is my home institution and employer; I chose it because I have direct engagement with its policies. Stanford and Michigan represent elite universities. UCF may seem like an odd choice for some readers; but, UCF is considered a leading mid-term public university with a mature and active IP management and commercialization policies. An initial point of contrast is the treatment of IP subject matter. One approach is to adopt a broad category, such as technology, with carve-outs that map onto the fields of patent and copyright. A second approach is to have separate policies for patent and copyright. A third approach is to treat all types of IP in an integrated fashion. A fourth approach is to use categories that map onto existing IP categories, such as works to refer to copyright and inventions to refer to patents. Syracuse illustrates the first approach; Stanford illustrates the second approach; Michigan illustrates the third approach; and UCF illustrates the fourth approach. A close look at the language from the publicized policies provide an illustration of the different styles of IP management. Syracuse’s IP policy is issued from the Office of the Provost. The policy contains a section entitled “Ownership and Management of Intellectual Property” and begins with a broad definition of technology and a carve-out: [T]echnology denotes inventions, discoveries, creations, technical innovations, information in various forms, including computer software, and tangible research property created in the course of research. Tangible research property includes, but is not limited to, notes, sketches, drawings, results of research or experiments, computer code or records, or any embodiment of the technology into any form. For purposes of this policy, technology does not include any copyright publication.4

4 3.23 Ownership and Mgmt. of Intell. Prop., syraCuse u., http://provost.syr.edu/faculty-manual/3 -23-ownership-and-management-of-intellectual-property/ (last visited Aug. 30, 2018).

72 Research handbook on intellectual property and technology transfer Copyright publications are subject to the following policy: Title to any copyright publication belongs to the member who has created the copyright publication, except in the case when it has been created under a sponsored program where there are ownership restrictions or in the case the copyright publication was created as part of a member’s explicit work assignment. Copyright publication includes, without limitation, written and artistic materials (such as articles, books, compilations, and visual and performing art works), whether or not protected by copyright. Under this policy copyright publication does not include software.5

The carve-out from the carve-out, software, is treated as “technology” and is subject to the ownership and management rules. Ownership and management of copyright subject matter incorporates work made for hire rules through a rule for publications “created as part of a member’s explicit work assignment.” Restrictions imposed through sponsored program financing also can shift the ownership rules prescribed by the policy. We will look at the implications of these ownership rules for management in more detail below. By contrast, Stanford University has adopted separate policies for copyright and patent. As for patents, Stanford’s policy states: All potentially patentable inventions conceived or first reduced to practice in whole or in part by members of the faculty or staff (including student employees) of the University in the course of their University responsibilities or with more than incidental use of University resources, shall be disclosed on a timely basis to the University. Title to such inventions shall be assigned to the University, regardless of the source of funding, if any.6

The policy also contains a definition of patents that integrates elements of United States and international law: A U.S. patent is a grant issued by the U.S. Government giving an inventor the right to exclude all others from making, using, or selling the invention within the United States, its territories and possessions for a period of 20 years. When a patent application is filed, the U.S. Patent Office reviews it to ascertain if the invention is new, useful, and nonobvious and, if appropriate, grants a patent – usually two to five years later. Other countries also grant similar patents. Not all patents are necessarily valuable or impervious to challenge.7

A definition of invention is also included: An invention is a novel and useful idea relating to processes, machines, manufactures, and compositions of matter. It may cover such things as new or improved devices, systems, circuits, chemical compounds, mixtures, etc. It is probable that an invention has been made when something new and useful has been conceived or developed, or when unusual, unexpected, or nonobvious results have been obtained and can be exploited.8

5

Id. Inventions, Patents, and Licensing, Research Policy Handbook § 9.1, stan., https://doresearch .stanford.edu/policies/research-policy-handbook/intellectual-property/inventions-patents-and-licensing (last visited Aug. 30, 2018). 7 Id. 8 Id. 6

Bayh-Dole beyond patents 73 The policy distills the doctrinal complexities of when an invention is patentable into several digestible paragraphs: Not all inventions are patentable. Questions relating to patentability are often complex and usually require professional assistance. 1. General criteria for patentability An important criterion of patentability is that an invention must not be obvious to a worker with ordinary skill in that particular field. It must also be novel, in the sense that it not have been previously publicly known or used by others in this country or patented or described in a printed publication anywhere. 2. Loss of patentability Inventions that are patentable initially may become unpatentable for a variety of reasons. An invention becomes unpatentable in the United States unless a formal application is filed with the U.S. Patent Office within 12 months of disclosure in a publication or of any other action which results in the details of the invention becoming generally available. 3. Circumstantial impairment of patentability Many other circumstances may impair patentability, such as lack of “diligence.” For example, unless there is a record of continuous activity in attempting to complete and perfect an invention, it may be determined that the invention has been abandoned by the initial inventor, and priority given to a later inventor who showed “due diligence.” [4]. International variation of patentability regulations Regulation covering the patentability of inventions and application filing procedures vary from country to country and are subject to change. It is important to note that an invention is unpatentable in most foreign countries unless a patent application is filed before publication.9

Finally, Stanford’s policy on patentable inventions also refers to unpatentable inventions: An invention, although unpatentable for various reasons, may still be valuable and important – for example, trade secrets and technical “know-how” encompassing proprietary information of a valuable and confidential nature. Agencies sponsoring research at Stanford usually require reports of all inventions, whether or not they are considered patentable.10

Stanford patent policy may be read as providing educational and legal guidance. The purpose is to inform those covered by the policy (faculty, staff and students) about the subject matter subject to disclosure and potential legal protection. Stanford’s copyright policy follows a similar structure. Key features are the treatment of pedagogical materials, institutional works and materials associated with online teaching. As to the first, the policy takes an approach favoring the instructor: In accord with academic tradition, except to the extent set forth in this policy, Stanford does not claim ownership to pedagogical, scholarly, or artistic works, regardless of their form of expression. Such works include those of students created in the course of their education, such as dissertations, papers and articles. The University claims no ownership of popular nonfiction, novels, textbooks, poems, musical compositions, unpatentable software, or other works of artistic imagination which

9 10

Id. Id.

74 Research handbook on intellectual property and technology transfer are not institutional works and did not make significant use of University resources or the services of University non-faculty employees working within the scope of their employment.11

However, the University claims ownership over so-called “institutional works”: The University shall retain ownership of works created as institutional works. Institutional works include works that are supported by a specific allocation of University funds or that are created at the direction of the University for a specific University purpose. Institutional works also include works whose authorship cannot be attributed to one or a discrete number of authors but rather result from simultaneous or sequential contributions over time by multiple faculty and students. For example, software tools developed and improved over time by multiple faculty and students where authorship is not appropriately attributed to a single or defined group of authors would constitute an institutional work. The mere fact that multiple individuals have contributed to the creation of a work shall not cause the work to constitute an institutional work.12

Works of non-employees may also be subject to university copyright ownership: Under the Copyright Act, works of non-employees such as consultants, independent contractors, etc. generally are owned by the creator and not by the University, unless there is a written agreement to the contrary. As it is Stanford’s policy that the University shall retain ownership of such works (created as institutional rather than personal efforts, as described in Section 2.B, above), Stanford will generally require a written agreement from non-employees that ownership of such works will be assigned to the University.13

Online courses and other recorded teaching materials receive separate treatment: Courses taught and courseware developed for teaching at Stanford belong to Stanford. Any courses which are videotaped or recorded using any other media are Stanford property, and may not be further distributed without permission from the cognizant academic dean (or, in the case of SLAC, by the director). Blanket permission is provided for evanescent video or other copies for the use of students, or for other University purposes. Prior to videotaping, permission should be obtained from anyone who will appear in the final program.14

Describing IP subject matter in much more detail than the Syracuse policy, the Stanford policy sets forth separate management rules for patent and copyright and for types of creations within each type of IP. Michigan’s policy avoids such nuances by treating different IP holistically. Key provisions as to IP management are: A. Intellectual Property made (e.g., conceived or first reduced to practice) by any person, regardless of employment status, with the direct or indirect support of funds administered by the University (regardless of the source of such funds) shall be the property of the University, except as provided by this or other University policy. Funds administered by the University include University resources, and funds for employee compensation, materials, or facilities. Rules in this Policy regarding ownership of copyrights are subject to

11

Copyright Policy, Research Policy Handbook § 9.2, stan., https://doresearch.stanford.edu/ policies/research-policy-handbook/intellectual-property/copyright-policy (last visited Aug. 30, 2018). 12 Id. 13 Id. 14 Id.

Bayh-Dole beyond patents 75

B.

C.

D.

E.

F.

G.

ownership rules directly addressed in the University’s copyright ownership policy, entitled “Ownership of Copyrighted Works Created At or In Affiliation With the University of Michigan” or successor Policy that is approved by the Regents. It is the obligation of Employees engaged in consulting and other activities with outside entities to ensure that their activities and agreements with third parties are not in conflict with the provisions of this Policy or other commitments involving the University. OVPR shall set and administer rules regarding the ownership of Intellectual Property made during outside employment activities (e.g., consulting). Employees should inform those outside parties with whom they make agreements of their obligations to the University. The University generally will retain ownership of Intellectual Property produced by Employees while participating in sabbaticals or other external activities if they receive salary from the University for such activity. Exceptions to this rule may be approved by the Vice President for Research. It is the responsibility of any such Employee to seek review by his or her appointing department (or equivalent) and OVPR in advance of entering into any intellectual property ownership agreements that may be associated with these activities or where such Employee is receiving partial salary. The University will not generally claim ownership of Intellectual Property created by students. (A “student” is a person enrolled in University courses for credit except when that person is an Employee.) However, the University does claim ownership of Intellectual Property created by students in their capacity as Employees. Such students shall be considered to be Employees for the purposes of this Policy. Students and others may, if agreeable to the student and OTT, assign their Intellectual Property rights to the University in consideration for being treated as an Employee Inventor under this Policy. The University will own Intellectual Property made by a former University employee if the Intellectual Property was made both (1) with substantial University faculty guidance or University resources and (2) during activity directly relating to and closely following employment. For example, if a graduate student researcher completes a research project and is no longer technically an Employee, and an invention is conceived during the creation of a dissertation or similar activity relating to the research involving faculty guidance, the University will own the patent rights related to the invention. This rule does not affect a graduate student’s ownership of the copyright on the dissertation itself. All Intellectual Property made under sponsored research agreements and material transfer agreements shall be owned by the University except where previously agreed otherwise in writing based on the circumstances under consideration. Such exceptions shall be approved and negotiated by OVPR; Intellectual Property subject to such an exception shall nevertheless be subject to the disclosure requirements of this Policy. Trade and service marks not incorporating previously existing University marks and that are related to University Intellectual Property and technology transfer activities are within the scope of this Policy as they are owned by the University and will be managed by OTT. University marks, including the University of Michigan Seal, are governed by other University policy.15

Consistent with this general provision, the specific Michigan’s copyright policy, referenced in Paragraph A, establishes copyright ownership by the university: The Default: Under U.S. copyright law, the University holds the copyright (as “works made for hire”) in copyrighted works authored by its EMPLOYEES who are acting within the scope of their employment. Otherwise, the University does not hold copyright in a work, unless the copyright has been transferred legally to it by written assignment or other process of law.16

15

Standard Practice Guide Policies § 303.04, u. MICh., http://spg.umich.edu/policy/303.04 (last visited Aug. 30, 2018). 16 Id. at § 601.28.

76 Research handbook on intellectual property and technology transfer Against this default rule, Michigan adopts a liberal policy of copyright transfer: B. Transfer of SCHOLARLY WORKS: In light of the default, the University, hereby, transfers any copyright it holds in SCHOLARLY WORKS to the FACULTY who authored those works—with the following conditions and exceptions. 1. Conditions—When the University transfers copyright in SCHOLARLY WORKS to FACULTY, under this policy, it reserves the nonexclusive right to: A. use SCHOLARLY WORKS for educational or administrative purposes consistent with its educational mission and academic norms and B. preserve, archive, and host SCHOLARLY WORKS in its institutional repositories, such as Deep Blue, where FACULTY can control the timing and scope of access to their copyrighted works. 2. Exceptions—The University does not, under this policy, transfer its copyright in SCHOLARLY WORKS: A. that are authored as required DELIVERABLES under a sponsored activity agreement; B. when that would put the University in violation of or conflict with an applicable contract or law; C. that are specifically commissioned by the University or are created as part of an administrative assignment to, for, or on behalf of the University; D. that are software under Regents Bylaw 3.10; or E. that are or have been transferred to the University in a writing (other than the Regents Bylaw 3.10 acknowledgment, which FACULTY sign as a condition of employment).17

Michigan’s IP policy streamlines ownership in the hands of the university and adopts rules for transfer. If there is a disagreement about ownership between University and faculty or staff, then the policy adopts certain dispute management procedures Finally, the UCF policy sets forth a general policy towards IP and university ownership of inventions and works. However, UCF adopts separate guidelines for patents, trade secrets and copyrights. While UCF asserts general ownership over patents and trade secrets created by its faculty, staff and students, the regulations distinguish between scholarly and non-scholarly publications. UCF ownership is claimed for the second type of publications. Centralized ownership is the default with exceptions for scholarly publications.18 Another important dimension to compare the IP policies of the various universities is the stated objectives as implemented in a particular choice of administrative design. For example, Syracuse University’s policy statement begins with the general set of goals: Syracuse University is dedicated to teaching, research, and the dissemination of knowledge. When these activities have been supported by the University and have resulted in the creation of properties that have economic interest and value, Syracuse University will retain title to, or have a fair and equitable income interest proportional to the University’s investment in, those properties that will reflect the legitimate interest of University investment as well as the traditions of academic freedom and pursuit.19

Faculty governance of IP matters is entrusted to the Committee on IP as follows:

17 18 19

Id. Technology Transfer, u. Cent. fla., https://tt.research.ucf.edu/ (last visited Aug. 30, 2018). See syraCuse u., supra note 4.

Bayh-Dole beyond patents 77 The Vice President for Research appoints a Committee on Intellectual Property including faculty from diverse academic units, at least one academic professional staff member, and at least one student. A normal term for service will be three years with membership renewable for one term. The Senior Vice President for Business and Finance and the Vice President for Research will serve on the committee as ex officio members. A senior member of the faculty serves as chair of the committee. The committee advises the Vice President for Research on the interpretation, administration, and implementation of this policy. Any appeals of the decisions of the Vice President for Research are directed to the Chancellor.20

Overall management of IP is as follows: The Office of Sponsored Programs (OSP) has the primary responsibility within Syracuse University for managing and administering matters involving technology developed at Syracuse. OSP will consult with members regarding the best means for development and transfer of the technology created by members. It may be necessary, in accordance with the terms of this policy, that any assignments, licenses, transfers, applications, registrations, or any other documents that are necessary to evidence the University’s ownership in technology be executed by members.21

The TTO works in conjunction with the OSP to administer the guidelines in identifying invention disclosures and pursuing IP strategy. University policies are set of course by the Chancellor pursuant to the goals of the policy. At Stanford University, the OTL assumes a more centralized role in administering the Stanford’s IP policy. According to its mission statement: OTL has signature authority on behalf of the University for license agreements, material transfer agreements, industrial contracts, and other agreements that pertain to intellectual property. University faculty and other inventors are not authorized to sign agreements that obligate the University to assign or license intellectual property rights to another entity. OTL and OTL personnel have no financial stake in University licenses. If the administrative overhead portion of licensing revenue exceeds OTL’s budget, the remaining funds are allocated to the OTL Research Incentive Fund under the control of the Dean of Research, for the support of research and education across the University. As a result, we believe OTL functions effectively as an unbiased agent serving the goals and interests of the University, as well as those of inventors and others. Inventors have many vital and valued roles during the licensing process, while final decisions are made by OTL. The management of University intellectual property is complex because many different interests must be considered. OTL works at the interface of science, business, and law within an intersection of academia, industry, and government. We know that the key to our success is our ability to work well with our diverse constituencies—inventors, departments, schools, industry, the U.S. Government, and the University.22

This centralized role is consistent with Stanford’s mission statement with respect to technology licensing: Stanford’s Office of Technology Licensing (OTL) has a mission to promote the transfer of Stanford University technology for society’s use and benefit while generating unrestricted income to support

20

Id. Id. 22 See OTL and the Inventor: Roles in Technology Transfer, stan., http://web.stanford.edu/group/ OTL/Old/inventors/inventors_policies.html#signature (last visited Aug. 30, 2018). 21

78 Research handbook on intellectual property and technology transfer research and education. This mission statement guides OTL’s decisions for how to best transfer technology to benefit the public and the University. Stanford has a long, successful history in technology licensing marked by collaborative relationships with inventors and by flexibility in negotiations. OTL is committed to helping faculty, staff, and students navigate the processes of patenting and licensing with the goal of transferring their research to industry in order to benefit society. Stanford University has an extensive and robust set of approaches and policies that support good stewardship of University intellectual property assets while encouraging innovation, entrepreneurship, and strong relationships with inventors and companies of all sizes. OTL is responsible for managing the intellectual property assets of Stanford for the public good and we assume the risk and expense of licensing on behalf of the University. OTL appreciates the inventors’ input throughout the process and carefully considers various interests before ultimately making licensing decisions that it believes will effectively transfer the technology for society’s benefit. An essential element of OTL’s approach is that financial interests are not the primary consideration when making licensing decisions. OTL exercises professional judgment to promote the efficient and effective transfer of technology while conforming to University guidelines and policies. Royalties come when the technology transfer is successful.23

Ensuring successful technology transfer for the public good requires an office that steers inventions and creative work from the space of the researcher to the hands of the consumer. Michigan’s TTO has a concise mission statement: “Our mission is to effectively transfer University technologies to the market so as to generate benefits for the University, the community and the general public.”24 The Office also provides a concise self-description of its organization: U-M Tech Transfer is the University unit responsible for the commercialization of University research discoveries and reports to the U-M Office of Research. We enhance these research discoveries to encourage licensing and broad deployment with existing businesses and newly-formed U-M start-ups. U-M Tech Transfer has earned a reputation for performance among the top 10 of all universities.25

Further detailed provisions provide some familiar facets of Michigan’s TTO operations: Our tech transfer team offers a full set of services to ensure effective technology transfer: • Invention Reporting Facilitation—We work with researchers and faculty to provide advice about potential tech transfer issues during research activities and to assist in the invention reporting process. • Patenting and Other Protections—We provide guidance in planning an effective patent, copyright, or trademark strategy and handle all implementation details during the protection stage. • Start-up Assistance—We provide proactive assistance in analyzing potential opportunities to form a start-up based upon U-M technology and encourage this interaction during the early invention reporting process. Our skilled New Business Formation Staff also provides hands-on business assistance, project planning and links to funding and people resources. • Licensing—We assist in technical and market assessments and actively market U-M technologies to industry partners. We create secrecy, evaluation, material transfer, option and license agreements with these industry partners to effectively get our technology into commercial use. • Legal Support—Our two tech transfer staff attorneys, in partnering with the Office of the General Counsel, provide legal guidance and assistance for all of our activities.

23 24 25

Id. See u. MICh., supra note 15. Id.

Bayh-Dole beyond patents 79 •

Decision Support—Our business and administrative staff provide information and guidance to conduct our business and expedite decisions by internal and external partners. We also advise interested parties on U-M policies & procedures including conflict of interest issues.26

One striking feature of the organization is the discussion of “partnering with the Office of General Counsel.” None of the other mission statements mention this, although such partnering may be implicit within the broader structure of the University. An interesting question for further research is exploring how such explicit (or implicit) partnerships work in practice, and the ways in which specialization occurs with respect to legal matters and types of IP. For example, trademark issues may be the exclusive domain of the Office of Legal Counsel while the TTO deals primarily with patents. Furthermore, all litigation matters may be directed to the Office of General Counsel. Finally, UCF has a broad mission statement: The UCF Office of Technology Transfer (OTT) brings discoveries to the marketplace through intellectual property protection, marketing and licensing processes and connects UCF researchers with companies and entrepreneurs to transform innovative ideas into successful products. Each invention is evaluated by a team of professionals with combined expertise in science, business and law. A customized intellectual property protection plan, and commercialization strategy is designed for each selected discovery based on market research and in-depth business opportunity analysis. The OTT manages hundreds of promising technologies and is actively seeking industry partners to bring these innovations to market. We support the mission of UCF’s President to be “America’s Leading Partnership University®” and to “achieve international prominence in key programs of graduate study and research.” Our faculty are integral to the success of the university. We support our world-class faculty by promoting the progress of their research for the benefit of the university, the community, economic development, and to further advance discovery and innovation. The benefits of licensing university technology to industry partners include paid royalties to the inventor’s college, department and to the inventor; as well as strengthened industry relationships. Industry partnerships can provide opportunities for sponsored research and product development. The Office of Technology Transfer at UCF is uniquely situated in a hub of innovation at Research Park. Our office is surrounded by companies that are world-renowned innovators and leaders and whom are positioned to acquire disruptive technologies.27

The Office is largely autonomous in its focus on managing the patenting and invention commercialization process: The OTT at UCF began in 1989, and actively works to protect and bring UCF researchers’ ideas to market. Our expertise combines the disciplines of science, business, and law and we manage over 900 active intellectual property cases in patents, trademarks, and copyrights. Our team of professionals evaluates each Invention Disclosure submitted to the OTT. Based on the disclosed information and internal research, a customized intellectual property protection plan, and commercialization strategy is designed for qualified technologies. Each technology is managed through the duration of its life cycle—from disclosure to deal—by the same licensing professional. Each licensing professional has a scientific area of expertise. As part of the Office of Research and Commercialization, the Office of Technology Transfer (OTT) is responsible for managing the UCF’s (UCF) intellectual property assets and supporting the commercialization of discoveries made at UCF. UCF faculty, students, and staff make new discoveries every day. Our team of professionals works with these inventors to protect and promote their discov-

26 27

Id. at About Our Organization. See u. Cent. fla., supra note 18.

80 Research handbook on intellectual property and technology transfer eries, so they may be used in publications, presentations and in the marketplace. UCF inventors are able to submit invention disclosures to our office either online or in our office. Once we receive the invention disclosure, we can meet with the inventor to review the discovery, gain additional insight, and uncover potential applications. We can then evaluate the invention for feasibility, novelty, and market potential to determine how to best proceed in the IP process. We continue to work with the inventors to identify industry partners to launch the technology into the market. Once the technology has been licensed to an industry partner, the OTT’s licensing professionals continue to monitor the technology milestones and the distribution of royalties to the inventor(s), and to the respective college and department for additional research funding.28

A Venn-Diagram graphic on the OTT’s website shows that the Office operates at the intersection of Science, Business and Law, in the overlap of these three labelled “Technology Transfer.” Note that the UCF statement of mission and organization does not refer to the Office of General Counsel, suggesting the autonomy and focus of the University’s OTT. The purpose of this section has been to set forth the details of how Bayh-Dole Act has been implemented by select universities. With this background, we turn now to the broader context of university IP in action and the development of theoretical and analytical perspectives of how IP management fits into the mission and organization of universities.

III.

UNIVERSITIES AND IP LITIGATION: CONFLICTING MISSIONS?

Past disputes can provide context for current debates. The first published opinion regarding university IP was in 1930. Relevant cases from that year involved a trademark dispute over the term “University Store”29 and a copyright dispute involving Yale University Press.30 The search uncovered over 700 reported opinions since 1930 involving IP and universities. This body of opinions shows the increasing importance of IP for universities. They also show that the number of IP opinions involving universities is small compared to the larger body of opinions involving IP disputes. Professor Rooksby in “The Branding of the American Mind” delineates university involvement in IP litigation, particularly in the areas of trademark and copyright. His thesis is that university litigation policy constitutes a distraction from the public minded mission of the University towards commercialization and profit. It is important, however, to view the actions of universities in a broader context. Universities are corporate entities, even if endowed with a public mission. That being said, as corporate entities, universities are in the role of managing assets to fulfill their mission. Professor Rooksby’s caution is well-taken and well-argued. However, it is important to consider the realities of the financial needs of universities and the need to modernize how they carry forth their mission. This section examines university IP litigation within this broader context and provides an initial foundation for the theoretical discussion at the end of this Chapter. Reported opinions arising from disputes over IP in universities span copyright, trademark, patent and tradesecret. Professor Rooksby examines disputes that arise, but were not

28 29 30

Id. Panitz v. U. Clothes, 40 F.2d 811 (D.C. Cir. 1930). Yale U. Press v. Row, Peterson & Co., 40 F.2d 290, 291 (S.D.N.Y 1930).

Bayh-Dole beyond patents 81 the subject of litigation or a final judicial opinion. Note, the majority of these non-litigated disputes involve students and untenured or non-tenure track faculty members at universities. Many of these defendants, particularly students, operate in the shadows of IP policies, and most often outside any formal agreements, which might provide the basis for formal adjudication. My search of the published judicial opinions uncovered over one hundred opinions since the 1930s, including the two examples discussed at the beginning of this section. With respect to copyright, these published opinions originated as disputes involving university presses and notice and take-down under the Digital Millennium Copyright Act (“DCMA”). Both sets of cases provided a formal basis for the disputes to enter into the legal system. University press disputes were often assignment and licensing disputes; notice and take-down were channeled through the formalistic mechanisms of the DMCA. Formalism is worth emphasizing here because of the availability of informal mechanisms within universities which can remove the disposition of a dispute from the transparency of the judicial system. With all their limitations, courts do provide for a public deliberation of social and legal policy issues, even within purely private rights-based disputes. What happens within the university processes stays within the university often without the benefit of sunshine. Recent debates over the disposition of student conduct involving sexual abuse and harassment (among students and between students and staff) highlight the ways university processes can avoid publicity. Perhaps over time, a more thoughtful balance of private interests (especially those of universities) and the public can arise as more sunshine is brought to disputes, whether involving highly sensitive matters such as sexual violence and abuse of position or commercial matters involving IP. But to echo Professor Rooksby’s points, seemingly commercial matters of contract and property can have implications for the public: [D]espite the promise of higher education as a cultural and knowledge commons, copyright and related campus policies all too often frustrate student and faculty innovation, serving to impede the flow and diminish the value of their cultural and knowledge outputs to society. Instead of finding higher education to be a site of unfettered intellectual vibrancy, many students have come to realize that their own college or university views their work products as being subject to institutional or third-party ownership. Faculty face similar incongruencies, as a lack of clarity in college or university policies or institutional overreach both undermine what should be regarded as clear rights of ownership involving faculty creations, whether curricular or extracurricular.31

One can question to what extent a university operates as a commons. In the final section of this Chapter, I question this idealized notion of a shared culture at a university. The battle over private interests is in conflict with the university’s mission of promoting knowledge and information for the benefit of the public. Trademarks provide an example of how commercial rights and public deliberation can intersect in the university setting. In Gerlich v. Leath, the Southern District of Iowa found against Iowa State University, which refused to license use of its trademarks to a student organization dedicated to the legalization of marijuana.32 The district court found this denial impermissible

31

Branding of the American Mind, supra note 2, at 203. Gerlich v. Leath, 152 F. Supp. 3d 1152, 1157 (S.D. Iowa 2016), cert. denied, 4:14-CV-00264-JEG, 2016 WL 10567632 (S.D. Iowa Feb. 17, 2016), and aff’d, 847 F.3d 1005 (8th Cir. 2017), and aff’d, 861 F.3d 697 (8th Cir. 2017). 32

82 Research handbook on intellectual property and technology transfer viewpoint discrimination by the state university. While the university at first worked with the organization in the use of the university logos on the organization’s t-shirts, which advocated the legalization of marijuana, the university recanted when it faced backlash from constituencies, including alumni and members of the public. The university decided that it “would refrain from approving any reorder requests for the previously approved t-shirts.” Furthermore, any future licensing requests from this particular student organization would require pre-approval from the university administration before the request was submitted to the licensing office. Finally, the university revised its licensing guidelines to deny licensing of the trademarks for uses that suggested “promotion or endorsement of dangerous, illegal, or unhealthy products.” After denial of licensing requests from the university, the student organization sued. The university, according to the district court’s ruling, had no substantial interest in denying the group’s request. Therefore, the denial was impermissible viewpoint discrimination under the First Amendment of the United States Constitution. Relevant to the intersection between private rights and the public interest is the court’s ruling that the university’s decisions were politically motivated. The licensing guidelines had been revised in “direct response” to political backlash in newspaper articles and communications from angered constituencies. Furthermore, the student group had been singled out through the pre-approval process that did not apply to other organizations seeking licensing approval. While the university did clarify its guidelines to deny licensing for the promotion of illegal activity, the court found that the student organization sought to change the law and not promote illegal uses of drugs. Also suspicious was the university’s flexible discretion in allowing licensing for student organizations that featured guns or swords in conjunction with the university logo even though these items might also be illegal or dangerous. Finally, the court rejected the university’s argument that it was protecting its own “government speech” by limiting use of the university trademarks. A year after the district court’s decision, the United States Supreme Court ruled that denial of trademark registration can constitute impermissible viewpoint discrimination in its Tam v. Lee decision. But contrary to the Supreme Court’s treatment of trademarks as private speech, the Iowa district court recognized the public dimensions of university logos. They were not simply a species of university speech which the university administration had the sole discretion to authorize. Instead, state university trademarks had a public dimension. Once the state university developed a policy of licensing, it had to exercise its approval authority on politically neutral terms.33 Note, however, that the district court’s ruling puts the university in an all or nothing position with respect to trademark licensing. Under the court’s reasoning, the university could decide not to license the trademarks to any group. However, to take the court’s reasoning further, if the university’s total licensing ban decision was based on suppressing all student speech, then a total licensing ban might be deemed impermissible as politically motivated. Banning all licensing might be ruled as suppressing all viewpoints through the use of the trademarks except for the viewpoints of the university administration. In any case, a total licensing ban might not be desirable. Universities benefit from creating associations with diverse student groups, even ones espousing politically unpopular positions. Trade secrets afford another area of university IP law to engage with matters of public interest. However, courts have in general favored universities in protecting their trade secrets. Most often trade secret issues arise through freedom of information requests made to public 33 John Roberts Manufacturing Company v. University of Notre Dame Du Lac, 258 F.2d 256 (7th Cir 1958) (explaining unfair competition action involving university insignia rings).

Bayh-Dole beyond patents 83 universities under state law. State public records or freedom of information laws create an exception from disclosure for IP, such as trade secrets. As a result, public requests for information on animal testing,34 for patient records (from university medical research or hospital facilities)35 and for contracts36 have been denied because the underlying information is deemed a protected trade secret. Alumni donor information has also been found to be a protected trade secret despite the argument that alumni lists are distinguishable from customer lists, a common form of business trade secret. Most controversial is the treatment of course syllabi. Some public minded groups have sought to police the content of teaching by requested course syllabi. However, courts overall have treated syllabi as confidential information protected from disclosure.37 Two lessons follow from this line of cases. First, commercial value is a requirement to protect certain confidential information as a trade secret. In finding that some university-based information is a trade secret, courts are highlighting aspects of university activity that are essentially commercial. The case of alumni lists is particularly illuminating. Alumni are “customer-like” and the information in an alumni donor list may be readily analogized to that in a customer list. That analogy is by no means foreordained. The similar treatment of the two suggests how a university operates in many ways like a business, creating goodwill and relationships deemed worthy of legal protection. The second lesson is the treatment of information within a university. As a “knowledge commons,” the university produces information that is public in nature. Trade secret law demarcates differences among types of university information, designating some information as proprietary and others as open. It is perhaps noncontroversial that business information, like contracts, fall on the private side of the divide. But, even that case should be challenged. Transparency is needed to detect discrimination in salaries or in public-private partnerships that have broader implications related to conduct or the public. Syllabi are particularly troubling. There is a sense within universities that requests for course information is a form of harassment by groups that seek to police for left-leaning academic content. At the same time, though, there is public interest in knowing what is taught in the classroom. For example, a science class may be teaching controversial religious precepts (such as creationism), or a history class may be teaching outdated or simply incorrect information. Protecting syllabi as trade secrets turns the classroom into a commercial forum and detracts from the knowledge commons, potentially. To once again echo Professor Rooksby’s comments:

34

See, e.g., Robinson v. Indiana Univ., 659 N.E.2d 153, 157 (Ind. Ct. App. 1995). See, e.g., U. of Connecticut v. Freedom of Info. Comm’n, 36 A.3d 663, 669 (Conn. 2012) (explaining database of subscribers and donors as protected trade secret). 36 See, e.g., Besser v. Ohio State Univ., 732 N.E.2d 373 (Sup Ct Ohio 2000) (explaining hospital records requested as challenge to hospital merger). This case states that the information is not a trade secret with the exception of “one page of the preliminary business plan and the memorandum and related list containing the names of the top patient-volume medical personnel at Park Medical Center.” State ex rel. Besser v. Ohio State Univ., 732 N.E.2d 373, 381 (Ohio 2000). 37 See Encore College Bookstores v. State University of New York at Farmingdale, 663 N.E.2d 302 (N.Y. 1995) (holding booklist information is exempt from FOIA request); Mohawk Book Co. v. State University of New York, 732 N.Y.S.2d 272 (N.Y. App. Div. 2001) (explaining course syllabi is not exempt from FOIA request). 35

84 Research handbook on intellectual property and technology transfer Faith is based on trust, which requires accountability, which demands transparency. The public may soon have little faith in the ability or desire of for-profit companies to serve society’s best interests, in part because their operations are generally kept secret.38

What is missing in this analysis is identifying the locus of decision-making within the university. In other words, in whom should the public place its trust within university decision-making? If the university professor is the trusted decision-maker for course content, then limits on policing professor decision-making might be a desirable goal, particularly if the prying eyes are seeking to identify content that is deemed undesirable or controversial. But there are limits to how much trust one should be willing to give to one class of persons. If alumni relations offices are the trusted loci for establishing alumni connections to promote giving, then alumni records may be shielded from scrutiny, not because of the commercial nature of alumni-university relations, but to limit interference with internal practices. But, again, there might be limits to how much trust is desirable. University IP policies require careful considerations of university governance, and it is that inquiry which Professor Rooksby invites as a scholarly mission for those who study IP and the university as organizations. And it is to questions of university governance that this Chapter turns in the final two sections. To recap the argument so far, the Bayh-Dole Act imposed a structure of rules pertaining to university patents; it left open questions for other areas of IP. This Chapter has shown how universities have pursued IP policies beyond patents both through its internal TTO and through litigation. What broader principles of university governance arise from these disparate practices? The next section finds answers in the work of AUTM; the following and final section offers some analytic insights and theoretical conclusions.

IV.

WHAT AUTM TEACHES

While universities may have wide latitude in designing their IP policies, certain best practices have emerged out of the experiences of the range of universities in the United States. One indication of a trend towards best practices is the promotion of guidelines by AUTM. Founded in 1974, the Society of University Patent Administrators helps move research from the lab to the marketplace for public use and benefit. Moreover, AUTM is a nonprofit organization dedicated to bringing research to life by supporting and enhancing the global academic technology transfer profession through education, professional development, partnering and advocacy. One important document disseminated by AUTM to its members is the Technology Transfer Manual, now in its third edition. The recommended best practices illustrate how universities can integrate their technology transfer policies across the range of IP. One example of emerging norms for IP management is the discussion in the Manual of the goals of the TTO. As one author notes: Most universities historically launched their technology transfer operations in a reactive manner rather than from a strategic or proactive perspective. As a result, licensing activity tended to be some-

38

Branding of the American Mind, supra note 2, at 280.

Bayh-Dole beyond patents 85 what reactive, very rules- and compliance-focused, extremely risk averse, driven much more from the licensee partner’s perspective, rather than from the goals and objectives of the academic licensor.39

Universities need to adopt, the author concludes, the goals of economic development as part of their teaching, research and service missions. What this means is to both identify and manage their IP assets more strategically with the aim of meeting the needs of constituencies locally, nationally and internationally. As one author describes the shift: Many [universities] have acknowledged the addition of a fourth leg (economic development) to the traditional three- legged stool (teaching, research, and service) often used to describe the mission of the university. More and more universities have developed a strategic focus for their technology transfer activities—and many no longer articulate or prioritize licensing revenue as one of the top priorities for their efforts (e.g., many now give higher priority to objectives such as company creation, supporting faculty recruitment and retention, enhancing research funding, creating an entrepreneurial culture, attracting venture investment to their regions, and related objectives).40

But this aggressive movement towards commercialization brings universities more intimately into the commercial sphere. University governance needs to adjust to this broader role of the university and serve as an effective counterbalance. TTOs are increasingly viewed as business units of the university. One example, from AUTM, is the University of Pittsburgh. University of Pittsburgh believes that, to be a successful business unit, its Office of Technology Management (OTM) must provide superior service to both the faculty, staff and potential industrial partners. For the university community, the services that OTM provides include: ● managing the intellectual property and completing good business deals with startup ventures and other licensees; ● providing education and counseling on intellectual property and technology transfer. For industry, the services that OTM provides include: ● acting as a bridge between science and business; ● providing access for the business community to intellectual property assets of the University of Pittsburgh; ● working with industrial and independent investors to match their interests with the available portfolio of intellectual property or by introducing them to world-class University of Pittsburgh researchers who are developing the next generation of innovations; ● aiding companies in forging business relationships between university innovators and federal government contractors.41

39 2005 AUTM President Mark Crowell, A Philosophy of Licensing and Technology Transfer for Academic and Nonprofit Research Institutions, 3 autM teCh. transfer praC. Manual 1, 2006, (last https://www.autm.net/AUTMMain/media/ThirdEditionPDFs/V2/TTP_V2_P1_C1.pdf?ext=.pdf visited Oct. 17, 2019). 40 Id. 41 Christopher C. Capelli, Technology Transfer Office as a Business Unit, 3 autM teCh. transfer praC. Manual 2.1, 2006, https://www.autm.net/AUTMMain/media/ThirdEditionPDFs/V2/TTP_V2 _P1_C2-1.pdf?ext=.pdf (last visited Oct. 17, 2019).

86 Research handbook on intellectual property and technology transfer These functions shape IP management and the organization of decision-making: After an invention has been made by the innovator at the university, an invention disclosure (ID) is submitted to the OTM. At the University of Pittsburgh, the ID is submitted electronically so that it becomes part of the database that is used for tracking and reporting of all inventions. Once a month, a separate committee, called the Technology Transfer Committee (TTC), meets to determine which IDs should be considered for filing of a patent application, which should be held for further data, and which may be marketed as unpatented research tools or copyrights. The committee is composed of various science and engineering faculty appointed by the provost and the senior vice chancellor, health sciences for specific terms, as well as representatives from OTM and the Office of the Provost. Several members of the current TTC are inventors. The committee began operating in January 1993. The TTC has become extremely effective in its selection of the most promising inventions for patent application. In addition, the TTC has been most helpful to inventors in helping them focus on a better or more complete presentation of the invention. OTM provides the organizational support for the TTC.42

How these bureaucracies operate is the subject of future research. But their emergence demonstrates how TTOs become more corporate and integrated with the larger operations of the university. Changes in goals and structures require a rethinking of how universities operate as organizations to manage IP challenges beyond Bayh-Dole Act. Even the model of University of Pittsburg still seems to be heavily focused on patents and inventions. The challenges of rethinking the organization of a university, structures attendant to this organization, and their normative implications are the subject of the next and final section of this Chapter.

V.

LOOKING FORWARD

The American university has undergone many challenges as the organization navigates the pulls of commercialization and the demands of the many constituencies it serves.43 At any point in time, a university may be serving many goals, some purely private and some purely public.44 As universities seek revenue sources from athletics, teaching and research, they act not all that differently from for-profit corporations. At the same time, as universities seek to diversify their faculty and student bodies and serve the needs of local communities, they act like charitable organizations. Universities cannot fit into one criterion, such as public purpose. Instead, universities are multivalent, and managed along multiple criteria. This thesis is the principal one of this Chapter. Interactions between private interests and public rights within universities reflect how researchers and scientists negotiate the need for practical and pure scientific knowledge.45 IP law reflects this complex dynamic, and it should continue to do so as universities shift between

42

Id. See Laurence R. Veysey, The Emergence of the American University 252 (1965) (tracking the development of the university’s role from that of providing religious education to that of pure research within a liberal culture). 44 Id. at 346–8 (setting forth various business models for universities). 45 See Steven Shapin, Never Pure 213–14 (2010) (analyzing the dilemma facing the industrial scientist navigating the tensions between the culture of universities and that of industry). 43

Bayh-Dole beyond patents 87 models of pure research and pure commercialization. The multivalent model presented in this Chapter offers a positive account of university governance that can guide IP policy. What distinguishes this Chapter is its focus on the university as an organization. While the university is an institution, embodying specific values of a community (whether local, national or global), a university is a collection of individuals coming together to interact in a community. As an organization, a university has to choose its governance structure, including its management of IP rights and the attendant relationships among researchers, teachers, students and administrators.46 When I ask whether a university pursues pure research, pure commercialization, or a mix, I am asking a question about how a university is organized. After World War Two, as the centers for university excellence shifted to England and the United States and away from devastated Germany, new visions of the university came to the fore in the United States.47 The mega-university took the bureaucratized university to new heights, serving large populations of students and providing secure positions for academics and staff. Scientific research and development became the foundation for innovation policy as federal policymakers focused on ways to avoid the catastrophic downturn of the Great Depression. Federal agencies would provide funding for university researchers. In turn, the university researchers would generate new ideas, new products and new inventions to feed industry. Even if universities did not commercialize products, they provided the resources for commercialization that would invariably feed the American consumer. But this scheme did not rule out the possibility of new companies and industries springing forth from within the university. However, there was a sense that the business of universities was not business, even if universities would sometimes be in close partnership with entrepreneurs.48 Against this historical background, we can identify three models for the contemporary university, particularly as we understand its relationship to IP. The first is the model of the university as the producer of pure research. The second is the model of the university as a commercial entity with a public purpose. The third is the model of the university as a pure commercial entity, no different from a for-profit corporation. Below, each model is examined. The analysis will set the foundation for our understanding of specific patent and copyright doctrines as applied to universities. A.

Model of Pure Research University

One idealized view of the university is as a producer of pure research49 with its core constituencies, faculty and students, pursuing questions independent of commercial or financial concerns. It would be hard to deny the inherent value of free and open inquiry, untethered from concerns of profit and internal rates of return. The interesting question is to what extent universities can, in practice, match this ideal. Resource constraints and scarcity of time may limit how far faculty and students can thrive in a rarified environment of free-wielding inquiry, 46 See, e.g., Intellectual Property Management Policy, Makere u., Mar. 13, 2008 (on file with author) (setting forth IP and governance rules in a university in Uganda and illustrating the importance of organization rules for universities in developing countries). 47 Id. at 312–13. 48 See, e.g., Wendy Schacht, The Bayh-Dole Act: Selected Issues in Patent Policy and the Commercialization of Technology (2009). 49 I do not mean to exclude teaching from the mission of the university by using the word research. I am using that word expansively to include inquiry, and teaching would be part of the broader meaning.

88 Research handbook on intellectual property and technology transfer guided solely by the rigors of particular disciplines. Santiniketan, the lovely rural university created by Bengali poet and novelist Rabindranath Tagore, is the closest I have seen to such a utopia. Liberal arts colleges, tucked away in United States hinterlands, sometimes emanate pure intellectual pleasure and engagement even if marred by pressures of upward mobility and maintenance of social standing. St. John’s University, at both its Santa Fe and Annapolis campuses, requires commitment to a four-year great books program that immerses students in the development of Western Civilization. If one were to construct a world from nothing, the need for some institution that allows for unadulterated thinking would be readily apparent. and that institution would have many of the characteristics of actual universities. While the temptation to exult pure research may stem from the desire to seek knowledge for knowledge’s sake, there is a practical reason to focus on pure research. Concentration on fundamental questions allows disciplines to flourish and evolve, whether that discipline is in the natural sciences and the search for understanding the world in which we live, or in the humanities and the search for how thinking and emotions evolve. This practical turn does not tarnish the purist model of the university. Human inquiry is not solely about having one’s head in the clouds, but also about being aware of the ground one walks upon. Pure research, to put in bluntly, can be both theoretical and applied. However, the purist model starts to tarnish with finance considerations. To live the life of pure research requires resources. Some institutions, like Santiniketan, may have the benefit of healthy endowments, but such endowments have to be maintained, leading to dull, practical questions of where to invest, how much, and where to place the returns. Institutions without endowments have more basic questions to ask about sources of money to run a going concern. Practical research may readily become one revenue stream, requiring engagement with the world of commerce. Once that happens, the luster of the pure research model fades, and the choice has to be made whether the university becomes a profit center or continues with its idealized mission. The second model suggests a way that the institution can accomplish both. B.

University As Commercial Entity With Public Purpose

A second idealized view permits universities pursuing commercial ends for a public purpose, which can be construed in many ways. As mentioned at the end of the previous section, the dual purpose university satisfies the need for profit and the pursuit of pure ideas. A university can be run as a business through the identification of revenue streams. These streams can include the commercialization of products and services developed within the university such as courses, patentable inventions, copyrightable content and branded merchandise. But what keeps the university from turning into an amusement park, or a cruise ship on land, is the demand of channeling profits towards public goals. The most likely candidate for these public goals would be the pursuit of pure research. But like running water, currency can move towards many destinations, and the ocean of pure knowledge may be only one. As a nonprofit organization, universities need to put their profits back into the organization rather than making a payout to residual claimants, whether shareholders, partners or members. By putting profits back into the organization, a university uses commercialization presumably to finance its public purposes. The challenge for this model is the implication for the organization of for-profit universities. There is nothing within the model that rules out the possibility of a for-profit university, so long as the entity uses it profits for public purposes. Defining these public purposes for

Bayh-Dole beyond patents 89 a for-profit university is the difficult issue. Public purposes might include scholarships for students, research support for faculty or funding for local community projects. But public purpose might also include international programs and other initiatives that extend beyond the traditional domain of the university. Conceptually, public purpose entails redefining the residual claimants of the surplus from a for-profit university to include a broader class of beneficiaries in the community. Doctrinally, the for-profit university could best be understood as a type of benefit corporation with the beneficiaries being defined by the founders and set forth in the corporate documents. One concern with the public purpose model is that the definition of public purpose can expand to include interests that might seem more akin to private ones. For example, public benefits may align with private interests of founders or professors, such as the local symphony, regional art museum or with political causes and campaigns. Furthermore, the commercial goals of universities to accumulate financial surpluses might lead the university to focus solely on commercialization efforts, losing sight of any broader public benefit, however altruistically set forth in the founding documents. A cynic might say that universities inevitably collapse into the model of pure commercial activity, no different from other for-profit business organizations. But even without accepting that cynical position, a realist might still predict the inevitable collapse of the pure or modified models into the third model described in the next section. C.

University As Pure Commercial Entity

The third idealized view is the university as a pure commercial entity, no different from a for-profit corporation. To call this model idealized may seem misguided as reducing a university to the status of any other business entity eviscerates the institution of any noble ideals of learning and research. But the university as a locus of pure self-interest is idealized in the sense of serving as a rarified model for the purposes of analysis, a benchmark against which to gauge policy choices. This model is also idealized in the sense that it is wholly unrealistic, ignoring not only important virtues, but practical details of the university mission. Among some, there may be superficial appeal in treating universities like any other business entity, driven by the profit motive and the necessity of meeting payroll. One implication of this conception is that universities should be left to fail, no matter how big they are. In other words, if they cannot produce certain measures of success, whether by profits, graduates or research, they should fail. For those familiar with post-Thatcher academia, this picture should be familiar as British universities and academies are subject to unavoidable scrutiny of university outputs. With the example of contemporary British universities, it seems that this model is not pure fantasy and is one that seems to have been adopted. Should this model be ruled out on its face? Faithful adherents to the pure research model might say yes. I may be one of these adherents, but I am also willing to play the advocate for the pure commercialization model. Organizational success is important for society. One measure of success is survival in a competitive environment. Organizations that survive a competitive environment have characteristics of efficiency in delivering outputs that society finds desirable. Therefore, universities need to demonstrate their success by thriving in a competitive environment just like any other entity. The healthy competition argument ignores the many ways competition could be destructive. First, education and research generate positive externalities or benefits that cannot be fully captured through market competition. As a result, if solely competitive forces determined

90 Research handbook on intellectual property and technology transfer outcomes, then too little education would be provided, and too little research would result. Second, the creation of universities requires large initial, or fixed, costs. These fixed costs require some degree of scale of production in the marketplace for entities to be profitable. Competitive forces acting alone can make it difficult, if not impossible, for entities to generate scale in production. As a result, competitive forces may tend to drive out most universities which fail to reach a size that would be more conducive to success. While scale effects and externalities may arise in many industries, the two together make competitive forces unhealthy and even unworkable in producing viable, socially desirable universities. In addition, there are potential problems of moral hazard and adverse selection in university markets. For example, education requires initial investments by students who defer present compensation for future earnings. Universities capitalize on these investments by accepting tuition payments currently with the expectation of returns to the training and education presented in the classroom. But once universities receive the tuition payments, there is an incentive not to fulfill the promise of training, especially if the educational inputs are uncertain or hard to completely measure. Therefore, universities might shirk in providing training (moral hazard), and bad universities might drive out good ones (adverse selection). What these limitations show is that universities may be subject to regulation in order to deal with the infirmities of market transactions. This conclusion does not deny the validity of the pure commercialization model. It simply states universities would need to be regulated in similar ways as other entities that produce positive externalities, incur high fixed costs, and are subject to moral hazard and adverse selection. The form of those regulations would depend upon characteristics of the industry, with particular attention to the geographic scope of the market (regional versus national), the distribution channels in the marketplace, allocation of information among buyers and sellers, organizational forms and other factors. D.

No One Stable Model: How the Three Models Interact

In constructing the governance and regulation of universities, the specific details of social interactions among faculty, students, staff and administrators would guide how regulation is designed and what transactions are the target of oversight. In the course of assessing these social interactions, a regulator would come across the practical details of university life. Students need housing, access to books and support in the education process. Faculty need resources to pursue teaching and research and tools for governance in interactions with students and each other. Regulation comes up against these social interactions and the cultural values of education and scholarship. Universities, no matter the depth of commitment to commercialization, are political and social institutions; thus, public mindedness and civic virtue come into play for university governance. In this way, our three models may converge or at least blur. Just as material necessity leads the pure research model towards the forces of commercialization, the attention to markets and competition leads to the need for social and cultural norms that allow universities to cohere into the locus of governance and regulation. University athletics provides one example of how these three models apply to ongoing and compelling policy debates. How should athletics be regulated? Should athletics serve as a basis for commercialization or should they be seen as purely intramural? Answers will rest on how one conceives of the university. The pure research model, in its extreme form, may support skepticism of athletics, especially in the all-consuming form. But, those who favor the pure research model because of

Bayh-Dole beyond patents 91 its appeal to the human mind would recognize the need for a healthy body to nourish mental activities. Therefore, college athletics can serve to support pure research through distraction, entertainment and exercise. Furthermore, there might be a limited practical and commercial benefit from investment in athletics. So, even within the pure research university, athletics can flourish but as a secondary venture, one that must yield when research is threatened. For the university engaging in commercialization that is public minded, athletics has a role not only as a source of a revenue, but as a basis for public engagement in competition and team spiritedness.50 However, university commitment to athletics may come at the expense of the public interest. Town-gown relations may lead to tensions between privileged athletes and targets of abuse in the local community. Within the university campus, the special status of college athlete may create divisions among students and cause rifts between students and faculty. Furthermore, an overemphasis on athletics may undermine public values of education and research because attentions may be distracted from the classroom to the gridiron. Athletics may be the lodestone for the purely commercially minded university. Merchandising, television rights, ticket sales—each serves as just one source of many for the generation of revenue. The returns for investing in athletics can in turn finance research and educational efforts, at least for the successful large-scale universities. But the market for college athletics will undoubtedly need regulation as students can be the victims of exploitation, and competition over intangible reputation and prizes can distort incentives in a winner-take-all market. Some of these regulations will overlap with the concerns raised in the pure research model and in the commercialization with public purpose model. However, if we allow universities to wholly focus on commercialization, to operate like any other firm, then a heavy focus on athletics may require more regulation to address within university and across university battles over revenue streams and IP rights. The example of athletics demonstrates how the three models would apply to the issue of IP doctrine and the question posed in the title of this Chapter. The pure research and commercialization with public purpose models would support special rules for universities to potentially limit the adverse roles of patent and copyright for the goals of universities. A pure commercialization model might support a more expansive role for copyright and patent. But, as the discussion in this section also shows, these models may have overlapping application. Consequently, actual rules for universities within copyright and patent illustrate different applications of these models, emphasizing some goals and downplaying others, depending upon the context.

50

See, e.g., Charles T. Clotfelter, Big Time Athletics in American Universities 95–6 (2011).

5.

University as knowledge-based enterprise: organizational design and technology transfer Jarrett B Warshaw

I.

INTRODUCTION

In recent years, the concept of design has suggested to advocates a compelling framework for reorganizing scientific research and education at colleges and universities. The value of design-based perspectives and strategies, adopters of the approach report, lies in directing attention to the reflexive relationship between academic structure and the creation and exchange of knowledge.1 To compete in science for resources and prestige, many research universities in the US and worldwide feature evolving continua of STEM-oriented units (“SOUs”). These units are comprised of centers and institutes, schools, and departments; and central administrators, academic leaders (provosts, deans, directors, chairs) and faculty can, through coordinated or semi-autonomous actions, form and design newer SOUs to link to each other and to older units throughout the enterprise.2 Such orthogonal, crosscutting linkages form a matrix to enhance the flow of personnel, funding, and research across a university and in collaboration with governmental laboratories, industry, other universities, and intermediary, non-profit organizations.3 The configuration of SOUs, by design, could thus facilitate the advancement of knowledge of economic and social relevance. There are a number of external influences that undergird the increasing focus of university stakeholders on redesigning academic structure. Since the 1980s in the US, the federal government has passed a portfolio of policies that leverage privatization, deregulation, and market competition to build competitive advantage in the global knowledge economy. Specific policies include streamlining the process of patenting federally-funded research via the Bayh-Dole Act, loosening anti-trust regulations, and broadening the legal scope of what can and should be patented.4 To address “grand challenges” pertaining to areas such as global health, the envi-

1

Michael M. Crow & William B. Dabars, Designing the New American University (2015). Historically, scientific research in countries such as the Netherlands, Germany, and Japan took place outside of universities in independent and government-supported institutes. Over the past decade, in response to national and regional economic competitiveness campaigns in science and technology, many once-external research organizations have become folded within university organizational structures and operations. See National Innovation and the Academic Research Enterprise: Public Policy in Global Perspective (David D. Dill & Frans A. Van Vught eds., 2010). The US features several independent research institutes (see Paula Stephan, How Economics Shapes Science [2012]), but its research universities have long housed centers and institutes to support knowledge-production and -exchange. 3 Roger L. Geiger & Creso M. Sá, Tapping the Riches of Science: Universities and the Promise of Economic Growth (2008). 4 Jacob H. Rooksby, The Branding of the American Mind: How Universities Capture, Manage, and Monetize Intellectual Property and Why It Matters (2016). 2

92

University as knowledge-based enterprise 93 ronment, and alternative energy,5 the federal mission agencies, most prominently the National Science Foundation (“NSF”), National Institutes of Health (“NIH”), Department of Defense (“DOD”), and Department of Energy (“DOE”), increasingly offer strong, selected funding opportunities for large-scale, multi-sector, interdisciplinary, and use-inspired research. Many states have over this period of time aggressively pursued science and technology-based regional economic development, subsidizing joint university-industry research and development (“R&D”) activities and centers.6 The scientific community, championing the innovative possibilities of interdisciplinary research, has advocated for flexible, nimble university designs analogous to those used in industry and national laboratories to engender bold, novel approaches to scientific problems.7 And the competitive dynamics in industry have shifted over the years as well, in which corporations are pursuing open innovation strategies and networking with other firms, universities, and faculty to generate and share new knowledge. Forces within the scientific profession reshape the organization of science as well. One source of change entails the increasing assumption that team-based collaboration is “state of the art” for conducting and advancing knowledge and scientific careers.8 Of course, the production of science in general and the management of laboratories in particular has long-entailed using “quasi-firms” of faculty, postdoctoral researchers, staff, and graduate and undergraduate students over whom principal investigators preside.9 Yet the newer generation of STEM faculty, Bozeman and Youtie (2017) observe, are socialized and trained in university settings that encourage collaboration and thus contribute to the shift in the scientific profession away from the lone-investigator model. A second source of change, emanating from the scientific profession, includes escalating competition among scientists and their teams across and within universities for resources and status.10 The jockeying for position—on campus, in science, and in the political economy of research funding—is reflected in newer unit-level and professional strategies of faculty in these settings.11 As such a perspective suggests, the interdependencies within and across SOUs strengthen because of competition rather than the technical demands of science per se. 5

Cristopher H. Hayter, “The Grand Challenge Model of R&D” in The Oxford Handbook of Local Competitiveness 237 (David B. Audretsch, Albert N Link, & Mary Lindenstein Walshok, eds., 2015). 6 James C. Hearn, T. Austin Lacy, & Jarrett B. Warshaw, State Research and Development Tax Credits: The Historical Emergence of a Distinctive Economic Policy Instrument, 28 eCon. developMent Q. 166 (2014); Jarrett B. Warshaw & James C. Hearn, Leveraging University Research to Serve Economic Development: An Analysis of Policy Dynamics in and across Three U.S. States, 36 J. hIgher eduC. pol’y and MgMt 196 (2014). 7 National Academies, Committee on Facilitating Interdisciplinary Research & Committee on Science, Engineering, and Public Policy (2005). 8 Barry Bozeman, Daniel Fay, & Catherine P. Slade, Research Collaboration in Universities and Academic Entrepreneurship: The State-of-the-art, 38 J. teCh. transfer 1 (2012); Barry Bozeman & Jan Youtie, The Strength in Numbers: The New Science of Team Science (2017). 9 Henry Etzkowitz, Research Groups as “Quasi-firms”: The Invention of the Entrepreneurial University, 32 res. pol’y 109 (2003). 10 Brendan Cantwell, Laboratory Management, Academic Production, and the Building Blocks of Academic Capitalism, 70 hIgher eduCatIon 502 (2015); Sheila Slaughter & Larry Leslie, Academic Capitalism: Politics, Policies, and the Entrepreneurial University (1997). 11 Jarrett B. Warshaw, Structuring to Advance Science: STEM-centered Organizational Innovations in the Research University (2016) (unpublished PhD. dissertation, University of Georgia) [hereinafter Structuring to Advance Science]; Jarrett B. Warshaw, “Authentic” Faculty and Academic Capitalism: The Scientist as Administrative Leader, Paper presented at the annual meeting of the American

94 Research handbook on intellectual property and technology transfer As Slaughter and Leslie (1997) observe: “Competition for resources requires that [faculty] collaborate with each other, that they capitalize on each other’s strengths while holding each other accountable for the production and quality efforts that are essential to the collective well-being.”12 The design of SOUs, as individual units and as linked sets of structures, unfolds in part from political processes and strategic action. With these considerations in mind, this Chapter offers a review of research and theory on STEM-focused organizational change in the academic structure of research universities. When university stakeholders consider design-based strategies by which to reorganize science on their campuses and to enhance technology transfer, they may seek information about the broader outcomes and implications associated with these initiatives. There are knowledge-gaps, though, due to the growth of multiple literature streams and theoretical bases on the topic, making it difficult to view reform-efforts, dynamics of change, and attendant outcomes “on the whole.” Additionally, there are relatively few empirical studies on design in the context of higher education. This Chapter offers an initial step toward an integrative perspective on the topic. Its contributions thus lie in: synthesizing the relevant science policy and higher education literatures;13 advancing conceptual and practical understandings of the changes taking place at research universities; and shedding light on emerging lines of research to inform policy and practice in this arena. Because the Chapter focuses on SOUs as examples of the prospects and limitations of design-related change in higher education, it precludes discussion of other important reforms. Many campuses have developed research parks,14 incubators,15 proof-of-concept centers,16 technology transfer offices (“TTOs”),17 and other units (e.g., economic development offices).18 In concert with these initiatives, a number of institutions have stratified academic departments based on their perceived value in producing intellectual property (“IP”) that can be profitable in the marketplace. A campus-level political economy, researchers and analysts suggest,

Educational Research Association, New York, NY (Apr. 2018) [hereinafter “Authentic” Faculty and Academic Capitalism]. 12 Slaughter & Leslie, supra note 10, at 221. 13 As Slaughter (2014) notes, science policy literature and theory concentrates on STEM fields and disciplines, and the connection of academic science to industry and economic development, but in relation to postsecondary research it does not necessarily address higher education as a whole or the range of actors, whether organized or not, within colleges and universities. Meanwhile, the higher education literature itself is comprised of numerous strands and sources of influence, including economics, sociology, psychology, and political science. Potential exists, then, to bring together the science policy and postsecondary streams of literature to inform a more general understanding of STEM-oriented organizational design at research universities. See Sheila Slaughter, “Retheorizing Academic Capitalism: Actors, Mechanisms, Fields, and Networks” in Academic Capitalism in the Age of Globalization 10 (Brendan Cantwell & Ilkka Kauppinen eds., 2014). 14 Michael Luger & Harvey A. Goldstein, Technology in the Garden: Research Parks and Regional Economic Development (1991). 15 Cristos Kolympiris & Peter G. Klein, The Effects of Academic Incubators on University Innovation, 11 strategIC entrepreneurshIp J. 145 (2017). 16 Cristopher Hayter & Albert Link, On the Economic Impact of University Proof of Concept Centers, 40 J. of teCh. transfer 178 (2015). 17 Irwin Feller, “Technology Transfers from Universities” in 12 Higher Educ. Handbook Of Theory And Res. 1 (John Smart ed., 1997). 18 Sheila Slaughter & Gary Rhoades, Academic Capitalism and the New Economy: Markets, State, and Higher Education (2004).

University as knowledge-based enterprise 95 privileges STEM fields and disciplines over the arts, humanities, and social sciences.19 Such reforms and their consequences merit research and policy attention. The Chapter is structured as follows: sections II through IV present an overview of the concept of design and addresses, by way of literature and theory on postsecondary organization, prospects and limitations about its application to research universities. Sections V through VIII discuss SOUs and their connection to technology transfer, as well as their dynamics of collaboration and competition within matrix forms of academic organization. Section IX highlights an emerging line of research on 501(c)(3) non-profit foundations in higher education and their relationship to strategies in university competitions in science. By way of conclusion section X pushes toward broader implications for institutional and public policy. Before proceeding, it may be helpful to clarify the assumptions that inform this Chapter. Specifically, the Chapter assumes a broad conceptualization of “knowledge” as codified and tacit (not yet formalized as IP or easily expressed or communicated) and as encompassing research and educational activities in STEM fields and disciplines.20 The Chapter also assumes a dynamic view of technology transfer,21 which includes an array of outcomes (patents, licenses, spin-off firms, start-ups, etc.) and a broad range, type, and less-linear, unstructured flow of knowledge-exchanges (faculty consulting, employment of graduates, funding, serendipity, and informal interactions between faculty and industry). Importantly, the concept of design, as articulated here, offers a framework for thinking about and approaching technology transfer-relevant organizational change; it is not a panacea or proven “best practice.” As design-based strategies gain the attention of university stakeholders, and as they can and do inform reorganizations of science, they warrant close, critical attention.22

II.

THEORETICAL PERSPECTIVES

The concept of design has already permeated higher education and reached the general public.23 “Design” has come to constitute a buzzword that is divorced from its theoretical foundations, making a discussion of those foundations important for clarifying its prospects and limitations for university stakeholders. As “design” is increasingly promoted through undergraduate and graduate programs such as Stanford University’s “d.school” and Boise State

19 Kelly Ochs Rosinger, Barrett J. Taylor, Lindsay Coco, & Sheila Slaughter, Organizational Segmentation and the Prestige Economy: Deprofessionalization in High- and Low-resource Departments, 87 J. of hIgher eduC. 27 (2016). 20 A third aspect of university mission centers on service and community engagement; though service is not covered in this Chapter, for a helpful discussion, see Thomas Gais & David Wright, “The Diversity of University Economic Development Activities and Issues of Impact Measurement” in Universities and Colleges as Economic Drivers: Measuring Higher Education’s Role in Economic Development 31 (Jason E. Lane & D. Bruce Johnstone, eds., 2012) (linking and measuring university service activities in relation to economic development). 21 Samantha R. Bradley, Cristopher S. Hayter, & Albert N. Link, Models and Methods of University Technology Transfer, 9 foundatIons and trends In entrepreneurshIp 571 (2013). 22 See Jeffrey R. Young, MIT Dean Takes Leave to Start New University Without Lectures or Classrooms, Chron. of hIgher eduC., available at http://chronicle.com/article/MIT-Dean-Takes -Leave-to-Start/235121 (last visited Feb. 1, 2016) (discussing innovative designs for teaching and educational activities). 23 Bill Burnett & Dave Evans, Designing Your life: How to Build a Well-lived, Joyful Life (2016).

96 Research handbook on intellectual property and technology transfer University’s College of Innovation and Design, the concept is likely to gain more legitimacy and spread further via professional associations, dissemination of research, executive training and certification programs, and alumni networks. Advocates suggest a wide range of applications of design to solve problems of performance, because the concept offers a general (and, in their view, generalizable) framework by which to infuse creativity and innovative change across organizational settings and in our daily lives. The pushback, among critics, centers on its fad-like qualities and characteristics.24 Specifically, “design” is critiqued for its ill-defined and jargon-like nature, for its diffusion from industry to other sectors, and for the forceful rush to apply the seemingly simple and straightforward solution to complex organizations.25 This section offers theoretical perspectives on design and postsecondary organization, aiming to reconcile design-based change within the university context.

III.

CONCEPTUALIZING DESIGN

Crow and Dabars (2015) offer a helpful definition of organizational design as they encourage institutional stakeholders to apply the concept to reconfigure the academic enterprise. As they observe: Institutional design, in our usage, refers broadly both to the process of design and its product, the organization of a knowledge enterprise and the social formations and knowledge networks its configuration engenders. The dynamics of this relationship in the American research university may appear at first glance to be a perfunctory administrative consideration … But any institutional platform constructed to support the growth of knowledge is only the product of a sequence of decisions that determine its structure and functions, which may over time require calibration or reconfiguration.26

Suggesting a structural-bureaucratic perspective of organizations,27 their definition focuses on how context shapes and is itself entwined with knowledge-production and -exchange. The concept is valuable, in their view, because it attends to the formal arrangement of academic units throughout a research university, which stakeholders can amend. As such, by leveraging changes in that structure to expand and deepen interactions and communication, research and educational activities could be optimized of benefit to the economy and public good. Despite its assumption of top-down managerially driven change, the Crow and Dabars (2015) definition nevertheless offers a flexible means of analysis.28 For example, organizational design can take place daily at the level of industrial, university, and federal laboratories. As Tucker and Sampat (1998) observe, design-driven change occurs when industrial labs “decide to link their central R&D facility to their production unit, university labs consolidate labs in biology with those in chemistry, government labs change the boxes and lines of

24

See R. Birnbaum, Management Fads in Higher Education: Where They Come From, What They Do, Why They Fail (2000). 25 Lee Vinsel, Design Thinking is a Boondoggle, Chron. of hIgher eduC., available at https:// www.chronicle.com/article/Design-Thinking-Is-a/243472 (last visited May 25, 2018). 26 Crow & Dabars, supra note 1, at 179. 27 Max Weber, “Bureaucracy” in From Max Weber: Essays in Soc. 196 (1946). 28 Crow & Dabars, supra note 1, at 179.

University as knowledge-based enterprise 97 programmatic organization charts.”29 They also suggest the utilization of design-based concepts to retool the national innovation system of R&D laboratories throughout the US. Thus design-based applications can be infused within and between units at various levels and layers of a single organization and be deployed to reshape a constellation of organizations within a broader system. Conceptualizations of design are spun, most centrally in Crow and Dabars’ (2015) usage, from knowledge-based theories of the firm and structuration theory. From an economic perspective, knowledge-based theories of the firm posit that a corporation’s greatest asset and competitive advantage is its ability to create, transfer, integrate, and exploit knowledge internally and to protect that knowledge from external competitors.30 As Bolton and Dewatripont (1994) explain, such a conceptualization treats the “internal organization of firms … as a communication network that is designed to minimize both the costs of processing new information and the costs of communicating this information among its agents.”31 Their analytical point brings helpful attention to efforts of firms to minimize transaction costs associated with generating, communicating, and acting upon knowledge, and it suggests an additional need to attend to strategies to maximize value rather than solely to reduce costs and other barriers. The firm as a whole, then, can be perceived as an “epistemic community” of various experts and specialists whose organizational configurations can be calibrated to capitalize on, combine, and integrate tacit and codified forms of knowledge.32 Structuration theory expands understandings of design by offering a sociological perspective on the relationship between agency of individuals and groups and social systems.33 Within the context of neo-institutional theories of compliance to powerful norms, “rules,” and reward systems,34 structuration theory offers a nuanced account for how and why organizational settings and fields come to stabilize over time.35 It suggests that the context in which individuals and groups are situated shapes the extent to which they perceive and direct cognitive attention to alternative possibilities, methods, approaches, and forms of work and organization. To be clear, the theory does not posit that context controls the range or expansiveness of human thought and practice, but it highlights the relationship between the structure of organizations

29

Christopher Tucker & Bhaven Sampat, “Laboratory-based Innovation in the American National Innovation System” in Limited by Design: R&D Laboratories in the U.S. National Innovation System 41 (Michael Crow & Barry Bozeman, eds., 1998). 30 Julia Porter Liebeskind, Knowledge, Strategy, and the Theory of the Firm, 17 strategIC MgMt J. 93 (1996); D.J. Teece, “Knowledge and Competence as Strategic Assets” in Handbook of Knowledge Management 129 (Clyde Holsapple, ed., 2003). For analyses of research universities seeking to guard IP and patents via litigation, see Rooksby, supra note 4; Jacob H. Rooksby, University Initiation of Patent Infringement Litigation, 10 J. Marshall rev. of Ip l. 624 (2011); Jacob H. Rooksby & Brian Pusser, “Learning to Litigate: University Patents in the Knowledge Economy” in Academic Capitalism in the Age of Globalization 74 (Brendan Cantwell & Iikka Kauppinen eds., 2014). 31 Patrick Bolton & Mathias Dewatripont, The Firm as a Communication Network, 109 Q. J. of eCon. 809 (1994). 32 Lars Håkanson, The Firm as an Epistemic Community: The Knowledge-based View Revisited, 19 Indus. and Corp. Change 1801 (2010). 33 Anthony Giddens, The Constitution of Society: Outline of the Theory of Structuration (1984). 34 Paul J. DiMaggio & Walter W. Powell, The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality In Organizational Fields, 48 aM. soC. rev. 147 (1983). 35 For a discussion on field destabilization see Neil Fligstein & Doug McAdam, Theory of Fields (2012).

98 Research handbook on intellectual property and technology transfer and fields and the relative degree of innovation and change in those settings. A key consideration for leaders and managers of organizations focuses on how to configure and coordinate a knowledge-based enterprise to push toward bold, novel forms, types, and combinations of knowledge-production. A conceptualization of design may further benefit from problematizing the delineation of where organizational boundaries begin and end. Theory and research on “invisible colleges” and external epistemic communities of faculty and their disciplinary colleagues suggests that an inwardly focused, within-institution perspective may be naïve and shortsighted.36 In the context of higher education, individual organizational units and faculty boundary-span.37 They bring to the university external interests, values, goals, and resources, connect a variety of organizations and collaborators closer together, divide accountability among numerous sources/resource providers, and exchange knowledge and resources.38 To facilitate matrix forms of organization constitutes a central goal of applying design-based strategies. The formulation and usage of matrix designs has roots in industry and government. As noted in the Introduction, a matrix entails purposeful orthogonal, crosscutting linkages and coordination among different units and operational domains within an organization.39 A manager of product X, for instance, may oversee a team comprised of personnel that features representatives from separate units such as R&D, manufacturing, marketing, and finance. In the 1970s, the matrix approach spread across industry sectors as prominent corporations—notably General Electric, Bechtel, Citibank, Dow Chemical, Shell Oil, and Texas Instruments—used it and to what appeared to be lucrative success.40 To its advocates, the concept of small, fluid structures in which employees are organized around a specific technical problem, product, or area of strategic emphasis is deemed a crucial component of the long-term vitality of the firm as a whole.41 While the organization of operations and people around selected products and projects brought flexibility, fluidity, and nimbleness to these settings, use of matrix-based designs was all but abandoned nearly a decade later. Whether accurate in their assessment or not, executives blamed the matrix for declines in organizational performance. What is more, the matrix had engendered problems that went unresolved. These included confusion in reporting relationships (some personnel reported to two different bosses), competition and “turf” disputes among the managers of the different

36

Crow & Dabars, supra note 1; Sheila Slaughter & James C. Hearn, Final Report for Centers, Universities, and the Scientific Innovation Ecology: A workshop. NSF Grant BCS-0907827 (2009). 37 Jason Owen-Smith, Research Universities and the Public Good: Discovery for an Uncertain Future (2018); Jan Youtie & Philip Shapira, Building an Innovation Hub: A Case Study of the Transformation of University Roles in Regional Technological and Economic Development, 37 res. pol’y 1188 (2008). 38 Creso M. Sá. & Anatoyl Oleksiyenko, Between the Local and the Global: Organized Research Units and International Collaborations in the Health Sciences, 62 hIgher eduC 367 (2011). 39 Jay Gailbraith, Matrix Organization Designs: How to Combine Functional and Project Forms, 15 Bus. Horizons 29 (1971); Jay Gailbraith, “Organization Design: An Information Processing View” in Org. plan.: Cases and concepts 49 (Jay W. Lorsch & Paul R. Lawrence, eds., 1972); Erik W. Larson & David H. Gobeli, Matrix Management: Contradictions and Insights, 29 Cal. MgMt rev. 126 (1987); see also Geiger & Sá, supra note 3. 40 Stanley Davis & Paul Lawrence, Problems of Matrix Organizations, harv. bus. rev (May 1978), available at https://hbr.org/1978/05/problems-of-matrix-organizations (last visited Oct. 17, 2019). 41 Tom Peters & Robert H. Waterman, In Search of Excellence (1982); Slaughter & Leslie, supra note 10, at 232.

University as knowledge-based enterprise 99 functional areas, and lack of consensus on goals and timelines/indicators of when to conclude and disband. Using the concept of design to form a matrix, advocates suggest, holds great potential for research universities. Industry and federal laboratories have employed it, mobilizing interdisciplinary teams of scientists to swarm around a selected topic or scientific problem. Structurally, it presents a core way in which university leaders aim to spur additional interdisciplinary research on campus that may not have occurred otherwise.42 Centers and institutes, through the design-based lens, can be instrumental in bringing together other departments, faculty, students, and resources throughout the enterprise. The use of incentive grants and cluster hires and joint-appointments may strengthen the crosscutting ties among these units. Despite its prospects for constructive change to knowledge-production and -exchange on campuses, the concept of design—and its relationship to forming a matrix of SOUs—needs to be reconciled with dynamics of postsecondary organization.

IV.

EXTENDING THE CONCEPT OF DESIGN TO RESEARCH UNIVERSITIES

As applied to higher education, the concept of design is typically used to inform decision-making about the restructuring of academic departments. Such efforts focus on facilitating interdisciplinary research and education, for emerging fields and disciplines connect to the development of potentially lucrative and transformative industries and discoveries for the economy and human health.43 These initiatives often entail substantive structural and governance reforms that privilege administrative-driven, short-term solutions (creating new centers and institutes) rather than faculty-led, long-term initiatives within academic departments. Jacobs (2014) raises strong warnings about the erosion of disciplinary departments and swings in the power dynamics and social relations on campuses, but he captures well a core concern among advocates of interdisciplinarity: “Interdiscipilnarians often rail against the ways that disciplines undermine intellectual freedom. The heartfelt complaint is that a relatively small number of intellectual figures at leading institutions set the agenda for researchers throughout the discipline, thus unduly constraining creativity and scholarly advances.”44 At issue is the institutionalized correlation between departments and disciplinary fields that critics view as stifling innovative advancements in the production of knowledge.45 Concepts of postsecondary organization offer nuance and texture to applying design-based change to research universities. For example, the concept of design can be read as advocating top-down, directive forms of administrative leadership and decision-making in academic organizations. Such an assumption at best reduces, and at worst ignores, the relative impor-

42 Creso M. Sá, “Interdisciplinary Strategies” in U.S. Research Universities, 55 hIgher eduC. 537 (2008). 43 Stephan, supra note 2. 44 Jerry Jacobs, In Defense of Disciplines: Interdisciplinarity and Specialization in the Research University 215 (2014). 45 National Academies, supra note 7.

100 Research handbook on intellectual property and technology transfer tance and benefits of loose coupling, which is defined as infrequent, indirect interactions and communication among units within an organization.46 Through the lens of loose coupling, colleges and universities are viewed as disjointed, “bottom-heavy” types of organizations. These institutions have evolved into “professional bureaucracies”47 in which faculty and academic departments, rather than administrative leaders per se, are responsible for carrying out the day-to-day work of research, teaching, and service.48 The specialized training of faculty, who are disciplinary experts and members of the profession that trains all other professions, endows them and their departments with latitude. They pursue semi-autonomous actions independent of each other and, often, of central administration and thus demonstrate a form of loose coupling.49 When the infrequent, indirect interactions and communication continues over time, a campus can become a loosely coupled system. To be sure, such organizational settings have notable disadvantages. These include programmatic duplication, cost inefficiencies, and vulnerabilities to extreme, emergency or crisis situations (e.g., Hurricane Katrina) that necessitate a coordinated campus-wide action plan to ensure safety. But there are distinctive advantages vis-à-vis loose coupling and its connection to encouraging organizational change. Individual departments can pursue changes that are often substantive, creative, and entrepreneurial, but, because of loose coupling, might be less visible to central administrators and external observers.50 What is more, these changes can be implemented rapidly or unfold over the course of many years due to politics and competition within disciplines.51 Stakeholders seeking evidence of change over the short-term might assume that adaptation is not occurring quickly enough or not occurring at all and turn to radical or transformational approaches that cause more harm than positive, marginal gains. Another core benefit of loose coupling entails buffering. As academic departments pursue change in relation to their disciplines and fields and to their particular external environments, and as some of their changes are maladaptive, other departments and the institution as a whole remain largely intact. Meanwhile, individual departments can receive benefits from other departments’ changes and reforms through “halo” effects and cross-subsidies that college deans and/or central administrators redistribute. There are additional forms and types of loose coupling in higher education to consider. The goals of colleges and universities, of teaching, research, and service, are ambiguous and elusive to define, and they may be loosely coupled or in competition with one another (e.g., the pursuit of research over teaching and service). These institutions tend to feature unclear technology, with tenuous connections among their core inputs (personnel, students, resources), processes

46

Karl E. Weick, Educational Organizations as Loosely Coupled Systems, 21 adMIn. sCI. Q. 19 (1976). 47 Henry Mintzberg, The Structuring Of Organizations: A Synthesis of the Research (1979). 48 A. Etzioni, Modern Organizations (1964); Clark, B.R., Contradictions of Change in Academic Systems, 12 hIgher eduC. 101 (1983). 49 Central administrators can and often do influence and shape the strategies, work, and resources of academic departments. In the context of loose coupling, presidents, for instance, may not have formal control over or coordination of the academic enterprise, but may nevertheless influence it vis-à-vis setting the broad premise or vision for change, making strategic hires across operational domains, and recalibrating resource allocations, incentives, and reward systems throughout the enterprise. Michael D. Cohen & James G. March, Leadership and Ambiguity: The American College President (2d ed., 1986). 50 Clark Kerr, The Uses of the University (4th ed., 1995). 51 Jacobs, supra note 44; Thomas S. Kuhn, The Structure of Scientific Revolutions (4th ed., 2012).

University as knowledge-based enterprise 101 (research, teaching, and service), and outputs and outcomes. Such characteristics stand in sharp contrast to the private sector in which scientific management, discerning the very best approaches and combinations of activities to produce optimal results, is more likely to inform daily operations and decision-making than in colleges and universities. Additionally, many postsecondary institutions face strong degrees of uncertainty in their external environments (especially in relation to resources), and to deflect outside scrutiny of its internal operations, gaining managerial discretion, they maintain the appearance through organizational structure of legitimacy and thus loosely couple themselves from external audiences.52 That legitimacy, as supported through looking like a college or university should in its structure, helps these institutions uphold their charters with society as a whole.53 Loose coupling shapes institutional decision-making and in ways that resemble a “garbage-can” process of choice.54 Colleges and universities reflect characteristics of “organized anarchies,” as they have ambiguous and conflicting goals (e.g., research, teaching, and service), feature unclear technology (e.g., inputs are loosely coupled to processes are loosely coupled to outputs/outcomes), and face strong uncertainty in their external environments (especially in relation to funding). When university leaders, for example, aim to engage stakeholders in decisions, they encounter a process that is often non-linear and somewhat unstructured. They encounter, that is, a distinctive form of bounded-rationality that limits the extent to which they optimize decision-making from issue to issue.55 In any given decision-making opportunity, the garbage-can model suggests, there are three streams that emerge and interact. The first entails the problem stream, encompassing a variety of symptoms and issues that vie for attention and that are posed as the main problem. A second stream is comprised of the solutions that are put forth, which may or may not connect to or address the problems or issues as defined. The third stream consists of fluid member participation, whereby stakeholders “float” in and out of various decision-making opportunities. Within such a context, decisions can be made by flight (waiting until the garbage detaches itself from the problem at hand), by oversight (committing to solutions before problems attract garbage), or by resolution (using a “rational” model in which stakeholders consider all available information, identify the problem, and find the optimal choice to best address the issue). As colleges and universities pursue entrepreneurial directions and organizational forms, they could increasingly become linear, top-down, and corporatized in their decision-making. Clark (1998) observes that entrepreneurial universities typically feature the expansion of administrative capacity and steering throughout the enterprise in order to manage and oversee directly market-oriented goals, activities, and resources.56 Administrative leaders and professional staff

52

John W. Meyer & Brian Rowan, Institutionalized Organizations: Formal Structure as Myth And Ceremony, 83 aM. J. of soC. 340 (1977). 53 David H. Kamens, The College “Charter” and College Size: Effects on Occupational Choice and College Attrition, 44 soC. of eduC. 270 (1971); David H. Kamens, Legitimating Myths and Educational Organization: The Relationship Between Organizational Ideology and Formal Structure, 42 aM. soC. rev. 208 (1977). 54 Cohen & March, supra note 49. 55 Herbert A. Simon, Administrative Behavior: A Study of Decision-making Processes in Administrative Organizations (4th ed., 1997). 56 Burton R. Clark, Creating entrepreneurial universities: Organizational pathways of transformation (1998).

102 Research handbook on intellectual property and technology transfer can and often do make constructive contributions to their campuses,57 but as they constitute an increasing proportion of all academic labor, and as full-time tenure-line and tenured faculty represent less than half of all current academic appointments, they could sideline faculty engagement in institutional decision-making.58 Contrary to calls for radical or transformational change to higher education, researchers and analysts suggest that much change and innovation has occurred in the context of loose coupling, garbage-can decision-making, and what critics deem to be antiquated faculty tenure and governance processes.59 Certainly these institutions and their departments may benefit from becoming especially nimble and adaptive in emerging contexts. A key implication, then, concerns the assumptions informing the use of design-oriented strategies and actions. The application of the concept of design has constructive potential for research universities and in the STEM arena, but requires reconciliation with fluid, less-structured forms of change and decision-making in higher education. Cameron (1984) offers a widely cited definition of organizational adaptation as “modifications and alterations in the organization or its components in order to adjust to changes in the external environment.”60 It suggests organization-wide change to restore equilibrium between a campus and its external environment. Importantly, such a definition comes with flexibility. Organizational adaptation can be proactive or responsive, which differentiates it from other types of change (planned change or organizational development) whereby the impetus for change originates within the organization and is deliberate but not externally connected. The concept of design, at heart, offers a framework that could be piloted in selected enclaves or segments of organizations and scaled to fit specific institutional contexts. To the extent that its deployment enhances dynamic, contingent-based forms of strategy and planning,61 the concept of design may be helpful to institutional stakeholders seeking to strengthen the connection of academic structure to technology transfer.

V.

ACADEMIC STRUCTURE AND MODES OF PRODUCTION

The design of SOUs in their own rights, and as linked, interdependent entities, suggest adaptations to advance innovative modes of academic production. For example, the adoption of

57

Kevin R. McClure, Building The Innovative and Entrepreneurial University: An Institutional Case Study of Administrative Academic Capitalism, 87 J. of hIgher eduC. 516 (2016); Barbara Sporn & Richard Miller, Adaptive University Structures: An Analysis of Adaptation to Socioeconomic Environments of U.S. and European Universities (1999). 58 Gary Rhoades, “Extending Academic Capitalism by Foregrounding Academic Labor” in Academic Capitalism in the Age of Globalization 113 (Brendan Cantwell & ILkka Kauppinen eds., 2014). 59 Dominic J. Brewer & William G. Tierney, “Barriers To Innovation in U.S. Higher Education” in Reinventing Higher Education: The Promise of Innovation 11 (Ben Wildavsky, Andrew P. Kelly, & Kevin Carey eds., 2011). 60 Kim S. Cameron, Organizational Adaptation and Higher Education, 55 J. of hIgher eduC. 123 (1984). 61 Marvin Peterson, “Using Contextual Planning to Transform Institutions” in Planning and management for a changing environment: A handbook on redesigning postsecondary institutions 127 (Marvin Peterson, David Dill, Lisa Mets, & Associates eds., 1997); Marvin Peterson, Improvement to Emergence: An Organization-environment Research Agenda for a Postsecondary Knowledge Industry (1998).

University as knowledge-based enterprise 103 research as mission and activity was considered a revolutionary change in higher education.62 As the research economy of funding grew over time, due to emerging social demands and resource-providers,63 research universities shifted from Mode I of basic, “pure” research to Mode II of applied and translational forms of knowledge-production and -exchange.64 In their contemporary contexts, many research universities pursue both modes of knowledge production and in increasingly interdisciplinary areas, anchor regional economic development, and serve as “innovation hubs” entwined with an array of intermediary and external partners.65 The evolving forms of academic structure and production, reflecting goals of applying fluid and flexible organizational designs and approaches to carrying out the core work of research and teaching, merit attention. Building on the discussion of the concept of design, this section focuses on the origins, structures, functions, finances, and outcomes associated with SOUs. These centers and institutes, schools, and departments serve as examples of STEM-oriented organizational changes taking place at research universities. To consider their formation and development may illuminate the prospects and limitations of using design-based strategies in this arena.

VI.

CENTERS AND INSTITUTES

In the US, the 19th century colleges and universities had developed the earliest types of centers and institutes in higher education at that time. These included observatories and museums to store equipment and materials to advance teaching and learning not otherwise accommodated in the core academic units on campus.66 Because nearly all of America’s earliest colleges and universities were founded as instructionally-focused institutions,67 whose mission was to disseminate what was already known rather than to advance new knowledge, they deployed the center and institute model to expand the research function. Exceptions, however, were Clark University, University of Chicago, and Johns Hopkins University, which were, by 62

Christophr Jencks & David Riesman, The Academic Revolution (1968). Roger Geiger, Organized Research Units—Their Role in the Development of University Research, 61 J. of hIgher eduC. 1 (1990) [hereinafter Organized Research Units—Their Role in the Development of University Research]; Roger Geiger, Research and Relevant Knowledge: American Research Universities Since World War II (1993). 64 Michael Gibbons, Camille Limoges, Helga Nowotny, Simon Schwartman, Peter Scott, & Martin Trow, The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies (1994). 65 See Owen-Smith, supra note 37; Youtie & Shapira, supra note 37; but see Maryann Feldman & Pierre Desrochers, Research Universities and Local Economic Development: Lessons from the History of Johns Hopkins University, 10 Indus. and InnovatIon 5 (2003). 66 See Organized Research Units—Their Role in the Development of University Research, supra note 63; Geiger, supra note 63; Patricia Gumport, “The Federal Role in American Graduate Education” in Higher Education: Handbook of Theory and Research, Vol. VII 102 (John Smart ed., 1991) [hereinafter Federal Role in American Graduate Education]; Patricia Gumport, “Graduate Education and Research: Interdependence and Strain” in American higher education in the twenty-first century: Social, political, and economic challenges 110 (Michael Bastedo, Phillip Altbach, & Patricia Gumport eds., 4th ed., 2016) [hereinafter Graduate Education and Research: Interdependence and Strain]. 67 Frederick Rudolph, The American College and University: A History (2nd ed., 1990); Laurence Veysey, The Emergence of the American University (1965); John Thelin, A History of America Higher Education (2nd ed., 2011). 63

104 Research handbook on intellectual property and technology transfer design, created to serve research and graduate education. Buttressed by a growing research economy of funding from industry, private foundations, and, though to a lesser extent, from state and federal governments, the leading universities of the time formed organized research units (“ORUs”). The surge in and formalization of federal R&D funding to higher education would occur in the post-World War II era. In the intervening years, ORUs advanced research in capital-intensive areas of science and in relation to emerging societal demands for different types, forms, and applications of specialized and abstract knowledge.68 In their early iterations, centers and institutes were typically located at the periphery of the campus administrative hierarchy. They were kept at arm’s-length from the core academic department structure because of their focus on research rather than on teaching and because of their ties to particular external funders. For faculty not involved in centers and institutes, they perceived these structures as “prostitution” of the campus vis-à-vis selling itself to industry and external interests/goals and thereby poisoning its wellspring of knowledge.69 Whether effective at meeting their scientific and funding goals or not, the number of centers and institutes in higher education has grown over time. That growth is rooted in the influx of post-World War II funding targeted to graduate education and research in higher education. Indeed, federal R&D funding has increased over time even as the rationales for that support have also changed.70 Within the context of escalating competition for extramural funding,71 estimates from the early 2000s suggest anywhere from a dozen to hundreds of ORUs at any given campus.72 Once-peripheral centers and institutes are increasingly linked to core academic structure, serving unique interstitial, boundary-spanning roles in the production and exchange of knowledge at interstices of organizations and disciplines and fields. Centers and institutes—ORUs—are organizational responses to solve problems of science and funding. A strong impetus for forming and developing centers and institutes lies in increasing institutional shares of the roughly $30 billion in federal R&D funding allocated each year to colleges and universities. Federal R&D funding constitutes the largest source of resources for research in higher education and, with exceptions of particular programmatic funding from the Department of Agriculture and contracts more generally, is allocated through a competitive, peer-review process.73 These resources are concentrated among an elite and prestigious few,74 undergirding patterns of stratification in higher education,75 but many less-competitive insti-

68 See Organized Research Units—Their Role in the Development of University Research, supra note 63; Geiger, supra note 63; Peter Rossi, Researchers, Scholars and Policy Makers: The Politics of Large Scale Research, 93 daedalus 1142 (1964). 69 Stanley Ikenberry & Renee Friedman, Beyond Academic Departments (1972). 70 Sheila Slaughter & Gary Rhoades, The Emergence of a Competitiveness Research and Development Policy Coalition and the Commercialization of Academic Science and Technology, 21 sCI., teCh., and huM. values 303 (1996). 71 Slaughter & Leslie, supra note 10; Slaughter & Rhoades, supra note 18. 72 Graduate Education and Research: Interdependence and Strain, supra note 66. 73 Stephan, supra note 2. 74 Roger Geiger, Knowledge and Money: Research Universities and the Paradox of the Marketplace (2004). 75 Barrett Taylor, “The Field Dynamics of Stratification Among US Research Universities: The Expansion of Federal Support for Academic Research, 2000–2008” in Higher Education, Stratification, and Workforce Development: Competitive Advantage in Europe, the US, and Canada 59 (Sheila Slaughter & Barrett Taylor eds., 2016).

University as knowledge-based enterprise 105 tutions still seek to compete despite the financial costs and the relatively low probabilities of success.76 Each center and institute has a distinctive trajectory of development and structure, mix of funding, source of accountability, degree of authority, and set of epistemic/technical demands. A faculty member or team of colleagues could, independent of departmental, college/school, or institutional efforts or priorities, start a center or institute. Department chairs, associate deans or deans, and Vice Presidents for Research, Provosts, and Presidents may launch centers and institutes with faculty involvement (e.g., through a competitive, proposal review process) or without faculty input during the front-end planning or designing phase. The latter could occur when, say, Vice Presidents for Research and Provosts identify an opportunity for centeror institute-level funding (from NSF) and then seed-fund and seek to organize, from across the academic landscape, a team of faculty to compete for the award. Meanwhile, states can create and fund selected centers and institutes, locating them at colleges and universities, to serve their regional economic development goals and the broader public good. And philanthropists and donors may endow centers and institutes, directors’ positions, and even faculty lines to attach permanent sources of funds to them.77 A representative example of how centers or institutes are founded and by whom is difficult to pinpoint. Nevertheless, it may be helpful to consider the Center for Materials Research (“CMR”) at Stanford University, an ORU founded in the 1960s. CMR helped to advance the field of materials science of importance to national defense during the Cold War. As historian Stuart Leslie (1993) observes, the reluctance of academic physicists to engage in defense-related research after the Manhattan Project of World War II opened a pathway to professionalization for materials scientists.78 At Stanford in 1961, amid broader efforts of faculty and laboratories to organize around funding opportunities for materials science, the then-newly forming CMR brought visibility and viability to the emerging field. Because of its usefulness to national defense, CMR contracted that year with the Department of Defense’s Advanced Research Projects Agency (“DARPA”). By 1965, the center was built on the Palo Alto campus in part from the four-year $2.6 million commitment from DARPA and $1.5 million from Jack McCullough, co-founder of the San Francisco Bay area’s oldest electronics firms Eitel-McCullough. Establishing a mix of federal, industry, and institutional support was critical, covering CMR’s $1 million annual operating budget and funding its innovative facilities for crystal synthesis and preparation, X-ray analysis, thin-film preparation, and electron microscopy, among others. The robust funding supported as well the recruitment of new faculty who were expected to compete for independent contracts to generate additional external funding.79 At Stanford, as elsewhere, federal funding was instrumental in codifying

76

States provide some targeted funding for R&D in colleges and universities, but mostly have given broad-based forms of support to subsidize institutions and, in turn, students. Of course, that broad-based form of support has decreased over time relative to the enrollment growth in public higher education, putting pressure on institutions to diversify their revenue streams through entrepreneurship, external grants and contracts, and the like. 77 See, e.g., the Simons Center for Geometry and Physics at Stony Brook University. 78 Stuart Leslie, The Cold War and American Science: The Military-Industrial-Academic Complex at MIT and Stanford (1993); See also Daniel Kevles, The Physicists: The History of a Scientific Community in Modern America (1978) (discussing the role of academic physicists in World War II). 79 The contracting of classified defense research in the era of the Cold War at Stanford University prompted backlash analogous to issues arising at MIT. Leslie (1993) notes that MIT earned the nickname

106 Research handbook on intellectual property and technology transfer new organizational structures, attracting top faculty in emerging, interdisciplinary fields, and subsidizing technology to give competitive edge and prestige to the home institution. Strikingly, centers and institutes reflect an impressive range and variety of types and forms. As noted above, CMR constitutes a “standard” center as bricks-and-mortar structure on campus with space and facilities to house it and facilitate its work. Other types of centers and institutes include “paper” units (existing in letterheads only), “adaptive” units to be assembled, dissolved, or reassembled, and “virtual” units that require minimal institutional funding, promote themselves via websites and other materials, and utilize already-existing facilities.80 Centers and institutes also feature a diverse array of formal positions within an administrative hierarchy, reporting to provosts and vice presidents, deans, associate deans, school directors, and/or chairs. But reporting relationships differ from funding and budgetary forms of support and coordination. For example, a center or institute may fall within the purview of the provost but not receive any center-level funding or subsidies from the institution or any other department on campus. The director may not even earn a pay-bump for her or his administrative role. Thus the center promotes a collaboration among colleagues on campus, reports on its progress to central administration, and does not receive any funding for operations or cost-sharing support (e.g., funding for graduate research assistants) until it captures external grants and contracts. Despite their importance to research on campus, centers and institutes face strong financial and managerial challenges. Consider the prominent NSF Engineering Research Centers (“ERCs”) and Industry/University Collaborative Research Centers (I/UCRCs). To apply and receive funding for these types of centers, university faculty submit proposals that include lists of industry partners who will participate in the projects. The partners, and any others recruited to join thereafter, pay membership fees to the centers, and the prices may be differentiated based on the levels of interaction the partners desire and the degrees of access to IP produced. Because these are NSF-sponsored centers, they come with specific, standardized parameters for ownership of IP and licensing and technology transfer agreements. The centers come with multiple phases of funding intended to wean the centers off governmental subsidies and to incentivize them to “show fruit or die” and become self-sustaining.81 Government and industry subsidizes them and other centers and institutes on campuses,82 but does not cover the full costs associated with conducting research and maintaining and upgrading facilities. As such, these centers are “fragile organizational innovations.”83 “Pentagon East” and “Pentagon on the Charles.” Leslie, supra note 78, at 235. At both Stanford and MIT, faculty and students protested the military’s reach on their campuses, outraged that military-industrial interests and money were polluting the values of educational institutions. Indeed, DOD funding led to organizational innovations in STEM at these elite universities, but, as Leslie warns, infused within them its own militaristic agenda. See id. 80 Barry Bozeman & Craig Boardman, Academic Faculty in University Research Centers: Neither Capitalism’s Slaves nor Teaching Fugitives, 84 J. of hIgher eduC. 88 (2013); Ikenberry & Friedman, supra note 69; Gerald Stahler & William Tash, Centers and Institutes in the Research University: Issues, Problems, and Prospects, 65 J. of hIgher eduC. 540 (1994). 81 Barry Bozeman & Craig Boardman, NSF Engineering Research Centers and the University— Industry Research Revolution: A Brief History Featuring an Interview with Erich Bloch, 29 J. of teCh. transfer 365 (2004). 82 Hearn, supra note 6. 83 Irwin Feller, Catherine P. Ailes, & J. David Roessner, Impacts of Research Universities on Technological Innovation in Industry: Evidence From Engineering Research Centers, 31 res. pol’y 457, 473 (2002).

University as knowledge-based enterprise 107 Because of the institutional costs associated with executing grant-funded research and maintaining and improving facilities, centers and institutes can generate helpful indirect cost recovery from all eligible grants they and their faculty receive. For example, institutions apply to each grant-dollar received a percentage mark-up for overhead (facilities, electricity, heat, etc.). The elite research universities, which have strong histories of federal R&D funding, may have indirect cost recovery rates of more than 50% and that, institutional advocates suggest, helps these campuses to upgrade infrastructure to continue to compete for the best scientists and students and vie for additional funding. To critics, such rates seem exorbitant. But as Gumport observes: “In 2010, advocates for research universities asserted that they bore enormous unreimbursed indirect costs for federally sponsored research, with national estimates of the shortfall to be between $2.5 and $4 billion.”84 Additionally, she notes that another concern centers on the increase in faculty time devoted to administering grants, with estimates showing an increase from 18% to 42% from 1990 to 2010.85 Research universities may deploy resources to build capacity in anticipated directions of funding, undertaking costly construction and capital-related projects (e.g., renovating laboratories and hospital facilities). These institutions may overbuild in relation to fluctuations and abrupt shifts in the budgets and priorities within and across mission agencies and leave those newer facilities unused and unusable.86 The financial and resource dynamics of centers and institutes, and of research more generally, shape the constraints within which newer modes of knowledge-production and -exchange may be possible.87 A related issue for campus stakeholders who are not involved in STEM-related centers and institutes is the extent to which ORUs drain resources from other areas.88 That is, once centers and institutes are created and expand in their scope, scale, and size, they need resources on which to feed.89 Nevertheless, given that centers and institutes across a campus reflect such a wide variety of types, forms, and funding arrangements, institutions could calibrate the costs and risks involved in expanding the production of research on their campuses. For instance, at the University of Florida (“UF”), state investment helped to form the Emerging Pathogens Institute (“EPI”) in 2006.90 EPI focuses on the connections between the life sciences and medicine, with an emphasis on the spread and treatment of diseases among humans, animals, and the environment. As a bricks-and-mortar and stand-alone entity, reporting directly to the provost, it was built on a previously undeveloped, rustic part of campus. The institute houses office space and its administrative apparatus, as well as laboratories for 84

Graduate Education and Research: Interdependence and Strain, supra note 66, at 135. Id. 86 Richard Harris, Built in Better Times, University Labs Now Lack Research Funding, nat’l pub. radIo (Sept. 10, 2014), available at http://www.npr.org/2014/09/10/347305805/built-in-better-times -university-labs-now-lack-research-funding (last visited Oct. 17, 2019); Richard Harris, U.S. Science Suffering From Booms and Busts in Funding, nat’l pub. radIo (Sept. 9, 2014), available at http://www .npr.org/templates/transcript/transcript.php?storyId=340716091 (last visited Oct. 17, 2019). 87 Roberta Ness, The Creativity Crisis: Reinventing Science to Unleash Possibility (2015). 88 Susan Harlan, A Poem About Your University’s Brand New Institute, MCsweeney’s (Aug. 10, 2016), available at https://www.mcsweeneys.net/articles/a-poem-about-your-universitys-brand-new -institute (last visited Oct. 17, 2019). 89 Federal Role in American Graduate Education, supra note 66; Graduate Education and Research: Interdependence and Strain, supra note 66. 90 Structuring to Advance Science, supra note 11 85

108 Research handbook on intellectual property and technology transfer affiliate-faculty. Because it capitalizes on UF’s unique blend of serving as a land-grant institution and featuring a medical school on campus, and because it positions the university and faculty to compete for increasing shares of NSF and NIH funding, EPI operates within a supportive administrative context. The director, who is essentially the equivalent in authority and responsibility as a dean, uses a portion of EPI’s hard-money line from the state to subsidize the recruitment and hiring of departmental faculty. Departments can compete for top investigators at a relatively low price because of the institute’s subsidization, and both the institute and departments benefit financially from the overhead on grants that faculty win. EPI receives 7.5% of indirect cost recovery, while departments continue to receive their standard 10% overhead. The financial connection, by design, fosters a supportive network of collaboration to advance interdisciplinary research. Meanwhile, a core managerial challenge for centers and institutes entails the human-resource aspect. The tenure-homes of faculty are academic departments, and to recruit and retain faculty for centers and institutes is challenging due to added workloads91 and the bureaucracy they encounter as they seek to pursue research and collaboration with units outside of these tenure-homes.92 Additionally, center and institute directors confront misperceptions about the legitimacy of their units’ activities, such that many research-oriented centers and institutes also serve, rather than avoid, graduate and undergraduate students, offer training workshops and professional development opportunities, and even grant degrees.93 States have a long history of funding particular types of centers and institutes at colleges and universities,94 and they can kill those that no longer fit within emerging economic and sociopolitical agendas.95 The locus of accountability and authority can constrain center and institute directors and narrow the parameters within which they formulate strategies for research and teaching. The effectiveness of centers and institutes suggests their importance within the academic core. For example, industry members of the NSF-sponsored ERCs have reported an ability to leverage knowledge-production and -exchange to transform a $20,000 membership into a $4–5 million “state-of-the-scientific- as well as technological-art research program.”96 Industry partners typically renew memberships to I/UCRCs when their interests and goals are aligned and are satisfied with center communications, planning, and so forth.97 Faculty in university-industry centers are associated with strong levels of academic publishing, grant activity, and student support,98 and their involvement with industry heightens with govern-

91

Craig Boardman & Barry Bozeman, Role Strain in University Research Centers, 78 J. of hIgher eduC. 430 (2007). 92 Samuel Garrett-Jones, Tim Turpin, & Kieren Diment, Managing Competition Between Individual and Organizational Goals in Cross-Sector Researcha and Development Centres, 35 J. of teCh. transfer 527 (2010). 93 Geiger & Sá, supra note 3. 94 Elizabeth Berman, Creating the Market University: How Academic Science Became a Market Engine (2012); Wesley Cohen, et al., University-industry Research Centers in the United States (1994). 95 Scott Jaschik, UNC Board Kills 3 Centers, InsIde hIgher eduC (Mar. 2, 2015), available at https://www.insidehighered.com/news/2015/03/02/unc-board-kills-3-centers-amid-criticism-action -violates-academic-freedom (last visited Oct. 17, 2019). 96 Feller, et al., supra note 83, at 473. 97 Denis Gray, Mark Lindblad, & Joseph Simon-Rudolph, Industry-University Research Centers: A Multivariate Analysis of Member Retention, 26 J. of teCh. transfer 247 (2001). 98 Bozeman & Boardman, supra note 80.

University as knowledge-based enterprise 109 mental subsidies that attract additional private investment in basic research.99 The commitment of campus administrators to I/UCRCs, which by design come with a limited, mission agency-capped amount of indirect cost recovery, is based on various “objective” measures of center productivity, such as publications, additional grant activity, and student job placements, but also mediated by their relationship with the center directors.100 And in some contexts, such as in nanotechnology, a center’s success may be based on the number and productivity of other centers it seed-funds.101 Importantly, to evaluate centers and institutes poses methodological challenges. As Bozeman et al. (2012) observe of research on collaboration in science: One of the major limitations of current research is the tendency for researchers to either (1) focus intensely on either the world of the individual researcher while, unfortunately, ignoring the larger context within which the researcher operates, or (2) focus on collaborating organizations at a level of abstraction sufficiently general as to permit no consideration of the role of individual dynamics that may shape the outcomes of collaborating organizations.102

The ability to differentiate outcomes associated with individuals and groups, versus those clearly attributed to the centers and institutes themselves, would prove useful. In their promotional materials, centers and institutes usually tout success-claims about grant funding received or products and companies created, but it remains unclear whether the same group of investigators could have jointly produced these outcomes without affiliation to the centers and institutes. Among NSF-sponsored centers, there is wide variation in their structures, primary source of accountability, technical demands and goals, and prior managerial experience of directors.103 According to Denis Gray, who has evaluated NSF centers for more than 30 years, the precise question to ask is not whether, on the whole, I/UCRCs or ERCs are effective or not but rather: Is this type of Center [I/UCRC, ERC, etc.] as delivered by this kind of [university, industry, government, advocacy], catalyzed by this kind of leader, within these types of organizations, effective in

99 Craig Boardman, Government Centrality to University-Industry Interactions: University Research Centers and the Industry Involvement of Academic Researchers, 38 res. pol’y 1505 (2009). 100 Donald D. Davis & Janet L. Bryant, Leader-Member Exchange, Trust, and Performance in National Science Foundation Industry/University Cooperative Research Centers, 35 J. of teCh. transfer 511 (2010). 101 For example, nanotechnology entails science, engineering, and technology conducted at the nanoscale of about 1 to 100 nanometers and has broad applications. It can be used in developing a range of manufacturing processes and products from biosensors to clothing. Nanotechnology has relevance for improving human health, informing approaches to individualized medicine and precision in drug delivery in the human body. As such, a number of prominent research universities have overarching, “umbrella” nanotechnology centers and leverage them—personnel, resources, and facilities and equipment—to form other centers within specific scientific niches, arenas of external funding, and emerging economic markets. 102 Bozeman, supra note 8, at 36. 103 Craig Boardman & Denis Gray, The New Science and Engineering Management: Cooperative Research Centers as Government Policies, Industry Strategies, and Organizations, 35 J. of teCh. transfer 5 (2010); Craig Boardman & Branco Ponomariov, Management Knowledge and the Organization of Team Science in University Research Centers, 39 J. of teCh. transfer 75 (2004).

110 Research handbook on intellectual property and technology transfer reaching this goal [research, technology transfer, economic development, education], as assessed by this kind of measure, over this time frame?104

Feldman, Freyer, and Lanahan (2012) note that evaluative efforts should consider the typical metrics such as number of jobs created, spin-off companies formed, patents produced, and the like, but they indicate the need to capture broader social, cultural, and educational benefits and outcomes (e.g. special training opportunities for high school teachers and students).105 Even so, they add, evaluations are often limited because they do not adequately address the counterfactual. That is, it remains unclear analytically whether the outcomes associated with centers and institutes could have occurred under different structural and funding arrangements. While centers and institutes have historically helped to catalyze the research mission and enterprise in higher education, are they, in contemporary settings, necessary to anchor and drive interdisciplinary research and education? In their perceptive, well-cited work, Corley, Boardman, and Bozeman (2006) suggest that not all interdisciplinary, multi-institution collaborations require formal organizational structures.106 They posit that the degree of agreement among collaborators on core scientific problems and agendas, goals, techniques, and methods relates to the extent to which interdisciplinary collaborations feature formalized structural arrangements. The more experimental and ambiguous the epistemic terrain, such a perspective suggests, the more likely a strong organizational component is needed to ensure the viability of the project. Even with well-defined epistemic content and boundaries, multi-institution collaborations could result in elaborative forms of bureaucratic organization; these projects may be eligible to compete for center-level grants. Whether a center is needed or not to meet the technical demands of the work, it can be pursued and developed because of competition. That is, a multi-institution team becomes a center because there is funding (and prestige) for a center. What is more, the institutionalization of centers and institutes on campuses does not necessarily indicate effectiveness or productivity. Many centers and institutes have “lived on forever” with “more funding through various re-inventions.”107 Others could change their names, revise their scientific goals, and undergo a transition in leadership to reposition for new sources of funding in emerging opportunities in science. Either way, the center and institute as organizational form is unlikely to disappear any time soon, and many institutional leaders at the highly prestigious research universities continue to develop, adapt, and leverage them. The task at hand, then, is linking them to interdisciplinary schools and academic departments to launch and grow new disciplines and fields and achieve fluidity and continuity of research, teaching, and funding.

104

Slaughter & Hearn, supra note 36, at 5–6. Maryann P. Feldman, et al., “On the Measurement of University Research Contributions to Economic Growth and Innovation” in Universities and Colleges as Economic Drivers: Measuring Higher Education’s Role in Economic Development 97 (Jason E. Lane & D. Bruce Johnstone eds., 2012). 106 Elizabeth A. Corley, Barry Bozeman, & Craig Boardman, Design and the Management of Multi-institution Research Collaborations: Theoretical Implications from Two Case Studies, 35 res. pol’y 975 (2006). 107 Personal correspondence with Sheila Slaughter (Sept. 25, 2015). 105

University as knowledge-based enterprise 111

VII.

INTERDISCIPLINARY SCHOOLS

As part of their competitive strategies in the biomedical and life sciences, many research universities have folded core academic departments into interdisciplinary schools. The rise and prominence of the biotechnology industry,108 which developed in part from faculty consulting with firms and other academe-industry links, provides a strong incentive for colleges and universities to reorganize scientific research and education. The splintering of biology into microbiology and integrative lines, based on advancements in disciplines and fields and also on external funding opportunities, pushes toward reforms as well. The University of California, Berkeley first began to re-evaluate its 20 biology-related departments on campus in the early 1970s. According to observers, its diffuse, antiquated structure had hindered research, recruitment of top faculty and students, and prestige in then-burgeoning specializations of biochemistry, molecular genetics, and cell biology.109 By the late 1980s and early 1990s, Berkeley reorganized 20 units into four core departments and associated “divisions,” where scientists teamed together based on research interests (rather than departmental lines) and had access to laboratories with then-contemporary equipment and technology. These substantive organizational changes took about two decades to transpire, and they suggest that processes and political dynamics in colleges and universities may powerfully shape the viability of new structures and disciplines and fields to develop.110 The restructuring of biology at Berkeley has informed developments at other institutions, such as the University of Illinois at Urbana-Champaign (“UIUC”). There, by the late 1990s and early 2000s, the Dean of the College of Liberal Arts and Sciences and a selected group of faculty and departments launched the Schools of Molecular and Cellular Biology (“MCB”) and Integrative Biology (“SIB”).111 They did not dissolve departments per se, but grouped them into the different schools. MCB is comprised of the Departments of Biochemistry, Cell and Developmental Biology, Microbiology, and Molecular and Integrative Physiology. SIB features the Departments of Animal Biology, Entomology, and Plant Biology. Interestingly, the Department of Biochemistry had been previously located in the School of Chemical Sciences, where faculty in the department felt their unit received less support and fewer resources than other departments. The chair of the Biochemistry Department negotiated with the founding director of MCB and the Dean of the College to join the new school. Groupings of departments reflect disciplinary changes and also political maneuverings at the campus level. From a structural-functional perspective, UIUC constitutes a vast land-grant research institution and developed over time a striking number and range of biology-related departments and units on campus. To consolidate and coordinate these units, and to position them in relation to emerging opportunities in research, graduate education, and external funding, the school model had strong appeal among its advocates. But an interpretation through a critical lens draws attention to influences of money. A key factor in decision-making about academic

108

Martin Kenney, Biotechnology: The University-Industrial Complex (1986); Paul Rabinow, Making PCR: A Story of Biotechnology (1996). 109 Martin Trow, Biology at Berkeley: A Case Study of Reorganization and its Costs And Benefits (1999). 110 Simcha Jong, Academic Organizations and New Industrial Fields: Berkeley and Stanford After the Rise of Biotechnology, 37 res. pol’y 1267 (2008). 111 Structuring to Advance Science, supra note 11.

112 Research handbook on intellectual property and technology transfer structure was not so much the need for responsiveness to changes in the discipline of biology, but rather the increasingly stratified funding opportunities and competition to capture indirect cost recovery. MCB faculty are especially positioned to compete for NIH funding, which concentrates a disproportionate share of grants in the biomedical and life sciences arena. From their perspective the school design provides a distinctive resource advantage. When MCB faculty win grants, their indirect cost recovery that flows back to the school does not cross-subsidize SIB faculty who have different, and not always as robust, funding opportunities. Another strategic advantage for MCB includes the ability to increase productivity in and revenue flows from patenting, as their disciplinary area has strong prospects for technology transfer.112 That said, the consolidation of scientific activity in SIB has been helpful to faculty there as well. An entomologist won the National Medal of Science in 2014, and other faculty in the school have had strong records of grant funding from NSF and in areas of genomics research. The school design carries implications for the recruitment of faculty and the production of undergraduate and graduate education. Faculty hiring occurs at the school-level, with committees comprised of faculty representatives from each department. Searches are thus based around scientific problems or strategic areas of emphasis rather than on departments’ needing to fill lines that they otherwise lose if unfilled. The hired faculty member then decides with which department to affiliate for purposes of tenure and promotion. Teaching is structured through a school-level curriculum as well, without inter-departmental duplication of classes that reduces competition for students. During their first year of study, PhD students cycle throughout the array of laboratories across departments in their school, deciding thereafter with whom to work. At UIUC, MCB and SIB operate within an institutional context of revenue-centered management;113 academic units are responsible for generating revenues to cover or exceed their operating costs. Tuition receipts and credit-hour production matters, and to this end, MCB benefits in undergraduate enrollment because of perceptions among students of the connections between degrees from MCB and preparation for medical school. Other institutions have applied the school-based design approach across all academic departments. As part of Arizona State University (“ASU”) President Michael Crow’s vision for designing the New American University, he and a team of executive-level administrators and academic leaders reorganized ASU’s entire academic structure. In 2007, ASU launched 14 interdisciplinary schools organized around “faculties” (e.g., interdisciplinary areas). The schools aim to position faculty for cross-collaborative research and external funding opportunities and to reduce educational duplication and administrative costs.114 The School of Life Sciences provides an example of the changes at ASU. It includes the “faculties” of biomedicine and biotechnology, cellular and molecular biosciences, evolution, ecology, and environmental science, genomics and evolution, and organismal, integrative, and 112 Jason Owen-Smith & Walter Powell, Expanding Role of University Patenting in the Life Sciences: Assessing the Importance of Experience and Connectivity, 32 res. pol’y 1695 (2003). 113 Edward L. Whalen, Responsibility Centered Budgeting: an Approach to Decentralized Management for Institutions of Higher Education (1991); James C. Hearn, et al., Incentives for Managed Growth: A Case Study of Incentives-Based Planning and Budgeting in a Large Public Research University, 77 J. of hIgher eduC. 286 (2006). 114 Elizabeth D. Capaldi, Intellectual Transformation and Budgetary Savings Through Academic Reorganization, 41 Change 19 (2009).

University as knowledge-based enterprise 113 systems biology. According to the provost at the time, “The objective was to form a structure that could be easily reorganized around big programs and engage in use-inspired research.”115 In STEM, the School of Life Sciences is but one of several newer units that also includes the School of Sustainable Engineering and the Built Environment, the School of Electrical, Computer, and Energy Engineering, the School of Biological and Health Systems Engineering, and, among others, the School of Computing, Informatics, and Decision Systems Engineering. The 14 schools connect as well to centers and institutes to increase research expenditures and teaching activity without having a medical school.116 The UC-Berkeley and UIUC examples suggest the importance of coordinated and joint action among a range of stakeholders: deans, chairs, and prominent faculty members. Meanwhile, the case at ASU indicates constructive benefits of strong directive forms of executive-level leadership. Faculty would have to have buy-in to make such a sweeping change viable, but the push, as presented in the Capaldi (2009) and Crow and Dabars (2015) accounts, seems to have come from above and in ways that may have worked around shared governance and the faculty role in driving and shaping organizational change.117 It would be helpful to know more about the context and history of state- and institutional-level governance and how that context empowers, or constrains, ASU’s administrative leaders to design and implement change. It may be difficult to assess whether campus-level structural changes such as those at ASU address the embedded reward systems and incentives set by external colleagues and peers in traditional disciplines. Within the current configuration of academic structure on campus, tenure-line faculty at ASU are affiliated with an interdisciplinary school, but to earn tenure and promotion may need to demonstrate success according to traditional disciplinary measures and norms. Their colleagues on campus could continue to hold junior faculty to the standards of scientific excellence as framed within traditional disciplines (in which they trained and were themselves evaluated). The external reviewers, usually prominent senior faculty in a discipline or field, may also continue to uphold standards of assessment irrespective of whether the candidate “going up” is based in an interdisciplinary school or academic department.118 Campus structural changes may be loosely coupled from academic judgment and decision-making.119 The school-based design is associated with a number of success-claims. For example, Crow and Dabars (2015) suggest that ASU’s strategic initiatives, which include academic restructuring and shifts in the structure and operations of technology transfer, yield strong positive, marginal benefits.120 They note that from 2002 to 2014, research expenditures grew by about 250% and with minimal growth in the number of faculty. During this stretch of time, the institution exceeded $200 million in research expenditures—a notable level of activity for

115

Id. at 23–4. See Geiger, supra note 74 (discussing medical schools and their connection to research and education). 117 Capaldi, supra note 114; Crow & Dabars, supra note 1. 118 External reviewers typically receive the departmental-level or school-based criteria that campus colleagues use to evaluate dossiers for tenure and promotion decisions. Even so, the external reviewers are selected for their expertise and specialization and to draw on that expertise and specialization to assess the credentials and future potential of candidates. They may consider the departmental and school criteria but lean toward their understanding of norms and expectations of the broader profession. 119 Michele Lamont, How Professors Think: Inside the Curious World of Academic Judgment (2009). 120 Crow & Dabars, supra note 1. 116

114 Research handbook on intellectual property and technology transfer an institution without a medical school. Crow and Dabars highlight as well robust technology transfer activity. In 2013, they observe, faculty submitted 250 invention disclosures and founded 11 new start-up companies. One may expect to find as well, within the school model, evidence of cost efficiencies per the reduction in programmatic and administrative duplication. Analogous, though, to issues of assessing centers and institutes, the outcomes associated with interdisciplinary schools warrant a strong note of caution. Anecdotes aside, we do not know whether the evidence and claims of successes may have happened without the structural reforms. That is, the uptick in research expenditures could be related to strategic hiring initiatives that bring to campus highly productive early-career faculty and selected eminent scholars/endowed chairs. We also do not know much about the faculty experience in interdisciplinary schools and how they reconcile, or not, conflict and tension between goals of the institution and of the profession. The training, preparation, and labor market prospects of graduates of interdisciplinary schools warrants attention as well. How do academic job candidates, for example, with degrees from interdisciplinary schools fair as they seek faculty positions in traditional departments elsewhere? Interdisciplinary schools may provide high quality training, but suggest academic structures at odds with departmental forms of organization and the lock that departments have on generating prestige and on regulating and shaping academic labor markets.121 The content and associated outcomes of interdisciplinary schools require analytical attention.

VIII.

ACADEMIC DEPARTMENTS

Since their development throughout the 19th and 20th Centuries, academic departments serve as the foundational units and building blocks of academic organization.122 They anchor research, teaching, and service throughout the institution, and they orient the recruitment, socialization, and evaluation of faculty and carry out core governance processes. Process theories of professionalization offer an oft-used account for the development and institutionalization of academic departments. As such a perspective suggests, departments are correlated with disciplines and fields, which are comprised of specialists and experts who corner particular niches within the labor market.123 In the industrial era, specialists rather than generalists received disproportionate benefits in labor markets because of their focused contributions to discreet aspects of manufacturing and production.124 For faculty in the US, they began to receive increasingly advanced levels of training (largely overseas in Germany), differentiating themselves from each other and

121

Jarrett B. Warshaw, et al., Does the Reputation of a Faculty Member’s Graduate Programme and Institution Matter for Labour Market Outcomes?, 30 J. of eduC. and work 793 (2017). 122 Burton R. Clark, The Academic Life: Small Worlds, Different Worlds (1987); James C. Hearn. “Sociological Studies of Academic Departments” in Sociology of Higher Education: Contributions and their Contexts 222 (Patricia J. Gumport ed., 2007); Marvin Peterson, “The Academic Department: Perspectives from Theory and Research” in New Directions for Institutional Research: Examining Departmental Management 21 (John C. Smart & James R. Montgomery eds., 1976). 123 Andrew Abbott, The System of Professions: An Essay on the Division of Expert Labor (1988); Steven Brint, In an Age of Experts: The Changing Role of Professionals in Politics and Public Life (1994). 124 Weber, supra note 27.

University as knowledge-based enterprise 115 from other professional domains by way of their research expertise. They legitimized research within their repertoire of professional work, and their ability to advance socially certified knowledge as adjudicated by peers brought prestige to themselves and to their institutions. The introduction of research into the mission and operations of colleges and universities, and the positioning of these institutions in the industrial-based economy to train and certify experts and specialists, influenced the creation of newer fields and disciplines and structures to accommodate them.125 Over time, departments developed around distinctive niches of faculty experts and the particular external disciplines, fields, and professional associations that legitimate them globally.126 These structural forms have endured for centuries because of dynamics (and politics) of professionalization and because they have become entrenched in university budgets and resource allocation models. Arguably, the post-industrial knowledge economy amplifies the importance and value of experts and specialists and also heightens the jurisdictional disputes and competition among and within professions. Despite criticism that traditional fields and disciplines, and by extension their institutionalization in departmental form, hinders innovative change, disciplines are themselves contested terrains. Building on foundational work by Abbott (2001) and Gieryn (1983, 1999), Jacobs (2014) offers the following conclusion about competition within and between disciplinary fields and the implications for knowledge-production: Thus, disciplines are not silos but rather can be thought of as sharing a dormitory space where they raid each other’s closets and borrow each other’s clothes. This system is dynamic; competition occurs on many levels within fields as well as across fields. The very structure of the disciplinary system tends to push in the direction of competition and over time will generally arrest any tendency toward intellectual fossilization.127

The dynamics of disciplines—of what is or is not legitimate knowledge—reflects the jockeying of position and jurisdictional disputes among professionalized faculty.128 These established, enduring forms of organization may also bring legitimacy to and facilitate the advancement of emerging disciplines and fields. In their study of global health initiatives at US and Canadian universities, Oleksiyenko and Sá (2010) found that Harvard and Johns Hopkins assimilated global health into existing academic departments, helping these institutions solidify a competitive position in this arena relative to Toronto and McGill whose pursuits occurred through somewhat peripheral ORUs but not core departments.129 The broader political economy has an instrumental role in advancing selected emerging fields and disciplines while suppressing others, shaping academic structure and production on campuses. Consider the rise of materials science in the Cold War era. “The [national] defense

125

See Clark, supra note 122. Gilis Drori, John W. Meyer, Francisco Ramirez, & Evan Schofer, Science in the Modern World Polity: Institutionalization and Globalization (2003); Evan Schofer, The Global Institutionalization of Geological Science, 1800 to 1990, 68 aM. soC. rev. 730 (2003). 127 Jacobs, supra note 44, at 35; see also Abbott, supra note 123; Thomas F. Gieryn, Boundary-Work and the Demarcation of Science From Non-Science: Strains and Interests in the Professional Ideologies of Scientists, 48 aM. soC. rev. 781 (1983); Thomas F. Gieryn, Cultural Boundaries of Science: Credibility on the Line (1990). 128 Brint, supra note 123. 129 Anatoly Oleksiyenko & Creso M. Sá, Resource Asymmetries and Cumulative Advantages: Canadian and US Research Universities and the Field of Global Health, 59 hIgher eduC. 367 (2010). 126

116 Research handbook on intellectual property and technology transfer establishment, … virtually created materials science as an academic discipline, funding all but a tiny fraction of American materials research during the Cold War years.”130 When physics departments retreated from conducting defense research, wary from eroding too much autonomy during the Manhattan Project, their reticence helped materials scientists who prospered and organized into research units and then academic departments. Even so, physics departments continue to maintain healthy levels of institutional funding and support as compared to women’s studies departments. The barriers to forming women’s studies departments were steep, because of skepticism, among critics, of the field’s core intellectual content and due to isolated pockets of advocates throughout a given campus. When an external network of colleagues across campuses championed the issue, connecting to broader advocacy groups and a fomenting social movement, it helped to galvanize the formation, development, and spread of these departments within colleges and universities.131 Departments form and develop, or are restructured or closed, based on their political viability and relative density of networks of support rather than based on their intellectual merits and value.132 Private, philanthropic foundations—key actors in the political economy of competition and funding for colleges and universities—shape academic structure and modes of production as well. In the late 1990s and early 2000s, the Whitaker Foundation spent all of its holdings to form Biomedical Engineering Departments on campuses.133 To make permanent the institutional commitments to educating students in this arena, the foundation sought to create academic departments rather than tenuously funded and resourced centers and institutes. To some faculty and academic administrators, biomedical engineering is not necessarily viewed as a discipline in its own right.134 The discipline contains, for example, Bioengineering Departments, which focus on engineering and agriculture, and Biomedical Engineering Departments, focusing on intersections of chemical, systems, and mechanical engineering, materials science, and medicine. Bioengineering and Biomedical Departments, according to critics, constitute applications of other engineering disciplines rather than discreet, core intellectual domains. Despite its contested disciplinary status, bio-related engineering departments spread throughout the US, as institutions adopted them whether or not they received Whitaker money. Part of the diffusion could be attributed to competitive norms of emulation in higher education; that is, a normative value is placed on adoption of these departments whether or not the departments are effective in their work and goals. Their spread may also be rooted in opportunities to position departmental faculty for NSF and NIH funding for research, to attract strong students seeking to enter medical school, to foster close relationships with medical schools and biomedical firms, and to demonstrate social responsiveness to issues of human health and the broader public good.

130

Leslie, supra note 78, at 213. Patricia J. Gumport, Curricula as Signposts of Cultural Change, 12 rev. of hIgher eduC. 49 (1998); S. Slaughter, “The Political Economy of Curriculum-Making in American Universities” in The Future of the City of Intellect: The Changing American University 260 (Steven Brint ed., 2002). 132 Patricia Gumport, The Contested Terrain of Academic Program Reduction, 64 J. of hIgher eduC. 283 (1993); Patricia Gumport & Brian Pusser, “University Restructuring: The Role of Economic and Political Contexts” in Higher education: Handbook of theory and research, Vol. VII 146 (John Smart & William Tierney eds., 1999); Slaughter, supra note 131. 133 Stephan, supra note 2. 134 Structuring to Advance Science, supra note 11. 131

University as knowledge-based enterprise 117 The ascendancy and relative prominence of academic departments on campuses is rooted in resource- and power-asymmetries. Salancik and Pfeffer (1974) and Pfeffer and Moore (1980) suggest that departments receiving external resources (grants and contracts) are more likely to win internal, institutional resources.135 On one hand, such a pattern may be related to the indirect cost recovery/facilities upkeep issue; but, on the other, externally-funded departments could receive disproportionate institutional resources because they help their universities diversify revenue streams and ease their dependence on and accountability to any one resource provider. Jockeying for scarce resources heightens the micropolitics of campuses,136 as reflected, in recent years, in the uneven distribution of authority among department chairs137 and in the relative inequality in opportunities and resources for laboratory management.138 Power matters in organizations as well.139 According to Perrow (1986), power is “the ability of persons or groups to extract for themselves valued outputs from a system in which other persons or groups either seek the same outputs for themselves or would prefer to expend their effort toward other outputs.”140 According to such a definition, power is not necessarily about increasing the size of the pie, but rather about reallocating the share of the various pieces or changing the “rules of the game” and intended end-goals entirely. To the extent that departments adapt their work and goals,141 they might increase the magnitude of and broaden their influence on their campuses. Forms of stratification may, in contexts of escalating competition, heighten further. For example, many universities privilege STEM departments and faculty (targeted with about 97% of the federal R&D budget) over the arts, humanities, and social sciences (whose fields constitute about 3% of the federal R&D budget).142 For each additional dollar that an institution invests in the arts, humanities, and social sciences, it incurs a “tax” in reducing the likelihood of securing federal R&D funding.143 But stratification is manifested within STEM and the arts, humanities, and social sciences, according to the relative value of IP (whether research or education/instruction-based). Academic departments pursue strategies to position within external and campus-level competitions for resources and status. The 1998 to 2003 $25 million contract between Novartis Agricultural Discovery Institute (Syngenta) and faculty of UC-Berkeley’s Department of Plant and Microbial Biology (“PMB”) offers an example of departmental strategy. As Berkeley’s administrative leaders sought institutional partnerships in the 1990s with biotechnology firms, they set a context for department-level changes. In 1997, a guiding coalition of four— 135

Gerald R. Salancik & Jeffrey Pfeffer, The Bases and Use of Power in Organizational Decision Making: The Case of a University, 19 adMIn. sCI. Q. 453 (1974); Jeffrey Pfeffer & William L. Moore, Power in University Budgeting: A Replication and Extension, 24 adMIn. sCI. Q. 637 (1980). 136 Jeffrey Pfeffer, “The Micropolitics of Organizations” in Environments and Organizations 29 (Marshall W. Meyer & Associates eds., 1978). 137 Barry Bozeman, Daniel Fay, & Monica Gaughan, Power To Do…What? Department Heads’ Decision Autonomy and Strategic Priorities, 54 res. In hIgher eduC. 303 (2013). 138 Cantwell, supra note 10. 139 Jeffrey Pfeffer, Power in Organizations (1981). 140 Charles Perrow, Complex Organizations: A Critical Essay (3rd ed., 1986). 141 Curtis L. Manns & James G. March, Financial Adversity, Internal Competition, and Curriculum Change in a University, 23 adMIn. sCI. Q. 541 (1978). 142 Rosinger, supra note 19. 143 Barrett Taylor, et al., Quasi-Markets in U.S. Higher Education: The Humanities and Institutional Revenues, 84 J. of hIgher eduC. 675 (2013).

118 Research handbook on intellectual property and technology transfer comprised of the Dean of Natural Sciences, the PMB chair, and two PMB professors—put forward an initiative to get companies to compete with each other to partner with the department and fund its research. As Rudy et al. (2007) observe, the four-point strategy entailed: 1. Selecting an industry research partner to optimize the financial, technological, and academic benefits for the department and for California’s agricultural economy; 2. Leveraging a process of competitive bidding among potential industry partners; 3. Structuring the partnership so that the industry partner provides unrestricted funds for departmental research of relevance to the public good and without oversight; and 4. Ensuring the industry partner extends access to technology and other materials (e.g., data) that benefit the department’s research and that is otherwise too costly for the department to secure on its own.144

Importantly, Rudy et al. note, the “reversal in the power dynamic between grantor and grantee was a major shift in the pursuit of university-industry relations and became commonly referred to—particularly by opponents of the subsequent agreement—as ‘auctioning off the department.’”145 Despite concerns of other university faculty on campus, the PMB-Novartis partnership went forward. It received approval from the highest levels of central administration. The partnership was reviewed during the course of its collaboration, especially regarding potential conflicts of interest, and the partnership was dissolved when the contract expired. Advocates suggest that the department was able to achieve its goals of: protecting autonomy to determine research directions, enhancing creativity, and promoting intellectual diversity across the full range of faculty members. Additionally, it suggests an entrepreneurial maneuver of a semi-autonomous academic department to generate unrestricted research funds and an arrangement by which the department held ownership of IP and Novartis was able to license it. At the same time, the development indicates that the market value of an academic department can be set by competitive bids from external stakeholders and organizations. Of course, there is variation in the extent to which academic departments in STEM interact with industry. Mendoza (2012) suggests that, on balance, departments can be “industry-friendly” by jointly pursuing unrestrictive grants, publication, and student placements.146 She concludes that in some settings the pernicious aspects of industry partnerships, aspects corrosive to the social values of academic institutions to pursue disinterested knowledge, may thus be mediated. Meanwhile, chemistry departments have, since their central role in World War I, long-supported and benefited from relationships with pharmaceutical companies.147 The connection to industry sponsors of research constitutes a crucial element in financing and resourcing academic departments and advancing professionalization as

144

Alan Rudy, et al., Universities in the Age of Corporate Science: The UC Berkeley-Novartis Controversy 47–8 (2007). 145 Id. at 48. 146 Pilar Mendoza, The Role of Context in Academic Capitalism: The Industry-Friendly Department Case, 83 J. of hIgher eduC. 26 (2012). 147 John Swann, Academic Scientists and the Pharmaceutical Industry: Cooperative Research in Twentieth-Century America (1988).

University as knowledge-based enterprise 119 a field and discipline. As Slaughter and Silva (1983) observe, the importance of social science research to industry led to the professionalization of social scientists.148 The financial importance of industry relationships to academic departments may become especially salient in particular types of university budgeting models. As noted, to incentivize SOUs to compete for external resources and balance their costs and revenues, many campuses now feature incentives-based planning and budgeting, also referred to as Revenue Centered Management (“RCM”). The model makes each unit responsible for generating revenues to cover costs and, potentially, to profit. RCM appeals to institutional leaders and some faculty members because it decentralizes the budgeting process, encouraging deans, chairs, and faculty to pursue new and diverse forms of revenues. Relatedly, the approach supports an entrepreneurial culture and portfolio of activities by which academic units can adapt quickly to opportunities or challenges in their particular external environments. If applied without modification to allow for some campus-wide cross-subsidization, RCM can thus increase programmatic and staffing duplication and impede resource flows to areas that may be struggling to produce revenues to cover costs. In such a scenario, a department could be closed due to its economic value rather than its academic merits. To be clear, RCM is not inherently biased toward STEM units and faculty per se. Some institutions generate strong revenue-flows from departments and programs in the liberal arts, seeking to invest in them to expand capacity relative to their enrollment market.149 In some settings and enrollment markets, the liberal arts are associated with prestige. Even in less favorable and financially tenuous contexts, academic departments in the arts, humanities, and social sciences may deploy revenue-generating strategies based on instructional activities. As Slaughter and Rhoades (2004) note of departmental strategies on campuses and in the knowledge economy more generally, “a number of fine arts colleges, traditionally not conceptualized as close to the market, have redefined themselves so that they train arts students in graphic design, digital animation, and web design, therefore connecting directly to the new economy.”150 They suggest other adaptations to competition for resources: Colleges of Education sell tests and measurements that departmental faculty have copyrighted: Classics Departments charge money for guided tours of Greece and Rome, and Anthropology Departments also offer tours and, for additional fees, opportunities to dig at prehistoric sites. But the emphasis on instruction for revenue generation may preclude these departments from opportunities to develop robust research agendas and resource streams,151 while in STEM the research economy provides funding and incentives for research and educational initiatives. Despite the widening use of RCM, it has been difficult to discern whether incentives-based models lead toward the intended outcomes.152 For example, it remains unclear whether the amplification of competition among departments for institutional and external resources

148 Sheila Slaughter & Edward T. Silva, Service and the Dynamics of Developing Fields: The Social Sciences and Higher Education Studies, 54 J. of hIgher eduC. 481 (1983). 149 Patricia Gumport & Stuart K. Snydman, The Formal Organization of Knowledge: An Analysis of Academic Structure, 73 J. of hIgher eduC. 375 (2002). 150 Slaughter & Rhoades, supra note 18, at 27. 151 William F. Massy & Robert Zemsky, Faculty Discretionary Time: Departments and the “Academic Ratchet”, 65 J. of hIgher eduC. 1 (1994). 152 James C. Hearn, Darrell R. Lewis, Lincoln Kallsen, Janet M. Holdsworth, & Lisa Jones, Incentives for Managed Growth: A Case Study of Incentives-Based Planning and Budgeting in a Large Public Research University, 77 J. of hIgher eduC. 286 (2006).

120 Research handbook on intellectual property and technology transfer helps campuses increase their shares of federal R&D funding and other contracts and grants they otherwise may not have won.153 The extent to which RCM leads to effectiveness and efficiency is unclear as well. While application of the model aims to improve organizational performance, it could have an unintended consequence that limits any positive, marginal gains. Depending on its content and form, it could exacerbate inequality throughout a campus, deepening stratification and precluding the cross-departmental pursuit of innovative synergies and collaboration.154 The strategic positioning of academic departments may gain increasing salience in times of financial hardship for institutions. Periods of resource turbulence magnify inequities that have been present all along, making manifest the underlying, latent tensions and fissures.155 For example, during economic recession in the 1980s and the related retrenchment in state funding, many public university leaders closed academic departments perceived as less central to revenue-generation and the economic interests of state policymakers.156 The departments targeted for closure were in the arts, humanities, and social sciences (e.g., education) and also included academic libraries. As Volk, et al., (2001) found, the use of market rhetoric to justify inequities in resource allocations—and by extension decisions regarding the viability of departments—mask gender biases that concentrate resources to areas with high proportions of men (e.g., physics) despite the lack of robust external resources available in those fields.157 They call into question the assumptions of resource allocation models, suggesting the need for reforms to account for gender and the myriad markets and missions of departments. To measure the effectiveness (doing the right things) and efficiency (doing things right) of academic departments has been of keen interest to institutional researchers and stakeholders. Keller (1983) suggests applying to program reviews a portfolio matrix of evaluation based on the Boston Consulting Group’s model.158 Treating academic departments as profit-centers, Keller notes, the portfolio matrix helps stakeholders to identify: ● ● ● ●

Cash Cows of stable departments that generate revenues; Dogs of low-growth with little income; Question Marks of growth potential but unproven in resource competitions; and Stars of strong growth and competitive positioning.159

Such an approach in corporate settings suggests to leaders and managers the need to sell-off dogs so that, as Birnbaum observes, “capital could be invested in stars and question marks,

153

See Geiger, supra note 74. Jarrett B. Warshaw & James C. Hearn, “Federal Spending on Higher Education: A System of Opportunity and Stratification” in Administration, Finance, and Budgeting in Higher Education and Student Affairs: Theory, Research, and Practice (Gabriel R. Serna & Joshua M. Cohen eds., in press). 155 See, e.g., James C. Hearn, et al., Privatization and Accountability Trends and Policies in U.S. Public Higher Education, 41 eduC. and sCI. 1 (2016). 156 Gumport, supra note 131; Cynthia Hardy, “Hard” Decisions and “Tough” Choices: The Business Approach to University Decline, 20 hIgher eduC. 301 (1990); Sheila Slaughter, Retrenchment in the 1980s: The Politics of Prestige and Gender, 64 J. of hIgher eduC. 250 (1993). 157 Cindy S. Volk, et al., Models of Institutional Resource Allocation: Mission, Market, and Gender, 72 J. of hIgher eduC. 387 (2001). 158 George Keller, Academic Strategy: The Management Revolution in American Higher Education (1983). 159 Id. 154

University as knowledge-based enterprise 121 while income from cash cows kept the whole group humming along.”160 The framework for evaluation has gained some support among strategic planners who advocated its potential for revising and investing selectively in program mixes and portfolios.161 Nevertheless, the particular measures, available data, and criteria by which to inform decision-making about departments and programs warrant close consideration. Analogous to studies of centers, institutes, and interdisciplinary schools, research on academic departments relies on several indicators of effectiveness and efficiency. Researchers and analysts may examine job placement of graduates, faculty productivity, number of graduates, morale/perceptions of clients (students), and feedback from other partners/resource providers (private industry, central administrators, etc.).162 Middaugh (2001) launched a multi-institution data collection and sharing project entitled the National Study of Instructional Costs and Productivity, dubbed the “Delaware Study” for his home institution the University of Delaware.163 These data track the relative cost efficiencies and productivity of academic departments of member institutions. They indicate, for example, direct instructional expenses and can be analyzed to inform departmental and campus-level policy and practice regarding faculty and instruction. As Hearn and Gorbunov (2005) observe, the Delaware Study has several limitations, including its omission of indirect costs, research-related costs, and costs associated with departments’ ancillary centers and institutes.164 Additionally, the data include relatively few qualitative indicators and thus preclude broader analyses of “quality.” Another way to examine the relative quality of academic departments on their campus is based on measures of centrality and connectedness. Hackman (1985) suggests that departments demonstrating centrality and connectedness are more likely than other departments to secure internal resource allocations.165 Centrality means that departments’ missions closely match and serve the missions of their home institutions. Connectedness indicates the extent to which one department relies on another (e.g., whether students from one department are required to take courses in another department). Arguably, though, quality could be manifested in high levels of centrality and lack of connectedness such that the department is self-sufficient or does not need to share students across other programmatic areas because of healthy enrollment demand. Any effort to capture and report “quality” is subject to political spin.166 Departments may selectively highlight or craft communication in a way to offer the appearance of needing additional resources or faculty-lines to support institutional mission and its relationship with other departments on campus. Data are also open to interpretation among administrators. They

160

Birnbaum, supra note 24, at 6. Daniel J. Rowley & Herbert Sherman, Academic Planning: The Heart and Soul of the Academic Strategic Plan (2004). 162 Kim S. Cameron & Mary Tschirhart, Postindustrial Environments and Organizational Effectiveness in Colleges and Universities, 68 J. of hIgher eduC. 87 (1992); Massy & Zemsky, supra note 151. 163 Michael Middaugh, Understanding Faculty Productivity: Standards and Benchmarks for Colleges and Universities (2001). 164 James C. Hearn & Alexander V. Gorbunov, “Funding The Core: Understanding The Financial Contexts of Academic Departments in the Humanities” in Tracking changes in the humanities: Essays on finance and education 1 (Malcolm Richardson ed., 2005). 165 Judith Dozier Hackman, Power and Centrality in the Allocation of Resources in Colleges and Universities, 30 adMIn. sCI. Q. 61 (1985). 166 Federal Role in American Graduate Education, supra note 66. 161

122 Research handbook on intellectual property and technology transfer do not tell decision-makers what to do and how. Efforts to collect, compile, interpret, and formulate academic strategy based on data—and who is involved in these processes and to what effect—merits further research.

IX.

AFFILIATED NON-PROFIT ORGANIZATIONS

An emerging line of research in higher education brings attention to the formation and use of 501(c)(3), non-profit foundations among US research universities. With the exception of for-profit institutions, colleges and universities are themselves non-profit organizations that receive tax-exemptions from the Internal Revenue Service and tax-deductions on charitable contributions and donations.167 Research universities are connected to numerous affiliated non-profit organizations (“ANPOs”) whose sole mission centers on supporting their respective institutions and/or that take on recognizable forms indicating their focused support of their respective institutions (e.g., athletics foundations, alumni foundations, etc.).168 From a legal perspective, ANPOs constitute 501(c)(3) entities as they do not redistribute revenues to shareholders, employees, and the like, reinvesting revenues in pursuit of mission, and they fall under the authority of state attorney generals and the federal tax code. In short, they can amass untaxed assets with somewhat limited governmental (and institutional) oversight. The rise and spread of ANPOs among research universities indicates structural adaptations to competition for resources. For example, ANPOs are connected to different aspects of institutional mission; some have clear ties to supporting research, while others serve educational, co-curricular, and service-related operations and activities. That some ANPOs can raise money and deploy their funds to particular segments of their institutions offers critical resource-support to institutions, which face escalating costs of research and teaching and also competitions for prestige and external funding. The Georgia Tech Research Institute (“GTRI”) provides an example. State legislators and regents founded GTRI as the State Engineering Experiment Station in 1934. It constitutes a non-profit organization and a division of applied research of Georgia Tech. While its operational focus, scope, and scale have broadened over the years, GTRI manages research grants and contracts from industry and government and taps into and also serves to strengthen the faculty, students, and scientific infrastructure of the university. When GTRI brings together governmental, industry, and academic stakeholders, it can deploy its resources to solve scientific problems of funders, support selected entrepreneurial pockets of faculty, students, and research initiatives, and elevate the research prestige and resource capture of the campus as a whole. Within the context of loose coupling, the formation and utilization of ANPOs does not necessarily indicate strategic coordination. ANPOs fall outside of formal university governance. External stakeholders may found these non-profit entities, with or without involvement of campus administrators and faculty. Meanwhile, central administrators and individual departments and faculty, either jointly or independent of each other, could develop ANPOs.

167

Warshaw & Hearn, supra note 154. “Authentic” Faculty and Academic Capitalism, supra note 11; Barrett J. Taylor, et al., Affiliated Non-Profit Organizations: Strategic Action and Research Universities, 89 J. of hIgher eduC. 422 (2018). 168

University as knowledge-based enterprise 123 An example entails an academic department whose chair forms an ANPO to accrue resources from services and activities that can be deployed as revenue-generation for that department. ANPOs merit attention for the ways in which they resource the research enterprise and commercialization169 and for their board members’ connections to private firms. The composition of the boards and how board members influence the reorganization of science at colleges and universities warrants consideration. Because the selection of board members to ANPOs occurs through these entities, rather than via institutional policies and procedures and state statutes, it increases the likelihood of strategic appointments (e.g., not just alumni donors). As such, board members could, by way of ANPOs, tie together multiple other organizations, including science corporations and colleges and universities. When corporate board members also serve on the boards of ANPOs and colleges and universities, they facilitate interlocking networks among these various entities and sectors. Likewise, university presidents, other administrative leaders, and faculty may serve on ANPO boards that connect to other organizations. The board members circulate knowledge and information, steward resources, and facilitate exchanges between academe and industry in ways that position segments of universities—and universities writ-large—within emerging markets of science and technology. To be clear, ANPO boards have not yet been studied in higher education while university-level trustees have garnered increasing scholarly attention. The research on university trustees and exchanges between universities and science corporations indicates strategic advantages for these institutions but also strong possibilities for institution-level conflicts of interest.170 Over time, the composition of trustees at the elite private research universities reveals a stable proportion of ties between board members and science corporations—even as the proportion of ties between board members and other types of firms has decreased.171 The ties of trustees to science corporations are reflected in the convergence of patenting profiles across sectors172 and in the positive, marginal gains in institutions’ federal R&D funding.173 The trustee-enabled exchanges between universities and firms unfold differently for private and public institutions. Trustees of public institutions are often political appointees of governors who may not necessarily be connected to science corporations (but are connected to political interests/agendas). In turn, the elite private research universities gain disproportionate access to information, resources, and, potentially, competitive markets for their faculty, students, and technology transfer.174 Barringer and Slaughter (2016) analyzed university-firm networks (by way of trustees) of the US members of the prestigious Association of American Universities (“AAU”) and related

169

Goldie Blumenstyk, Treatment that Brings a Tear to the Eye, 49 Chron. of hIgher eduC. A24 (2003). 170 Sheila Slaughter, et al., U.S. Research Universities’ Institutional Conflicts of Interest Policies, 4 J. of eMpIrICal res. on huM. res. ethICs 3 (2009). 171 Charles Mathies & Sheila Slaughter, University Trustees As Channels Between Academe And Industry: Toward An Understanding Of The Executive Science Network, 42 res. pol’y 1286 (2013). 172 Sheila Slaughter, et al., Institutional Conflict of Interest: The Role of Interlocking Directorates in the Scientific Relationships Between Universities and the Corporate Sector, 85 J. of hIgher eduC. 1 (2014). 173 Mathies & Slaughter, supra note 171. 174 Brian Pusser, Sheila Slaughter, & Scott Thomas, Playing the Board Game: An Empirical Analysis of University Trustee and Corporate Board Interlocks, 77 J. of hIgher eduC. 747 (2006); Slaughter, supra note 170.

124 Research handbook on intellectual property and technology transfer exchanges at two sampled institutions, MIT and University of Pittsburgh (Pitt).175 The content of exchanges includes committee/advisory board memberships; research and educational initiatives; academic appointments or lectureships for firm executives and employees; donations and philanthropic giving; and co-development of patents and publications. They found at MIT a deep engagement of trustees in the intellectual life of particular S&T-focused segments of the institution, noting: The university provided infrastructure and resources of various kinds, including symbolic, human capital, research, expertise and prestige for knowledge economy corporations and the corporations and other entities represented by the trustees did the same for the university.176

In contrast, the exchanges at Pitt were less concentrated in the S&T domain and more focused on channeling donations to support the finance, insurance, and real estate arena by way of the business school. A key conclusion, according to Barringer and Slaughter (2016), entails the increasing potential for institution-level conflicts of interest (from which trustees are traditionally supposed to buffer) and also the segmentation of universities based on their relative contributions to science corporations and vice-versa. “This may result,” they conclude, “in deeply segmented research universities, where entrepreneurial graduate education and faculty research are expanded and those who participate in these endeavors move into a variety of markets while other areas contract and/or subsidize entrepreneurial activity.”177 Governance mechanisms of ANPOs differ from boards of trustees, as they do not entail legal and fiduciary responsibility for the university as a whole. What is more, we know relatively little about the networks and content of exchanges of other AAU institutions and of those that aspire to enter the AAU. To the extent that some strategically deployed ANPO boards feature members acting in ways analogous to those at MIT and Pitt, shaping multiple organizations across sectors, modes of scientific production, and competitive markets, research is merited. The non-profit foundation model of design has informed the restructuring and repositioning of TTOs. As Rooksby (2011) found, some research universities transitioned their TTOs into 501(c)(3) organizations in order to enhance the strategic flexibility of working with faculty and industry and protecting patents.178 Not all research universities seek to engage in patent infringement litigation, especially because of the financial costs accrued over time and throughout the protracted legal process. But for those that do, the non-profit foundation approach to TTOs can prove beneficial. As these types of TTOs manage IP on behalf of the university, and as these TTOs thus pursue legal recourse to protecting that IP, they safeguard the image of their universities. That is, the university seems less directly involved in litigious action than does the 501(c)(3) TTO. Despite university leaders’ success-claims associated with leveraging the TTO non-profit organization to manage technology transfer for the campus, there have been relatively few empirical studies in this arena. The structural location of university-housed TTOs, and how

175

Sondra Barringer & Sheila Slaughter, “University Trustees and the Entrepreneurial University: Inner Circles, Interlocks, and Exchanges” in Higher Education, Stratification, and Workforce Development: Competitive Advantage in Europe, the US, and Canada 151 (Sheila Slaughter & Barrett Taylor eds., 2016). 176 Id. at 164. 177 Id. at 168. 178 Rooksby, supra note 30.

University as knowledge-based enterprise 125 that location within a campus administrative hierarchy influences technology transfer (e.g., patents), has been well documented.179 But stakeholders know relatively little, beyond anecdotal and descriptive reports, about the extent to which—and how—TTO non-profit organizations enhance “flexibility” and lead to positive, marginal gains on selected outcomes (e.g., number of disclosures, licenses, spin-off firms, jobs created, venture capital received, etc.). Such changes have already occurred at University of Arizona and at Arizona State University. The antecedents shaping diffusion of this TTO design across universities are important to study. A related, future line of research could focus on the morphing of university-based centers and institutes into independent (or semi-autonomous) non-profit entities. For example, Applied Research Institutes focus largely on research contracts and may pursue, in relation to Department of Defense-oriented funding, classified and non-classified projects. Applied Research Institutes, analogous to GTRI, could in effect become 501(c)(3) foundations equipped to generate resources, organize research, interact with its affiliated university, hold and manage IP, and keep at arm’s-length their universities from litigation and also criticism over classified, national security-oriented projects.180 But such institutes may not want to “go out on their own” because they lose full access to university resources and infrastructure (e.g., in-house counsel, IP policies, etc.). The conditions under which some centers and institutes morph into non-profits, and the outcomes associated with those transitions and organizational forms, are potentially fruitful arenas for empirical studies.

X.

CONCLUDING REMARKS

The concept of design, at heart, suggests a framework for action. Its advocates indicate the strong potential for application to research universities, which, in their view, are institutions that can and should be reorganized to optimize knowledge-production and -exchange. Arguably, “design” is itself undergoing a process of professionalization. It gains legitimacy from and ascends in popularity and usage via educational programs, faculty experts, and external professional and alumni networks. There are growing concerns among critics who take issue with the core assumptions informing design-based strategies of change in higher education. But due to mechanisms of professionalization, and due to the increasing expectations on and competitive interests of many research universities to transform the economy and human health, the idea of design will likely continue to gain attention of university stakeholders. This Chapter has aimed to address knowledge-gaps for university stakeholders on the prospects and limitations of design-based strategies in research universities. SOUs—centers and institutes, schools, and departments—offer illustrative examples of how and why academic structure changes and to what end or purpose. As university stakeholders consider and formu-

179

Janet Bercovitz, et al., Organizational Structure as a Determinant of Academic Patent And Licensing Behavior: An Exploratory Study of Duke, Johns Hopkins, and Pennsylvania State Universities, 26 J. of teCh. transfer 21 (2006). 180 See Leslie, supra note 78 (discussing student and faculty backlash at Stanford and MIT regarding classified research during the Cold War).

126 Research handbook on intellectual property and technology transfer late plans to reorganize scientific research and education on their campuses, they will confront several critical questions. For instance: ● ● ● ●

To what extent is “design” a fad in the making? What (or who) constrains the imaginative possibilities for design? Who does the designing—and how? And how do emerging policy contexts influence design?

Though not an exhaustive list, each of these questions engages core issues and tensions in reconciling design-based interventions with dynamics of postsecondary organization. In this section, each is addressed to inform future considerations for research, policy, and practice. A.

To What Extent is “Design” a Fad in the Making?

Higher education has, over the years, adopted a variety of new, promising management systems and processes from the corporate and governmental sectors. The impetus for adoption among colleges and universities comes from strategic choice or mandate (e.g., of state policymakers and governing boards). For example, many institutions have pursued Business Process Reengineering and Benchmarking, Total Quality Management, Zero-Based Budgeting, and Management by Objectives (“MBO”), only to abandon them years later. These initiatives, despite their transformational potential, have been fads: initiatives or efforts pursued with vigorous zeal for a delimited time.181 The concept of design, this Chapter suggests, has strong theoretical foundations in management, organizational sciences, and sociology. Its spill-over to research universities from industry and government labs, where matrix approaches organize the production of science, suggest one potential marker of what will become a fad. The appearance and promotion of design as the solution for a variety of complex, nuanced organizational and operational issues—and across organizational settings and sectors—indicates another potential fad-like quality. Fads are only identified as such after-the-fact: after they cycle into and through an organization. It is premature to say whether the concept of design meets the criteria of being a fad. Despite their promise of success in improving organizational performance, fads can have detrimental consequences and incur high financial costs. For example, to utilize design-based strategies to restructure an entire campus is nearly irreversible and can come with unintended consequences. University stakeholders could thus consider pilot programs before determining whether their design-based efforts should be applied more broadly. Such pilot programs could entail the tailored use of design to individual units on campus or to particular sets of units on campus. Likewise, faculty could create a cross-departmental degree program that draws on colleagues and mobilizes students from disparate majors. It may be difficult, in advance of implementation, to predict unintended effects (e.g., widening stratification among departments and disciplines). But a key strength of research universities is loose coupling, and to the extent that design-based strategies can be targeted and buffered, university stakeholders could learn if and how their approaches should be deployed across the campus as a whole.

181

Birnbaum, supra note 24.

University as knowledge-based enterprise 127 B.

What (or Who) Constrains the Imaginative Possibilities for Design?

For university stakeholders, the formulation of design-based strategies gets at fundamental issues of imagination. That is, by way of design, to what extent are stakeholders able to push toward novel, bold visions of what a university can and should look like? As this Chapter’s discussion of SOUs suggests, we could be witnessing (and facilitating) the institutionalization of the organizational form. Not all research and educational collaborations require or transition into formalized structures of centers and institutes, schools, and departments. But the idea of creating and leveraging a type of recognizably bureaucratic unit, which bestows legitimacy and positions collaborative teams for different types, amounts, and sources of funding, seems entrenched in decision-making in this arena. More generally, we tend to anthropomorphize organizations and ascribe to them qualities and characteristics analogous to those of people. Organizations have legal rights, endowing them with human-like protections for ownership and property, free speech (e.g., donating to political campaigns), and religious freedoms (e.g., restricting healthcare options for women).182 When university stakeholders consider the reorganization of science, aspiring to create fluid, flexible configurations and arrangements of academic structure, they tend to employ inherited, longstanding models and templates. For example, interdisciplinary schools present novel possibilities for knowledge-production and -exchange but have origins in the designs of other institutions.183 As such a perspective suggests, might even bolder, creative possibilities exist? And what are the conditions that enhance how we imagine postsecondary organization? A college or university could, contingent on its design and resources, reflect in its internal organization high degrees of flexibility and fluidity. Envision, then, a campus without lectures or even discreet classroom spaces, but instead a single, open, and expansive laboratory-type setting facilitating interaction, communication, and work on applied projects.184 There have been efforts over the years to devise newer public universities without academic departments and to formulate and coordinate research, teaching, and service in innovative ways. Yet these newer public universities typically face choices between receiving core funding from the state or receiving less state funding in lieu of increasing autonomy.185 In settings of close financial ties to the state, founding administrators and faculty usually default to traditional, inherited models and templates of organizational design.186 Resources, more so than public or private, non-profit control, shape the boundaries of imagination and the related institutional designs. There are limitations on imagining research universities anew, but also possibilities for designing novel, creative combinations with longstanding organizational forms and templates. For example, the strategic use and coordination of ANPOs indicates one avenue

182

Nina Totenberg, When Did Companies Become People? Excavating The Legal Evolution, nat’l pub. radIo (July 28, 2014), available at http://www.npr.org/2014/07/28/335288388/when-did -companies-become-people-excavating-the-legal-evolution (last visited Oct. 17, 2019). 183 Rogers Hollingsworth & Ellen Jane Hollingsworth, “Major Discoveries and Biomedical Research Organizations: Perspectives on Interdisciplinarity, Nurturing Leadership, and Integrated Structure and Cultures” in Practising Intersdiciplinarity 215 (Peter Weingart & Nico Stehr eds., 2000). 184 Jeffrey Young, MIT Dean Takes Leave to Start New University Without Lectures or Classrooms, Chron. of hIgher eduC., available at http://chronicle.com/article/MIT-Dean-Takes-Leave-to-Start/ 235121 (last visited Feb. 1, 2016). 185 Hearn, et al., supra note 155. 186 Brewer & Tierney, supra note 59.

128 Research handbook on intellectual property and technology transfer for design-based reform efforts. ANPOs are not innovations per se because they constitute recognizable, legitimate, legally endowed, and historic organizational forms. Yet they can potentially be deployed in ways to achieve unique configurations of resource support and knowledge-exchange. That said, creative use of ANPOs also raises potential conflicts of interest and legal concerns about the content of the exchanges between them (and their boards) and universities and segments of universities. Thus the legal, regulatory environment shapes the imaginative and tactical possibilities. C.

Who Does the Designing—and How?

As noted earlier in the Chapter, advocates of using design-based strategies in higher education have not necessarily discussed the details of governance and decision-making. A concern, then, for some university stakeholders entails the extent to which strong, directive forms of administrative leadership drive organization-wide changes and sideline the faculty. The involvement of university stakeholders in designing academic structure could be limited to particular types of SOUs. For instance, the campus-wide restructuring of departments into interdisciplinary schools requires some degree of administrator and faculty involvement. When administrative leaders issue calls for proposals to seed-fund centers and institutes, incentivizing change from the “ground up,” they set the “rules of the game” and reward systems by which to shape faculty efforts. Meanwhile, the disciplines, as they professionalize, typically look toward academic departments as the gold standard of organization. In loosely coupled systems, the organizational design imposes itself. Academic structure unfolds in non-linear, organic ways, suggesting to stakeholders the need to consider how to coordinate the design that emerges. To this end, we need to know more about leadership, management, and governance within and across SOUs. Leadership constitutes the “glue” of organizations, management focuses on day-to-day operations and work, and governance entails setting the purposes and goals and identifying processes and procedures to pursue those purposes and goals. Together they inform design and suggest several questions that merit attention: Who leads SOUs? How are SOU leaders selected and by whom? To what extent are SOUs led by career academic administrators or, as Warshaw (2018) suggests, by “authentic” scientists with strong research accomplishment and track records of external funding?187 How might the managerializing of academic faculty, transitioning into administrative positions of leadership, relate to dynamics of professionalization and competition? In relation to management, Bozeman and Youtie (2017) highlight several strategies for team-based science: In what ways do these strategies apply to SOU-level operations? That is, to what extent—and how—might unit-level managerial strategies suggest entirely different or novel combinations of those deployed at the team-level? Meanwhile, what are the forms, types, and approaches to decision-making and governance in SOUs? To what extent do they feature corporatized, top-down decision-making in light of their entrepreneurial and resource goals? How are decisions made in contexts, such as centers and institutes, where there may not be permanent faculty lines? And what about the role and involvement of professional staff in this arena? There are numerous examples in the STEM arena of external stakeholders, private foundations, and donors and philanthropists establishing the designs for SOUs. To win their

187

“Authentic” Faculty and Academic Capitalism, supra note 11.

University as knowledge-based enterprise 129 resources is contingent on following specified models of organization. The NSF and its leaders and administrators have designed the template and financial incentives for the ERCs and I/UCRCs to meet scientific and resource goals.188 NSF centers vary from each other in their structure and functions but are stilled based on externally-envisioned concepts and ideas. The Whitaker Foundation incentivized institutions to create Biomedical Engineering departments. Whether or not the discipline constitutes a recognizable field, Whitaker-funded and non-Whitaker-funded institutions created departments. Likewise, some philanthropists want to endow a center or institute because, in their view, such organizational forms broaden the reach and potential effects of any resources won. Such influences raise questions about the extent to which university stakeholders are themselves able to shape academic structure on their campuses. And it calls attention to ways in which external and university stakeholders compile, analyze, share, and act on data regarding design decisions. D.

How Do Emerging Policy Contexts Influence Design?

The recent ascendance of populist-strands of Republican conservatism among US states and nationally suggests threats to particular types of science research and funding. Despite the resurgence of skepticism about science, facts, and reality, the STEM arena continues to receive attention and resources. But some STEM disciplines are prioritized over others. Amid Republican control of Congress and of many state governments, we can anticipate continued cuts to work on climate change/global warming and stem-cell research because such research is viewed as harmful to industry and religious freedoms respectively. In times of Democratic “leftist” control, there are constraints on funding to other scientific areas, such as those perceived detrimental to human health, the environment, and animals.189 More generally, a number of communities in the US support the anti-vaccine movement, acting upon skepticism of medical advice and evidence on the importance of vaccinating children. And denial of evolution persists. Broadly considered, innovation occurs from positions of advantage and disadvantage. If the hostile rhetoric in the policy environment toward particular types of science intensifies further, it may unwittingly catalyze bold changes for both the well-resourced and tenuously funded projects and investigators. The bifurcation, though, could potentially become stratifying to the point that innovative activity is concentrated among fewer and fewer areas of scientific research and education, collaborative teams, and universities. That is, any positive, marginal benefits to the broader system of innovation could be increasingly constrained based on the polarization of science by partisan preferences and funding opportunities. Faculty, SOUs, universities, selected ANPOs, and multi-sector partners are, in this arena, entwined within an ecology of knowledge-production and -exchange. Though federal and state policies do not always overlap or cohere, they suggest in recent years an increasing emphasis on the use of competitive markets to achieve effectiveness and efficiency across organizations and industries. But political processes, researchers and analysts suggest, powerfully influence the selection of winners and losers,190 and perhaps, in the years ahead, politics will forcefully

188

Bozeman & Boardman, supra note 80. Mark Zachary Taylor, The Politics of Innovation: Why Some Countries are Better than Others at Science and Technology (2016). 190 Taylor, supra note 75; Hearn, et al., supra note 155. 189

130 Research handbook on intellectual property and technology transfer shape institutional designs more than technical goals of effectiveness and efficiency. That these processes amplify disequilibria within and across a constellation of organizations and sectors, constraining science, research and policy attention is warranted.

6.

Policy advocacy and organizational change at the Association of University Technology Managers (AUTM) Christopher S Hayter and Jacob H Rooksby

I.

INTRODUCTION

Since its founding, the Association of University Technology Managers, Inc. (now preferentially called “AUTM” after a February 2018 rebranding) has played a critical role in national policy discussions relating to technology transfer, intellectual property (“IP”), federal research support, entrepreneurship, and innovation. The nonprofit association is comprised of individuals responsible for the management of technology portfolios and an increasing variety of related support programs. AUTM’s mission is “to support and advance academic technology transfer globally”1 by: (1) (2) (3) (4)

providing members with knowledge, tools, and training for professional development, helping stakeholders understand the value and impact of technology transfer, developing partnerships with other organizations to advance AUTM’s goals and impact, and improving the regional and national innovation environments within which technology transfer occurs, otherwise known as innovation or entrepreneurship ecosystems.2

Since its establishment, AUTM has employed various advocacy strategies, defined as activism intended to sway public opinion, increase awareness, and influence policymakers toward a specific action related to policies, laws, or practices in support of these objectives.3 The Society of University Patent Administrators (“SUPA”), AUTM’s predecessor, was established in 1974 by Howard Bremer, a patent administrator at the Wisconsin Alumni Research Foundation; Ralph Davis, of Purdue University; and Norman Latker, patent counsel at the National Institutes of Health (“NIH”).4 SUPA was established to advance the interests of university personnel responsible for administering institutional patent agreements (“IPAs”). SUPA is perhaps best known for its critical role in the establishment of the University and Small Business Patent Procedures Act of 1980, otherwise known as the Bayh-Dole Act.5

1

See autM, available at https://www.autm.net/autm-info/ (last visited Sept. 14, 2018). For a recent review of the academic entrepreneurship literature, see Christopher S. Hayter, Andrew Nelson, Stephanie Zayed, & Alan O’Connor, Conceptualizing Academic Entrepreneurship Ecosystems: A Review, Analysis and Extension of the Literature, 43(4) J. teCh. transfer 1039 (2018). 3 See Christopher Prentice & Jeffrey Brudney, Nonprofit Lobbying Strategy: Challenging or Championing the Conventional Wisdom?, 28(3) voluntas 935 (2017). 4 Recollections: Celebrating the History of AUTM and the Legacy of Bayh-Dole, AUTM (2004). 5 35 U.S.C. §§ 200–212 (2012). For information on SUPA and its role in the formation and passage of the Bayh-Dole Act, see Elizabeth Popp Berman, Why Did Universities Start Patenting: Institution-building and the Road to the Bayh-Dole Act, 38(6) soC. stud. sCI. 835 (2008). 2

131

132 Research handbook on intellectual property and technology transfer The Bayh-Dole Act was passed to promote the dissemination and commercialization of technologies stemming from university research and is credited with many positive social and economic development outcomes. Further, the Bayh-Dole Act was responsible for the rapid development of the technology transfer profession, if not its establishment. The Bayh-Dole Act was not without detractors, however. It was opposed when it was introduced and legislative attempts to unravel or roll back its provisions have been put forward periodically over the past 40 years.6 The advocacy efforts of SUPA, and later AUTM (the organization officially changed its name in 1990), have thus focused on countering threats to the Bayh-Dole Act and helping policymakers understand the legislation’s impact and value. Further, AUTM has taken positions on and advocated for a myriad of other policy issues related to research, technology transfer, IP, development, commercialization, and entrepreneurship. The changing nature of AUTM’s advocacy efforts also reflects the changing nature of technology transfer. Traditional responsibilities in the profession include management and oversight of technology transfer milestones governed by federal (and at times state) laws and policy. Milestones include technology disclosure, technology evaluation, filing of provisional and full patent applications, patent maintenance, and technology licensing and marketing to industry.7 Not only have these milestones changed substantially, but so have the duties of the positions responsible for achieving them. Technology managers now process material transfer agreements, manage copyrights and other types of IP, manage early-stage technology development and entrepreneurial seed funds, organize entrepreneurship support classes, and serve as industry liaisons, among other responsibilities. Within these changing technological, policy, and professional contexts, this Chapter seeks to contribute to the nonprofit management and policy literatures by understanding AUTM’s impact on relevant federal and state policy discussions. It does so by examining the evolution of advocacy within AUTM through interviews with former AUTM board members and presidents, individual experts, and representatives from other higher education associations to understand these organizational changes from a variety of perspectives. The Chapter is divided into three sections. First, it briefly reviews the nonprofit advocacy literature that motivates the Chapter. The second section establishes a framework for examining AUTM’s evolution and does so over the course of three developmental phases that we identify. The final section provides additional analyses and concludes.

II.

CONCEPTUALLY MOTIVATING THE CHAPTER

Nonprofit organizations employ various operational strategies in fulfillment of their missions, including capacity building, service delivery, public education, and advocacy. Conceptualizations of nonprofit strategy have evolved from a dichotomous contrast between service delivery and advocacy toward hybrid approaches that combine service provision,

6 David C. Mowery, Richard R. Nelson, Bhaven N. Sampat, & Arvids A. Ziedonis, Ivory Tower and Industrial Innovation: University-Industry Technology Transfer Before and After the Bayh-Dole Act in the United States (2004). 7 For a review of the technology transfer literature and milestones associated with the “traditional linear model,” see Samantha Bradley, Christopher S. Hayter, & Albert N. Link, Models and Methods of University Technology Transfer, 9(6) found. & trends entrepreneurshIp 571 (2013).

Policy advocacy and organizational change at the AUTM

133

capacity building, and advocacy.8 With regard to the distribution of functions, recent research finds that most (51%) nonprofits located in the US focus primarily on service delivery while only a small number of secular nonprofits (8%) focus on the combination of capacity building and advocacy, supplemented by public education and research.9 Further, advocacy is the primary mission for some, albeit few, nonprofits.10 In sum, most non-religious nonprofits do not engage in policy advocacy. Advocacy is defined as “activism in support of an idea or cause which is intended to increase awareness, influence public opinion, direct decision makers toward a solution, and change policies, laws, or practices.”11 Scholars recommend that investigations of nonprofit advocacy be conceptualized by relative level of strategic commitment: to what extent are nonprofits involved in advocacy-related activities?12 Scholars frame motivations for nonprofit advocacy activities in terms of their intent: do advocacy activities promote organizational benefit (e.g., government funding, regulation) or do they relate to larger societal benefits, or both?13 Other scholars focus on nonprofit lobbying as a specific advocacy strategy that attempts to influence legislation directly or indirectly, though these conceptual boundaries have not been well defined.14 Within their respective level of engagement, nonprofits can pursue insider or outsider strategies for advocacy. Insider strategies assume that a nonprofit has access to policymakers through direct lobbying, meeting with legislators, and invited testimony, often undertaken by experts or professionals. In contrast, outsider strategies include protests, grassroots mobilization, and public education, strategies usually undertaken by service recipients and volunteers, at times in a confrontational manner.15 Further, nonprofits build coalitions with other organizations to strengthen their ability to participate in substantive public policy discussion, which usually leads to increasing levels of political activity.16 Despite a robust literature that examines various aspects of nonprofit advocacy, few scholars have examined how advocacy strategy changes over time. Though not limited to advocacy strategies, Mark Hager reviews the literature pertaining to life cycles within nonprofit organizations. He finds that while nonprofit scholars have examined phenomena within specific stages of organizational life, they have yet to holistically conceptualize nonprofit life

8 See Gary D. Bass, David F. Arons, Kay Guinane, Matthew F. Carter, & Susan Rees, Seen but Not Heard: Strengthening Nonprofit Advocacy, 34(2) adMIn. soC. work 213 (2010); see also George E. Mitchell, The Strategic Orientations of US-based NGOs, 26 voluntas 1874 (2015). 9 Mitchell, supra note 8. 10 See Michael Almog-Bar & Hillel Schmid, Advocacy Activities of Nonprofit Human Service Organizations: A Critical Review, 43 nonprofIt & voluntary seCtor Q. 11 (2014). 11 See Prentice & Brudney, supra note 3, at 937–8. 12 See Curtis D. Child & Kirsten A. Grønbjerg, Nonprofit Advocacy Organizations: Their Characteristics and Activities, 88 soC. sCI. Q. 259 (2007). 13 See Linda Donaldson, Advocacy by Nonprofit Human Service Agencies, 15(3) J. CoMMunIty praC. 139 (2007). 14 Prentice & Brudney, supra note 3, at 938. 15 William Gormley, Jr. & Helen Cymrot, The Strategic Choices of Child Advocacy Groups, 35 nonprofIt & voluntary seCtor Q. 102 (2006). 16 Bass et al., supra note 8; see Rachel Fyall & Michael McGuire, Advocating for Policy Change in Nonprofit Coalitions, 44(6) nonprofIt & voluntary seCtor Q. 1274 (2015).

134 Research handbook on intellectual property and technology transfer cycles.17 This Chapter thus seeks to contribute to the nonprofit management literature as well as understand the role and influence of AUTM in related national policy discussions involving IP, technology transfer, and innovation. The Chapter does so by examining the evolution of advocacy within the context of organizational development and change at AUTM.

III.

ADVOCACY EVOLUTION AT AUTM OVER TIME

This section examines the evolution of AUTM’s approaches to advocacy over three distinct periods. In addition to reviewing the relevant nonprofit management and advocacy research above, we reviewed archival documents and AUTM publications before conducting 15 interviews in support of this Chapter. To aide in the selection of participants, we constructed from public sources a list of former AUTM board members, key personnel from other higher education associations, federal policymakers, and scholars who specialize in the study of technology transfer. From this list, we contacted individuals by email to schedule interviews. Interviews were conducted over the phone or in person, ranging from 30 minutes to two hours in duration; most lasted about 45 minutes. Interviews were recorded and transcribed. During the interviews, participants learned about the purpose of the study and were asked to describe AUTM’s advocacy strategies and how these strategies had evolved over time. Interviews were guided by a protocol that included open-ended questions to elicit a variety of responses, as well as specific follow-up questions to ensure various aspects of advocacy were captured. Specific points included the internal organization of advocacy efforts, the programmatic or policy focus of these efforts, the role of AUTM members within these efforts, and partnerships with other organizations to advance advocacy goals. We inductively coded and analyzed the resulting data. In addition to the results reported below, respondents helped us define the evolution of AUTM’s advocacy’s strategy in terms of three phases—1990–2002; 2003–2012; and 2013–2018—based on key organizational developments and other mediating factors. A.

Early Emergence (1974–2000)

1. Early advocacy by SUPA As mentioned, SUPA not only represented the interests of university personnel responsible for IPAs, but it also was instrumental in the composition and eventual passage of the Bayh-Dole Act. Several individuals, such as Howard Bremer, George Pickar, Earl Freise, and Larry Gilbert, among others, played critical roles in establishing SUPA.18 Early SUPA leaders quickly developed strategy to advocate for what Howard Bremer called a “uniform federal patent policy.”19 SUPA members did this through letter writing, visits to Washington, DC

17 Mark A. Hager, “Life Cycles of Nonprofit Organizations” in Global Encyclopedia of Public Administration, Public Policy, and Governance 1 (Ali Farazmand ed., 2016). 18 AUTM, supra note 4. 19 Id. at 11.

Policy advocacy and organizational change at the AUTM

135

to meet with congressional leaders, offering congressional testimony, and by writing amicus briefs for relevant cases such as Diamond v. Chakrabarty.20 According to the study participants, university presidents from SUPA member institutions— as well as higher education associations— were supportive and active champions for legislation that would eventually become the Bayh-Dole Act. In other words, high-level administration support from universities helped legitimize the advocacy efforts of SUPA members who were in relatively new, specialized functions. Early SUPA leaders were also empowered by their presidents to meet directly with policymakers, such as Ralph Davis’s meeting with Senator Birch Bayh (D-IN) and his aide, Joe Allen. During the meeting, Davis, along with Howard Bremer and Norman Latker, convinced the Senator to support their efforts.21 During their trips to Washington, with the help of congressional staff, SUPA members worked with small business advocacy groups who were, up to that point, skeptical of the idea of supporting comprehensive patent legislation. Their concern related to how a patent bill might benefit large companies, potentially hurting small businesses. Small business advocates believed that if large companies maintained ownership of technologies stemming from federally-funded research then small businesses would be unable to compete in related markets. The SUPA-small business coalition was important to the eventual passage of the Bayh-Dole Act. First, the Select Committee on Small Business had opposed prior efforts to pass similar patent legislation. Thus, SUPA’s work with small business advocates eventually led to legislative support within the committee. Another reason earlier legislative efforts failed was the inclusion of big business; many members of Congress viewed the inclusion of big business in patent legislation as corporate welfare. Thus, the removal of large business from proposed legislation was critical to its eventual passage.22 The Bayh-Dole Act was eventually signed into law by President Carter on December 12, 1980. After passage of the Bayh-Dole Act, SUPA representatives focused on developing other aspects of the association, such as training materials and skills workshops. In other words, the association focused more on expansion of the profession and related membership while placing relatively less emphasis on advocacy. Contemporary policy advocacy 2. The evolving context of technology transfer in the late 1980s and throughout the 1990s drove additional interest in advocacy at AUTM. In 1988, SUPA members voted to change the name of the association to AUTM to better reflect the changing composition and responsibilities of its members. Members were no longer simply filing IPAs; rather, they were managing portfolios of different technologies, ensuring compliance with the provisions of the Bayh-Dole Act, and marketing inventions to companies for potential licensing. The profession also rapidly expanded during the 1990s. Drawing on AUTM data, recent research shows that the number of individuals employed by universities in support of tech-

20 This case is credited with dramatically expanding the field of patent-eligible subject matter and spurring the creation of the biotechnology industry. See generally Jacob H. Rooksby, The Branding of the American Mind: How Universities Capture, Manage, and Monetize Intellectual Property and Why It Matters (2016). 21 Berman, supra note 5, at 856. 22 Id. at 855–6.

136 Research handbook on intellectual property and technology transfer nology transfer grew from a small group in 1991 (fewer than 200 licensing professionals) to nearly 500 licensing professionals and an additional 500 full-time employees by the year 2000.23 Growth of the profession was fueled by attention surrounding mounting revenues from licensing at some research universities; co-evolution of computing, software, the Internet, and dot-com ventures; and the growth and maturity of the venture capital industry.24 Finally, a new generation of technology transfer leaders emerged who were particularly interested in policy advocacy. According to study respondents, Janna Tom from the University of California System, Lita Nelson from the Massachusetts Institute of Technology, Kathy Ku from Stanford University, and Mark Crowell from the University of North Carolina-Chapel Hill represented institutions—including a large public university system—that received a significant portion of federal R&D spending, had long histories of working with industry, and were themselves located in vibrant entrepreneurial regions. Thus, these institutions faced unique or emergent policy changes perhaps not yet experienced by others. By taking a more active role in policy advocacy at AUTM, these leaders believed they could improve policy issues associated with their day-to-day responsibilities while also serving the needs of the broader technology transfer community. According to study respondents, AUTM’s policy-related activities were first motivated by proposed changes to agency rules affecting the university technology transfer community. When introduced, rules are published in the Federal Register with typically 30 to 60 days allotted for public comment. AUTM began to respond. For example, letters were drafted and circulated, then submitted in response to a proposal to change IP policy at the Advanced Technology Program (“ATP”), a now defunct program that supported the development of pre-competitive technologies within companies and among multi-company partnerships, as well as companies working in partnership with universities. The proposal allowed companies to jointly own IP generated during ATP-funded projects. AUTM argued that the proposed rule ran counter to the Bayh-Dole Act and therefore should not be implemented. Despite AUTM’s objections, the rule was enacted. AUTM representatives also sent letters to and met with key lawmakers expressing support for the reauthorization of the Small Business Innovation Research (SBIR) Program in 1997, advocating for an increase in the percentage of the federal R&D budget devoted to the program. This support was not only a result of AUTM maintaining cordial relations with small business advocacy organizations, but also reflected the increasing importance of SBIR awards to university spinoff companies. AUTM’s interest in ATP and SBIR programs was also emblematic of a growing interest among members in the overall ecosystem that supports technology transfer and academic entrepreneurship. AUTM’s action in response to the rule change at ATP, as well as SBIR’s authorization, established a contemporary precedent for member advocacy and reflected the ongoing formation of policy advocacy capabilities within the organization.

23

Bradley et al., supra note 7, at 577; see also AUTM STATT database. See Philip Phan & Don Siegel, The Effectiveness of University Technology Transfer: Lessons Learned, 2(2) found. & trends In entrepreneurshIp 77 (2006). 24

Policy advocacy and organizational change at the AUTM B.

137

Organizational Development (2000–2012)

According to respondents, AUTM rapidly developed its advocacy capabilities during the 2000s, reflecting a broader organizational emphasis on governance and operations. For example, since its establishment (as SUPA), the AUTM board played an integral role in the planning and operations of the annual meeting—from picking menus and negotiating hotel contracts, to lining up speakers and designing meeting content. When AUTM membership grew in the 1990s and the size of the meetings increased dramatically, the all-volunteer board was overwhelmed. It eventually decided to move toward a strategic board model whereby staff would focus on organizing activities, such as the annual meeting, and the board could focus on governance and strategy. In 2002, AUTM hired the Sherwood Group, a professional association management company, to manage its organizational affairs. This decision was a critical milestone in the evolution of AUTM. The Sherwood Group brought on Vicki Loise, an experienced association manager, to become the executive director of AUTM in 2005. While AUTM had always benefited from the generosity of technology transfer professionals volunteering their time for the association, the board believed that having a full-time executive director would help the organization grow. While the AUTM president changed every year and the board rotated periodically, Ms. Loise became a consistent presence. As one interviewee put it, “Vicki was the face of AUTM.” One of the executive director’s early responsibilities was to meet with well-known higher education associations, such as the Association of American Universities (“AAU”) and the Association of Public and Land-grant Universities (“APLU”). Establishing a strong relationship with these associations was important for several reasons. First, in contrast to the individual-based membership of AUTM, higher education associations represent individual institutions, with college presidents often involved in key association decisions. AUTM members were employed by these institutions and did not want to see their association take policy-related positions that conflicted with the administration of their home institution. Second, higher education associations employed full-time government relations personnel who not only had an excellent understanding of the political dynamics of Washington, but they also were involved in many of the same policy issues that were of interest to AUTM. Finally, according to interviewees, AUTM had previously considered establishing a permanent Washington, DC presence that would pursue its own advocacy agenda. Higher education association leaders were concerned that—given the volunteer-driven structure of AUTM—it lacked understanding of the complicated Washington political environment and misguided advocacy efforts might damage their reputation. Ms. Loise assured higher education associations that AUTM was interested in working closely with them to find areas of common policy interest. Further, AUTM benefitted from the goodwill of individuals who had an excellent understanding of the Washington, DC policy environment relating to higher education. For example, Bob Hardy, Director of Contracts and Intellectual Property Management for the Council on Governmental Relations, helped AUTM representatives understand the nuances associated with higher education associations, especially related to IP issues. Further, John Vaughn, Senior Fellow at AAU, not only helped AUTM understand the Washington political environment, but he also became the organization’s advocate, articulating its expertise and value in specific policy areas.

138 Research handbook on intellectual property and technology transfer With a full-time executive director focused on day-to-day issues, several active presidents and board members could focus on strategic issues. For example, Arundeep Pradhan (AUTM president in 2009), the board, and Ms. Loise created a strategic plan that focused on improving governance of the association’s strategies for professional development offerings and policy advocacy. In 2005, the board had asked the policy committee, including long-serving members such as Janna Tom and others, to take the lead in defining AUTM’s stance on specific issues. While the policy committee had played an important role in AUTM’s advocacy efforts, its efforts were now viewed as an important component of AUTM’s overall strategy. The mid-2000s were also significant in terms of IP policy-related events. First, the NIH was petitioning to exercise its march-in authority for Xalatan, a prescription glaucoma drug, the rights to which were owned by Pfizer, Inc. The drug was partially based on IP stemming from research conducted at Columbia University. The price of the drug was significantly higher in the US than in Canada or Europe and petitioners argued that this disparity was cause for march-in. While the NIH eventually refused to exercise its march-in authority, the event was nonetheless cause for concern at AUTM given that a successful march-in case could deter future industry licensing activity from universities. Second, 2005 marked the beginning of discussions regarding federal patent reform, efforts which eventually led to the passage of the America Invents Act (“AIA”) in 2011. The policy committee responded in two different ways. First, it sought to implement a proactive advocacy strategy instead of only reacting to policy developments. While the committee continued to write policy position letters and respond to proposed agency rules in the Federal Register, it placed increasing effort on establishing relationships with policymakers. This change was manifest in periodic visits to Washington, DC by AUTM leadership to meet key members of Congress and their staff. Further, AUTM began to sponsor and host events for congressional staff to improve their recognition of the association and improve their understanding of related policy issues. AUTM also sought to work with federal research funding agencies, especially the National Science Foundation (NSF), the NIH, and the National Institutes for Standards and Technology (NIST, the agency responsible for federal administration of the Bayh-Dole Act) to understand their rulemaking processes and more quickly provide information and feedback to proposed rules. Second, the policy committee viewed its increasingly close relationship with higher education associations as an excellent opportunity to improve AUTM’s advocacy impact. As discussed, Bob Hardy and John Vaughn were already helping AUTM leaders understand the political dynamics of Washington. The policy committee built upon these relationships as well as its expertise in IP to form and utilize coalitions in support of policy advocacy. Comprised of attorneys, technologists, and licensing practitioners, the policy committee was becoming more confident in its ability to provide other organizations with valuable technical advice—no small feat considering the complex issues associated with patent reform and other IP-related issues. The policy committee worked closely with higher education associations—first sharing policy statements and later working to create joint letters. AUTM also began working with the Biotechnology Innovation Organization (“BIO”) on broader IP-related matters. The biotechnology industry has its roots in research universities, benefiting from the patenting and licensing of university technologies. According to study participants, AUTM’s emphasis on proactively forging close relationships with other associations, including BIO, improved AUTM’s stature and advocacy-related

Policy advocacy and organizational change at the AUTM

139

impact. While AUTM worked in coalition to express support for the first America Competes Act of 2007, respondents viewed the 2008 reauthorization of the SBIR program as a defining moment for AUTM’s work within the context of a coalition. Working relationships solidified, providing the basis for continuing joint work on the AIA until its passage in 2011. AUTM also became better equipped to defend legislative efforts that sought to repeal or weaken the Bayh-Dole Act. Finally, AUTM undertook several initiatives on its own, including international outreach to technology transfer professionals across the globe and efforts to support the passage of Bayh-Dole-Act-like legislation in jurisdictions outside the US. Coalitions with other organizations and technology thought leaders became a critical mechanism supporting AUTM’s advocacy efforts during these years; however, these coalitions did not always perfectly align on all issues. For example, in the runup to passage of the AIA, WARF—the oldest and arguably the most venerated technology transfer office (“TTO”) in the nation—broke rank from AUTM and the DC-based higher education associations in coming out against passage of the AIA, fearing that it would ultimately prove harmful to universities. Moreover, while some urged Stanford University to pursue its patent ownership dispute with Roche Molecular Systems, Inc. all the way to the US Supreme Court—which it did, and lost, in 2011—others within the technology transfer community, including Howard Bremer himself, felt Stanford’s case was weak and feared that the case could dismantle fundamental working understandings of the Bayh-Dole Act. AUTM was placed in the position of mediating these tensions, not always to the satisfaction of all members. Indeed, on a basic level, leadership of the organization hewed toward members from TTOs at large universities, as they vastly outnumbered members from single-person or small TTOs. This dynamic alone may account for much of the organization’s policy positions, which tend to align naturally with the desires of large university presidents from whom the higher education associations take their instructions. But as one interviewee—a longtime AUTM member—put it, “Consensus does not mean 100%.” Beyond the policy committee, AUTM also placed emphasis on understanding better the impact of technology transfer and conveying these impacts to various stakeholders. Around 2010, the board launched a “new metrics” subcommittee, led by Dana Bostrom, Kevin Cullen, and Ashley Stevens, charged with expanding AUTM’s data collection efforts. AUTM had long collected technology transfer-related data through its annual AUTM Licensing Activity Survey of members, the results of which are kept in an online database (called STATT) accessible only to AUTM members. The survey asks about patenting activity, licensing revenues, TTO characteristics, and number of university spinoff companies established each year. Technology transfer practitioners and policymakers focused on relative differences in revenues, patents, and spinoff companies among research universities, while scholars began to analyze and publish related academic studies in the late 1990s using these data. However, the unintended consequence of these data analyses was to turn focus away from other areas of technology transfer impact.25 In response, the new metrics subcommittee proposed that AUTM augment the annual survey with data collection that supported an institutional economic engagement index.26 Though the proposal was not implemented, it highlighted the difficulty

25 See Bradley et al., supra note 7 (discussing how academic scholars similarly focused on technology transfer as a linear patent-centric heuristic, neglecting alternative paths and conceptualizations). 26 See AUTM, available at https://www.autm.net/autm-info/about-tech-transfer/about-technology -transfer/new-metrics/ (last visited Sept. 16, 2018).

140 Research handbook on intellectual property and technology transfer of measuring the impact of technology transfer and led to a similar initiative at APLU that focused on the overall impact of research universities. AUTM also established the Better World project in 2005 to promote public understanding of the impact of academic research and technology transfer.27 The project initially produced an annual report of case studies focusing on specific successful technologies and their “real world” and market impacts. Beginning in 2012, the project transitioned to an online database format. The site not only features monthly stories, but it also enables users to search the story database and generate customized reports focusing on specific types of impact as well as where those impacts occur. In sum, both efforts sought to get stakeholders away from using disclosures, patents, licenses, spinoffs, and revenue as the primary or even the only metrics of success. C.

Institutionalization and Identity Refinement (2015–2018)

Beginning in 2014, two concurrent events provided a catalyst for additional governance changes at AUTM. First, Vicki Loise was promoted within the Sherwood Group (which merged with and has been called Kellen Company since 2015) to a vice president level, a position that does not typically include association management responsibilities. Further, AUTM’s contract with the Sherwood Group ended. After some negotiations, the AUTM board decided it would use the occasion of the contract expiration to recruit the organization’s first full-time, non-Sherwood staff hire, engaging Sherwood only for more routine, non-policy-oriented association management responsibilities, such as billing and accounting. After an initial hire of a top executive who lasted less than a year in the position, the AUTM board subsequently hired Stephen Susalka. With a PhD in neuroscience and over a decade of experience in university technology licensing at Wake Forest University and the University of Virginia, Susalka was appointed AUTM’s executive director in 2015 and made CEO of the organization in 2017. He reports directly to the AUTM board. Changes in AUTM governance are occurring while the technology transfer profession is itself evolving rapidly. Technology managers often report to university vice presidents for research and may manage a number of activities, including sponsored research, copyrights, entrepreneurial seed funds, and entrepreneurship support programs. AUTM leadership has devoted considerable attention to understanding what these developments mean for the association. In 2017, AUTM hired a branding agency to conduct market research and organize focus groups among AUTM members. The analysis found that the name AUTM did not represent the association’s increasingly diverse membership, especially non-university members (i.e., nonprofit and corporate members).28 In response to the analysis and several suggestions from the branding agency, the AUTM board proposed a name change that would entail rebranding the organization from being called AUTM to the Association for Research and Innovation Advancement (“ARIA”). The board asked the membership to vote on the proposed name change at the end of 2017.

27 See Better World Project, AUTM, available at http://www.betterworldproject.org/ (last visited Sept. 16, 2018). 28 See Branding FAQ, AUTM, available at https://www.autm.net/AUTMMain/media/Elections/ Documents/Branding_FAQ_ Website.pdf (last visited Sept. 16, 2018).

Policy advocacy and organizational change at the AUTM

141

In October of 2017, eighteen past presidents of AUTM issued a letter asking the membership to vote against the proposed name change.29 The letter argued that the proposed name lacked brand resonance, that the name ARIA was similar to or the same as other organizations (e.g., a Las Vegas hotel), and that the change would destroy AUTM’s valuable brand identity, which is centered on the capabilities and credibility of its members (i.e., university technology managers). The letter also argued that emphasis should be placed on visually rebranding AUTM and offering new subject-matter, innovation-oriented content to its members, such as start-up business development courses. The proposed name change to ARIA was voted down, with 71 percent of participating members voting against the name change. Around this time, in February 2018, the organization underwent a marketing makeover with the assistance of Bates Creative, a “brand experience agency” based in Washington, DC.30 In addition to unveiling a new logo, website, and tagline (“Transforming Ideas into Opportunities”), the organization announced it would be shedding its formal title of Association of University Technology Managers in favor of being preferentially known as AUTM. As AUTM’s Chief Marketing Officer later told us, the name AUTM “better reflects the professional diversity of our membership. The face of technology transfer is evolving and so are we.” Several study participants, however, had expressed disappointment with the effort to change AUTM’s name, feeling that the time and energy might have been better spent on pursuing strategic issues more beneficial to the organization. But the event also reflected broader tensions over AUTM’s evolving identity. Some objected to the organization’s recent emphasis on building international memberships and alliances and attracting non-university members to the annual meeting. As the organization has grown, it invariably has become less personal in the eyes of some. Other interviewees expressed frustration that they learn more from their university’s government relations staff about IP-related issues than they do from AUTM. While some would prefer that AUTM return to its historic emphasis on the nuts and bolts of patenting and licensing, as was the case with SUPA, the professionalization of the organization means that AUTM can pursue policy advocacy with more continuity and planning. And although the effort to change the organization’s name to ARIA ultimately was unsuccessful, there is no denying that AUTM’s current organizational structure enables it to participate more broadly and consistently in national and even international dialogues on the future of innovation. Gone are the days of AUTM being, as one interviewee put it, “kind of reluctant to take policy positions.” With an active, standing policy committee, a full-time CEO and marketing officer, and a longtime organizational management company in place to handle routine organizational concerns, AUTM has found its voice. The question remains, as demonstrated by efforts to change the organization’s name, what issues will that voice choose to speak on in the future, and how loudly and unified will it speak?

29 See From the Director: ARIA or AUTM? Why Past Presidents Say Keep AUTM, eMory U., available at https://scholarblogs.emory.edu/techtransfer/2017/10/from-the-director-aria-or-autm-why-past -presidents-say-keep-autm/ (last visited Sept. 16, 2018). 30 See https://batescreative.com/case-study/autm/ (last visited Oct. 17, 2019).

142 Research handbook on intellectual property and technology transfer

IV.

CONCLUSION

The story of AUTM is the story of organizational change in an industry that has matured dramatically since its effective inception 40 years ago with the passage of the Bayh-Dole Act. In this Chapter, we have identified three distinct periods in the organization’s history and described how they impacted the organization’s ability to respond to, or even get out ahead of, important IP and technology transfer policy issues during these eras. Every organization faces challenges as it grows and transfers leadership to the next generation. These changes impact an organization’s ability to advocate for itself and stake policy positions that are informed and consistent. AUTM has developed the organizational capabilities to better govern itself and, as discussed in this Chapter, construct and implement an advocacy strategy. What is less certain, however, is how the professional practice of technology transfer will evolve as major financial and structural changes continue to occur in higher education into the 21st Century. The roles of technology transfer professionals are expanding to include applications for sponsored research, early-stage venture finance, industry outreach, and entrepreneurship education and services, among other responsibilities. What do these changes mean for a profession that has, through most of its existence, focused on receiving disclosures, filing patent applications, and executing licenses? Perhaps more important to this Chapter, what do these changes portend for AUTM as a nonprofit, individual membership association engaged in policy advocacy? Regardless of potential answers to those questions, we conclude that AUTM’s ability to advocate for its membership seems more firmly rooted than ever before. Uncertain is what the fortification of the organization’s functioning and identity will mean for the future of laws and policies that impact innovation and technology transfer. After all, these complicated issues now draw vocal and well-organized advocates from many sectors outside of higher education, including governmental, nongovernmental, and corporate worlds. Whatever the outcome of those future debates, however, we predict that the strength and credibility of AUTM’s voice in the dialogues will only become richer as technology transfer comes of age.

7.

Conflicts of interest and academic research Jorge L Contreras and Marc Daniel Rinehart

I.

INTRODUCTION

On September 8, 2018, the front page of the New York Times bore the headline “Top Cancer Researcher Fails to Disclose Corporate Financial Ties in Major Research Journals.”1 The story revealed that world-renowned oncologist José Baselga, Chief Medical Officer of Memorial Sloan Kettering Cancer Center in New York, received more than $3 million from pharmaceutical and biotechnology firms between 2013 and 2017, and failed to disclose these payments in dozens of articles and speeches commenting, in many cases, on drugs being developed by those very companies. Dr. Baselga resigned a few days after these revelations were published. The Baselga incident was only the most recent in a long series of prominent controversies surrounding the influence that corporate funders may exert over academic research. Corporate funding of academic research began quietly in the 1950s when the tobacco industry sought to generate academic debate over findings that linked smoking with cancer.2 The results of this campaign were not generally known, however, until much later. In 1974, public concern over corporate support for academic research began to emerge when Monsanto was reported to have paid Harvard Medical School $23 million to conduct research on organ development.3 Other multi-million dollar transactions followed as corporations increasingly sought to benefit from research conducted at leading academic institutions.4 Alongside these institutional transactions, news began to spread that individual researchers could profit—sometimes substantially—from their academic research.5 Growing ties between industry and academia generated significant public opposition in the late 1990s and early 2000s, when student protests at institutions including University of

1 Charles Ornstein & Katie Thomas, “Top Cancer Researcher Fails to Disclose Corporate Financial Ties in Major Research Journals”, N.Y. tIMes, Sep. 8, 2018, A1. 2 See Allan M. Brandt, Inventing Conflicts of Interest: A History of Tobacco Industry Tactics, 102 aM. J. pub. health, 63, 67 (2012) (“the tobacco industry managed to sustain the widespread perception of an active and highly contested scientific controversy into the 1960s despite overwhelming evidence and scientific consensus that smoking caused serious disease …”). 3 Edward B. Fiske, “Monsanto Research Pact Aims to Cut Academic Controversy”, n.y. tIMes, Jun. 4, 1982, at A21. 4 See Fiske, supra note 3; Jennifer Washburn, University Inc.—The Corporate Corruption of Higher Education 4–5 (2005). 5 See Sheldon Krimsky, Science in the Private Interest—Has the Lure of Profits Corrupted Biomedical Research? 110 (pbk. ed., 2004) (finding that between 1985 and 1988, 31% of biology faculty at MIT had ties to a biotechnology company, with 20% at Harvard and Stanford); Eliot Marshall, When Commerce and Academe Collide, 248 sCIenCe 152 (1990).

143

144 Research handbook on intellectual property and technology transfer California Berkeley,6 University of California at Davis,7 and Yale Medical School8 led to the revision of significant corporate funding arrangements. Consumer advocate Ralph Nader expressed the fears of many in 2004 when he wrote: Academic science, with its custom of open exchange, its gift relationships, its willingness to provide expert testimony that speaks truth to power, its serendipitous curiosity and its nonproprietary legacy to the next generation of student-scientists, differs significantly from corporate science, which is ridden with trade secrets, profit-determined selection of research, and awesome political power to get its way, whether by domination or servility to its payers.9

A few years later, journalist Dan Greenberg, in his influential book Science for Sale: The Perils, Rewards, and Delusions of Campus Capitalism, posed the following question after interviewing researchers, university administrators and academic technology transfer officials across the country: Have today’s commercial values contaminated academic research, diverting it from socially beneficial goals to mercenary service on behalf of profit-seeking corporate interests? What are the gains and losses in the visibly tightening linkage of science and Mammon, and to whose benefit and whose detriment? Can academic institutions, with their insatiable appetite for money, reap financial profits from their production of valuable knowledge without damage to the soul of science and the public?10

In 2008, a string of highly-publicized failures of academic investigators to disclose significant industry compensation led to a Congressional investigation of academic conflicts of interest.11 A 2012 report by the non-partisan Congressional Research Service also recognized the threat of conflicts of interest arising from increased patenting and technology commercialization by universities.12 In response to these concerns, the academic research enterprise—researchers, institutions, funders and scientific journals—have given renewed attention to potential conflicts of interest. Many of these groups have adopted formal policies and procedures concerning conflicts of interest. In this Chapter, we summarize some of the procedures implemented by these different stakeholders in the research enterprise to reduce the impact of conflicts of interest.13

6

See Washburn, supra note 4, at 7–8 (describing controversy over corporate sponsorship by Novartis). 7 See id. at 95 (controversy over support of agricultural research by Allied Chemical). 8 See Donald G. McNeil, Jr., “Yale Pressed to Help Cut Drug Costs in Africa”, N.Y. tIMes, Mar. 12, 2001, A3 (describing student protests over agreement with Bristol-Myers Squibb regarding AIDS drug d4T). 9 Ralph Nader, Foreword in Krimsky, supra note 5, at xiii, xiv. 10 Daniel S. Greenberg, Science for Sale: The Perils, Rewards, and Delusions of Campus Capitalism 2 (2007). 11 See Gardiner Harris, “Top Psychiatrist Didn’t Report Drug Makers’ Pay”, n.y. tIMes, Oct. 3, 2008. 12 Wendy H. Schacht, The Bayh-Dole Act: Selected Issues in Patent Policy and Commercialization of Technology, Cong. researCh serv. 17–19 (Dec. 3, 2012). 13 Though our focus in this Chapter is on COI policies and regulations in the United States, issues surrounding COI are by no means limited to the United States. See, e.g., Mark A. Rodwin, Conflicts of Interest and the Future of Medicine: the United States, France and Japan 6 (2011) (“Conflicts of interest are endemic in private practice in countries with very different medical, legal and political systems”).

Conflicts of interest and academic research 145

II.

WHAT ARE CONFLICTS OF INTEREST?

Many definitions of “conflict of interest” (“COI”) exist.14 As defined in an influential 2009 report by the Institute of Medicine, a conflict constitutes any “set of circumstances that creates a risk that professional judgment or actions regarding a primary interest will be unduly influenced by a secondary interest.”15 Thus, a familial or romantic relationship, a personal animus, a professional friendship or rivalry, a religious belief, and even an unconscious bias, could all affect research outcomes and be considered conflicts of interest.16 However, for a host of practical reasons, formal conflict of interest policies typically focus on financial relationships and the potential for financial gain.17 Thus, financial conflicts of interest will be the focus of this Chapter.18 This being said, there is little consensus regarding the size of a financial interest that gives rise to a conflict of interest. As discussed below, different federal requirements at different times have required disclosure of financial interests in excess of $10,000, $5,000 and $200. While some academic investigators and clinicians receive tens of millions of dollars in outside compensation,19 others receive mere token amounts, free meals or travel reimbursement. Nevertheless, studies have shown that even relatively small levels of compensation can have an impact on research results and reporting.20 As suggested in section I, financial conflicts of interest can occur both at the individual level and at the institutional level. The policies described below are directed primarily to individual conflicts of interest. Identifying and addressing institutional conflicts of interest has proven far more difficult,21 though some policies have taken steps in this direction. Before addressing specific policies, it is worth considering how financial conflicts of interest can affect research. As an investigator plans a study, including the formulation of a hypothesis, the selection of methodologies and the design of study protocols, a significant interest could

14

See Mark A. Rodwin, Conflicts Of Interest In Medicine: Should We Contract, Conserve, or Expand the Traditional Definition and Scope Of Regulation?, 22 J. health l. & pol’y 158 (forthcoming 2019) (discussing and critiquing definitional inconsistencies). 15 Inst. of Med. (IOM), Conflict of Interest in Medical Research, Education and Practice 45–6 (Bernard Lo & Marilyn J. Field, eds. 2009) [hereinafter IOM 2009]. 16 See, e.g., id. at 46–7, Inst. of Med. (IOM), Conflict of Interest and Medical Innovation: Ensuring Integrity While Facilitating Innovation in Medical Research—Workshop Summary 8 (Sarah H. Beachy, Adam C. Berger, & Steve Olson, Rapporteurs, 2014) [hereinafter IOM 2014], Lisa Bero, Addressing Bias and Conflict of Interest Among Biomedical Researchers, 317 J. aM. Med. ass’n 1723 (2017). 17 See IOM 2009, supra note 15, at 47; Bero, supra note 16, at 1723 (meta-analyses have shown that financial conflicts of interest have a measurable effect on research outcomes). 18 But see Ross E. McKinney & Heather H. Pierce, Strategies for Addressing a Broader Definition of Conflicts of Interest, 317 J. aM. Med. ass’n 1727 (2017) (arguing for policies to address non-financial conflicts of interest). 19 See Bernard Lo & Deborah Grady, Payments to Physicians: Does the Amount of Money Make a Difference?, 317 J. aM. Med. ass’n 1719 (2017) (“The 5 physicians who received the largest total payments in the Open Payments database from 2013 through 2015 each received more than $28 million”). 20 See id. at 1720 (even meals costing between $12 and $18 have been shown to affect physician prescribing practices); Harvey V. Fineberg, Conflict of Interest—Why Does it Matter?, 317 J. aM. Med. ass’n 1717 (2017) (even small gifts such as return address labels in charitable solicitations and stamped return envelopes can significantly increase physician return rates). 21 See Greenberg, supra note 10, at 160 (“institutional conflicts of interest presented a tangle of problems that had heretofore been ignored”).

146 Research handbook on intellectual property and technology transfer influence the investigator’s choices. Because the role and responsibilities of the investigator vary with the type of research being conducted, the potential for bias in experimental design created by a conflict of interest varies as well. For example, an industry sponsor often designs experimental protocols with little to no input from investigators conducting the research in order to standardize inter-institutional research protocols and to meet criteria assigned by the Food and Drug Administration (“FDA”). In such cases, even an investigator with a significant financial interest related to the industry sponsor cannot impart bias on the design of the study.22 In such cases, even direct remuneration for consulting services does not appear to represent a direct and significant relationship with the investigator’s role and responsibility in the design of the study—and a financial conflict of interest would likely not exist. But if an investigator independently designed a study that compared the efficacy of two FDA-approved drugs and the investigator received direct remuneration from one of the manufacturers of the drugs, a potential for bias in the experimental design of the study would exist. The investigator would have a financial conflict of interest due to her outside relationship with the manufacturer as a result of her role in experimental design. Likewise, an investigator who does not form the hypothesis, design the experiments, or assign the methodology used in a research project, but who takes an active role in the conduct of the research may still be influenced by a significant financial interest. Such influence could manifest itself through myriad activities by the investigator during the research study. For example, an investigator working on a research study involving patients may determine exclusion and inclusion criteria for enrolling patients or may be involved in consent decisions. If the research study has an industry sponsor, the investigator also may recruit patients, conduct medical evaluations, and in some circumstances, analyze the resultant data. Each of these activities might be affected adversely if there is a conflict of interest. Financial conflicts of interest might also affect the conduct of a study through data analysis. An investigator might “cherry-pick” data by excluding data points identified as outliers or choose a statistical test for analysis that yields desired results.23 Finally, a biased investigator might draw overstated or unfounded conclusions, fail to disclose funding sources or financial 22 This is assuming that the researcher’s financial interest in the company was not related to assisting in designing the study for FDA approval. Does FDA criteria allow for some flexibility in experimental design that could be unduly influenced by bias created by a financial interest in the company? 23 It may seem intuitive that statistical analysis of data sets should be rigorous and be selected based on the design of the experiment. However, the field of statistics is becoming ever more complex as new technologies lead to more complex experimental designs with more variables to analyze. This has resulted in ambiguity in assigning proper statistical analyses to more complex experiments. Silberzahn et al. have shown that even experts in the field of statistics are vulnerable to bias and uncertainty when deciding best practices. In their experiment they gave several groups of statisticians the same data set to analyze using their own discretion in selecting the proper statistical evaluations. The distribution of the groups’ results was striking with some groups finding statistical significance to support the hypothesis while other groups found the results did not significantly support the hypothesis. In a twist, the researchers then had the groups share their methods and opened discussion so that all the groups considered the others’ approach. The groups, now armed with new approaches and considerations, re-evaluated the data with what should have been more similar statistical tests. The results of the second analysis showed a similar distribution and variability as the first analysis. The results of this study suggest that even without undue bias in the form of external financial interests, devising proper studies with appropriate analyses can prove difficult. See R. Silberzahn et al., Many Analysts, One Data Set: Making Transparent How Variations in Analytic Choices Affect Results, 1 advanCes Methods & praCtICes psyChol. sCI. 337 (2018).

Conflicts of interest and academic research 147 interests, fail to report results not supported by the significant financial interests, or fail to report null results. In each of these instances, the research in question would be undermined by these decisions. A final, yet critical, area in which conflicts of interest can affect research arises with respect to the treatment of human subjects. Human subjects research is heavily regulated in the United States under a variety of administrative schema including the so-called Common Rule, which imposes requirements on the conduct of federally-funded human subjects research.24 There has long been concern that researchers with financial conflicts of interest may make decisions regarding human subjects that are compromised, thus imposing risks to individual health and welfare.25 Accordingly, policies relating to conflicts of interest are also intended to protect human subjects in research. As described above, many conflicts of interests can bias research outcomes or impose risks on human subjects. However, even conflicts that do not have any tangible effect on research can give an appearance of bias. In many cases, this appearance of bias can be just as damaging to an investigator’s or an institution’s reputation as actual bias. Thus, as McCoy and Emanuel explain, there are no “potential” conflicts of interest.26 Rather, “a COI describes a situation in which there is a risk of bias and resulting harm, not a situation in which bias or harm necessarily occurs. Thus, a situation marked by risk of bias from a secondary interest is no less a COI because it does not result in bias or harm.” For these reasons, the policies discussed below address conflicts of interest whether or not bias or harm can be proved.

III.

GOVERNMENTAL CONFLICT OF INTEREST POLICIES

A.

Individual Conflicts of Interest

The first formal US agency rules addressing conflicts of interest were adopted by the National Science Foundation (“NSF”)27 and the Department of Health and Human Services (“HHS”) through its Public Health Service (“PHS”)28 almost concurrently in 1995. These rules applied to each institution receiving research funding from the agency (grantees). These early rules established an approach to dealing with conflicts of interest based primarily on disclosure, an approach that has persisted through the present day. In the case of PHS (including the National Institutes of Health (“NIH”)), a grantee institution had to require each investigator planning to participate in PHS-funded research to disclose to a designed institutional official any “significant financial interest” that “would reasonably appear to be affected by the research” or “in entities whose financial interests would reasonably appear to be affected by the research.”29 For this purpose, a “significant financial interest” 24

Federal Policy for the Protection of Human Subjects, 45 C.F.R. Part 46. See, e.g., U.S. Dept. Health & Human Serv., Financial Conflict of Interest—HHS Guidance 2004, available at https://www.hhs.gov/ohrp/regulations-and-policy/guidance/financial-conflict-of-interest/ index.html (last visited Apr. 3, 2019). 26 Matthew S. McCoy & Ezekiel J. Emanuel, Why There Are No “Potential” Conflicts of Interest, 317 J. aM. Med. ass’n 1721 (2017). 27 Investigator Financial Disclosure Policy, 60 Fed. Reg. 35820 (1995). 28 Objectivity in Research, 60 Fed. Reg. 35810 (1995), codified at 42 CFR Part 50, Subpart F (1995). 29 45 CFR Part 50, Subpart F, § 50.604(a)–(c) (1995). 25

148 Research handbook on intellectual property and technology transfer was anything of monetary value (including salary, consulting fees and honoraria), equity interests and intellectual property (“IP”) rights held by the investigator or members of his or her immediate family valued, in the aggregate, at more than $10,000 or representing more than 5% of the value of a single entity.30 The grantee was required to notify PHS of the existence of any such conflict (but not the nature of the interest or other details) and to assure PHS that the interest was “managed, reduced or eliminated.”31 In addition, the grantee was required to disclose specific information regarding the interest to PHS upon request.32 These rules, which left the review and administration of conflicts of interest almost entirely to grantee institutions, was viewed by critics as “well-intentioned but soft.”33 Moreover, the pace of industry support of academic research continued to increase. One study found that, from 1994 to 2003, private sector support of biomedical research in the US increased from $37.1 billion to $94.3 billion.34 Likewise, relationships between individual academic researchers and industry increased from 28% in 1996 to 52.8% in 2007.35 In addition, news stories continued to break regarding egregious instances of academic misfeasance and failure to disclose major corporate research funding.36 As a result, PHS revised its COI policy in 2011 after a 60-day public review and comment period.37 The 2011 rules made a number of changes to PHS’s 1995 COI rules. First, the threshold for a “significant financial interest” was lowered from $10,000 to $5,000.38 Second, the term “financial conflict of interest” was defined as “a significant financial interest that could directly and significantly affect the design, conduct, or reporting of PHS-funded research.”39 Third and most importantly, the 2011 rules require that investigators seeking PHS funding must disclose to their institution all significant financial interests reasonably related to their institutional responsibilities and that grantee institutions take measures to identify and manage conflicts of interest through a variety of means including: public disclosure of the COI, disclosure of the COI to individual human subjects, appointment of an independent monitor capable of protecting the research against bias, modification of the research plan or research team, reduction or elimination of the COI, or disqualification of conflicted personnel from the research project.40 The grantee institution is also required to disclose the COI to PHS and

30

Id. at § 50.603. Id. at § 50.604(g)(2). 32 Id. at § 50.604(g)(3). 33 Greenberg, supra note 10, at 160. 34 Hamilton Moses, Ray Dorsey, & David H.M. Matheson, Financial Anatomy of Biomedical Research, 294 J. aM. Med. ass’n 1333 (2005). 35 David Blumenthal, et al., Participation of Life-Science Faculty in Research Relationships with Industry, 335 N engl. J. Med. 1734 (1996); D.E. Zinner, et al., Participation Of Academic Scientists In Relationships With Industry, 28 health affaIrs 1814 (2009). 36 See, e.g., Sara Reardon, NIH Disclosure Rules Falter, 525 nature 300 (2015) (explaining that a 2008 U.S. Senate investigation revealed that an Emory University psychiatric researcher failed to disclose at least $1.2 million that he had received from drug companies). 37 Responsibility of Applicants for Promoting Objectivity in Research for which Public Health Service Funding is Sought and Responsible Prospective Contractors, 76 Fed. Reg. 53256 (2011), codified at 42 CFR Part 50 (2011). 38 45 CFR Part 50, Subpart F, § 50.603 (2011). 39 Id. 40 Id. at § 50.605(a)(1). 31

Conflicts of interest and academic research 149 certify that it has implemented a suitable management plan.41 The specific impact of the 2011 policy on some institutional COI policies is discussed in greater detail in section III, below. If an institution discovers that an investigator has failed to disclose a COI as required under the PHS rules, and that failure “appears to have biased the design, conduct, or reporting” of the relevant research, the institution is required to report the failure to the relevant PHS entity.42 The relevant PHS entity may also review the institution’s relevant records to assess the potential for COI-based bias and to verify that an investigator has complied with any management plan imposed to address a COI.43 In each of these cases, the PHS entity may impose specific corrective or remedial measures on the institution and investigator.44 In severe cases, additional federal remedies are available, including debarment and suspension of the institution and/or investigator from further PHS grant funding, and liability under civil fraud regulations.45 B.

Institutional Conflicts of Interest

Interestingly, neither the 1995 nor the 2011 PHS rules address institutional conflicts of interest. In the public comment solicitation preceding the adoption of the 1995 rules, PHS asked members of the public to comment on whether the new regulations should address institutional COI.46 More than 100 comments were received.47 According to PHS, “those addressing the issue were nearly unanimous in concluding that the regulations should not address the institutional conflict issue because of the need to carefully consider that issue through a separate process.” 48 PHS agreed and the 1995 rules contained no references to institutional COI. In the late 1990s, several highly-publicized incidents occurred involving federally-funded academic research programs, including the death of a teenage volunteer in a gene therapy trial at University of Pennsylvania that was tainted by significant corporate ties.49 These incidents reinvigorated federal efforts to monitor and control institutional conflicts of interest. In 2001, the Office of Human Research Protection (“OHRP”) published a Draft Interim Guidance as a first step toward updating the 1995 COI rules.50 This document was authored by OHRP’s newly appointed head, Dr. Greg Koski, former director of human research affairs at the Partners HealthCare system and a Harvard Medical School professor.51 Koski’s draft included the following controversial provisions:

41

Id. at § 50.605(b). Id. at § 50.606(a). 43 Id. at § 50.606(b). 44 Id. at § 50.606(a). 45 Id. at § 50.607 (citing other regulations). 46 60 Fed. Reg. 35810 at 35813 47 Id. at 35810. 48 Id. at 35813. 49 See Greenberg, supra note 10, at 103–5 (discussing Penn incident) and 149–50 (discussing shutdown of Duke University studies for procedural violations). 50 Off. Human Research Protection, Draft Interim Guidance—Financial Relationships in Clinical Research: Issues for Institutions, Clinical Investigators, and IRBs to Consider When Dealing with Issues of Financial Interests and Human Subject Protection (Jan. 10, 2001), available at http://ccnmtl.columbia .edu/projects/rcr/rcr_conflicts/misc/Ref/OHRP_CoI.pdf (last visited Oct. 17, 2019). 51 Greenberg, supra note 10, at 154. 42

150 Research handbook on intellectual property and technology transfer 1.6 While institutions clearly need to have policies and procedures for managing conflicts of interest among their employees, they should not lose sight of the need to manage their own conflicts of interest as well. Increasingly, academic institutions and corporate entities are entering into agreements that are mutually beneficial, and which may also bring the institution’s interests into direct conflict with those of research participants. For example, an institution may accept a principal equity interest in a biotechnology company as part of a cooperative endeavor to develop a new medical device. Clearly, in such a situation, both the institution and the corporate partner would stand to gain financially if the device proves to be safe and effective. Accordingly, the institution should carefully consider whether a clinical trial to evaluate safety and efficacy should be performed at that site, and if it should, what special protections would be needed. The financial interest of the institution in the successful outcome of the trial could directly influence the conduct of the trial, including enrollment of subjects, adverse event reporting or evaluation of efficacy data. In such cases, the integrity of the research, as well as the integrity of the institution and its corporate partner, and the well-being of the research participants, may be best protected by having the clinical trial performed and evaluated by independent investigators at sites that do not have a financial stake in the outcome of the trial, or carried out at the institution but with special safeguards to maximally protect the scientific integrity of the study and the research participants. 1.7 When institutions consider entering into such business agreements, they should consider establishing an independent advisory and oversight committee (institutional conflicts of interest committee), if one does not already exist, to determine whether the financial arrangements pose a conflict of interest, and if so, how those conflicts should be managed. 1.8 Any financial relationships that the institution has with the commercial sponsor of a study should be documented and the specific relationships submitted to the Chair/Staff of the IRB as described above. Items to be identified include: any equity interest in the commercial sponsor; any “up front” payments to the Institution beyond those payments directly applicable to carrying out a particular protocol; any funds given to the Institution (or an entity within the Institution, e.g. an Institute); any equity ownership in the commercial sponsor that was transferred to the Institution, including the percentage ownership of any patents related to articles under study in the protocol; any royalties; any licenses granted to the commercial sponsor by the Institution; whether or not the Institution stands to gain financially if the study shows the “article” to be successful for its proposed use.

Not surprisingly, Dr. Koski’s Draft Interim Guidance generated significant opposition from the academic community.52 He left OHRP and returned to Harvard shortly thereafter.53 Nevertheless, increasing public scrutiny led PHS in 2009 to consider a revision to its COI rules. Over a two-year period it held two rounds of public comment on the question of managing institutional conflicts of interest.54 PHS received a “wide range of responses”55 to this inquiry, including calls for further research and greater specification of precisely what rules might be imposed with respect to institutional COI. Given this input, PHS concluded that “requiring Institutions to have a policy on institutional conflicts of interest without providing additional guidance as to the nature and scope of that policy would lead to confusion and inconsistencies across Institutions. We also believe that substantial additional information and deliberations are needed to formulate such guidance.”56 Accordingly, no rules regarding 52 Greenberg, supra note 10, at 162; Greg Koski, Research, Regulations, and Responsibility: Confronting the Compliance Myth—A Reaction to Professor Gatter, 52 eMory l. rev. 403, 413 (2003). 53 Greenberg, supra note 10, at 167. 54 76 Fed. Reg. 53256 at 53257, 53278. 55 Id. at 53278. 56 Id.

Conflicts of interest and academic research 151 institutional COI were adopted, and PHS committed to “continue to consider the issue of institutional conflict of interest with the biomedical research community.” 57 As of this writing, no further guidance has emerged from any significant US funding agency on this topic.58 C.

Governmental Reviewers and Advisory Committee Members

The policies discussed above relate to conflicts of interest by investigators at grantee institutions and the institutions themselves. In addition, the federal government has a number of COI policies that cover government employees. Most importantly, 18 U.S.C. § 208(a) imposes criminal penalties on executive branch employees who participate in any matter in which he or she, or his or her immediate family, has a financial interest.59 This prohibition applies both to regular government employees as well as individuals serving on governmental advisory committees, panels and councils, who are generally regarded as “special” government employees (“SGEs”).60 These positions are often held by prominent academics and community leaders with expertise in the scientific area addressed by a particular governmental agency or institute. SGEs must file periodic certifications of financial interests pursuant to the Ethics in Government Act of 1978.61 Unlike the COI disclosures required of grantee investigators, which only pertain to significant financial interests giving rise to a conflict of interest, SGE disclosures made on form OGE-45062 must include all financial interests (including assets and income) of the SGE and his/her immediate family in excess of $200.63 If an SGE discloses a conflict of interest pertinent to his or her role on a governmental body, then he or she must either recuse himself or herself from the matters as to which the COI exists, or a responsible official must make a determination that the need for the individual’s service outweighs the potential conflict.64 In addition to SGEs, many academics serve on peer review panels and study sections that review and evaluate grant applications for federal funding. These individuals are not subject to the same conflict of interest disclosure requirements as SGEs, but must recuse themselves from consideration of any grant application in which they have a “real conflict of interest.”65 Conflicts by advisory board members have recently been in the spotlight following the results of a 2018 investigation by the journal Science, which found that of 107 physicians who advised the FDA on drug approvals, 40 received more than $10,000 in undisclosed compensation from the manufacturers of the drugs that they voted to approve (or their competitors), 57

Id. Discussion, however, continues. See Sandro Galea & Richard Saitz, Funding, Institutional Conflicts of Interest, and Schools of Public Health, 317 J. aM. Med. ass’n 1735 (2017) (“[S]chools of public health should not accept money if doing so pushes them to be something that is not consistent with their mission to promote the health of the public.”). 59 18 U.S.C. § 208(a). 60 See 18 U.S.C. § 202(a) (defining “special Government employee”). 61 Ethics in Government Act of 1978, codified at 5 U.S.C. § 101 et seq. 62 Off. Gov’t Ethics, Form 450: Confidential Financial Disclosure Report. 63 5 U.S.C. § 102(a)(1)(A). 64 18 U.S.C. § 208(b)(3). 65 45 CFR 52h.5. A “real conflict of interest means a reviewer or a close relative or professional associate of the reviewer has a financial or other interest in an application or proposal that is known to the reviewer and is likely to bias the reviewer’s evaluation of that application or proposal as determined by the government official managing the review …”. Id. 58

152 Research handbook on intellectual property and technology transfer and seven of these received more than $1 million each.66 Interestingly, most of these payments were made after the vote to approve the drug, sometimes over several years. As one ethicist commented, such payments represented a way of “postponing” the “reward” for approving a company’s drug.67

IV.

CONFLICT OF INTEREST POLICIES, PROCEDURES AND CHALLENGES—AN INSTITUTIONAL PERSPECTIVE

Although, as discussed above, many governmental funding agencies require research institutions to comply with their COI policies, these basic policies only set minimum thresholds and provide few practical guidelines for addressing conflicts of interest related to research.68 Institutions are responsible for interpreting these policies, devising strategies to identify and mitigate conflicts of interest, monitoring compliance and enforcing remedial or punitive measures in case of policy violations. Additionally, an institution may adopt its own COI policy for research not expressly covered by funding agency policies (e.g., internally funded research, research funded by private industry or philanthropic organizations, and research funded by state or federal agencies without COI policies). Some philanthropic organizations such as the Pew Charitable Trusts have issued “best practice” recommendations for COI policies.69 And other institutional charters, codes of conduct and public commitments may also impact COI policies.70 Accordingly, COI policies and processes vary widely from institution to institution.71 Not all financial conflicts of interest are the same in kind or degree, and while the effects of conflicts of interest on collective researcher behavior suggest industry relationships (sponsorship, direct remuneration, equity, etc.) tend to result in favorable research outcomes for indus-

66

Charles Piller, Hidden Conflicts?, 361 sCIenCe 17 (2018). Id. at 18. It is notable that payments made to these physicians prior to a drug’s approval were largely disclosed in journal publications, but not to the FDA. Id. at 17. 68 PHS policy provides definitions (45 CFR Part 50, Subpart F, § 50.604) and guidance on managing conflicts (45 CFR Part 50, Subpart F, § 50.605); however, institutions are ultimately responsible for determining directness, relatedness, and significance of financial interests compared to the research when evaluating real or perceived financial conflicts of interest. In other words, determining the size and scale of the COI is the responsibility of the institution, as are best methods to mitigate the COI when possible. Additionally, institutions may decide identification of potential COIs is better accomplished by a COI Committee versus dedicated COI staff. 69 Pew Charitable Trusts, Conflict-of-Interest Policies for Academic Medical Centers– Recommendations for Best Practices (Dec. 2013), available at https://www.pewtrusts.org/-/media/ legacy/uploadedfiles/phg/content_level_pages/reports/coibestpracticesreportpdf.pdf (last visited Oct. 17, 2019). 70 For example, the 2007 document In the Public Interest: Nine Points to Consider in Licensing University Technology, which has been adopted by more than 100 universities and research institutions around the world, expressly commits its signatories to disclose conflicts of interest in the context of university technology transfer transactions. AUTM, In the Public Interest: Nine Points to Consider in Licensing University Technology, Mar. 6, 2007, available at http://www.autm.net/AUTMMain/media/ Advocacy/Documents/Points_to_Consider.pdf (last visited Oct. 17, 2019) (Point 4: “Universities should anticipate and help to manage technology transfer related conflicts of interest”). 71 See Washburn, supra note 4, at 100. 67

Conflicts of interest and academic research 153 try, the reasons why are not as well characterized.72 This complexity and uncertainty presents to institutions the daunting challenge of crafting robust policies and procedures to address a problem of indeterminate size and scale. In this section, we identify some challenges that institutions face when defining conflicts of interest, how best to identify and manage real or perceived conflicts of interest, and practical implications for conflicted researchers (and their research). This Chapter uses the University of Utah’s COI policy and its related procedures as an exemplar of the types of policies that large research universities may implement, but the authors recognize that these may not be the most effective or comprehensive methods for addressing conflict of interest issues related to research. Accordingly, they highlight potential benefits of other strategies. A.

Institutional Conflict of Interest Oversight and Guidance

Prior to 2002, at the University of Utah, each department handled its own conflicts of interest. The departments relied upon faculty to individually self-disclose any conflicts of interest connected to their research. In 2002, the University established a centralized COI office (“COI Office”) with an associated committee (“COI Committee”) to review and manage COIs from all academic units, but the COI Office still relied heavily upon investigator self-reporting of financial interests on a study-by-study-basis. In 2012, in response to the 2011 amendments to the PHS policy described above, the University amended its own COI policy to the present version, which outlines the duties and responsibilities of both the COI Office and the COI Committee to better identify and manage conflicts. The COI Office reports to the Associate Vice President for Research Integrity, who reports to the Vice President for Research, who, in turn, reports to the President of the University. The COI Office currently employs a Director/Officer, a manager, an analyst and a part-time administrative assistant. The main responsibilities of the COI Office include identification of potential individual and institutional conflicts of interest related to research, identification of potential conflicts of interest related to business transactions and oversight of employee compliance with the University’s COI policy and management plans. The analyst reviews individual research projects and the investigators listed on them, identifies potential conflicts of interest, conducts annual compliance reviews of individuals with managed conflicts of interest and reports on the same. The manager assists with research analysis for institutional COIs, gathers information concerning financial interests and research protocols which may be relevant, assists with

72 Justin E. Bekelman, Yan Li, & Cary Gross, Scope and Impact of Financial Conflicts of Interest in Biomedical Research, 289 J. aM. Med. ass’n 4 (2003). Bekelman, et al. conducted a meta-analysis showing industry relationships in general yielded more pro-industry research outcomes. Considering human subjects protections (HIPPA), potential experiments assessing decision-making behavior of conflicted individual researchers and their research outcomes are unlikely to ever get IRB or FDA approval. However, circumstantial evidence from a study evaluating prescribing patterns in physicians suggests that the amount of money does have a measurable effect on behavior (the more money a physician received from industry, the more likely the physician prescribed the company’s drug). See Charles Ornstein, Mike Tigas, & Ryann Grochowski Jones, Now There’s Proof: Docs Who Get Company Cash Tend to Prescribe More Brand-Name Meds., propublICa, Mar. 17, 2016, available at https://www .propublica.org/article/doctors-who-take-company-cash-tend-to-prescribe-more-brand-name-drugs (last visited Oct. 17, 2019).

154 Research handbook on intellectual property and technology transfer non-compliance investigations, reviews business transaction conflicts of interest, aids with information technology infrastructure to improve efficiency and system design, provides customer service to constituents, and engages in education outreach to the University. The Officer engages in all of the tasks of the manager and the analyst. In addition, the Officer addresses necessary policy changes within the COI Office and across other offices that require financial disclosures (for example, School of Medicine regarding its Industry Relations policy, Human Resources and the like). The Officer coordinates with departments, academic units, and other offices to assist in drafting internal COI policies, designs information technology processes to eliminate redundancy in the system and to automate the process for non-traditional areas of management, including investigations, non-compliance and remediation, builds checklists and management plans, administers the committee, and reports to the NIH and other funding agencies when necessary. The Officer also liaisons with the Office of General Counsel on sensitive matters where the COI policy and its operation have legal significance. The COI Committee determines whether conflicts arise from significant financial interests based upon the recommendation of the COI Office. The Committee also designs management plans best suited to eliminate or reduce bias in research created by the conflict. The Committee conducts investigations of non-compliance and recommends corrective action and sanctions when necessary and appropriate. Committee members also represent the COI Office in their respective departments and serve as ambassadors for the policy and its goals. In 2018–2019, the Committee comprised twenty-one members drawn from faculty and staff across academic units, with required representation from critical units, including the School of Medicine and Health Sciences.73 Each member serves a staggered three-year term. A faculty member who has previously served on the Committee is selected as chair and serves a two-year term. The chair leads meetings, weighs in on policy disputes or interpretation issues, and votes with the Committee only when there is a tie. The Committee meets once every three weeks to discuss agenda items set by the COI Office. Such agenda items include reviewing compliance with management plans, reviewing significant financial conflicts of interest, resolving conflict of interest disputes, and conducting institutional COI reviews pertaining to senior administrators. Each conflict of interest review is assigned to a seven to eight-person panel. If the COI Office receives a complaint, which could include third parties raising issues about conflicts, out-of-compliance management plans, and other policy violations, the Officer investigates the basis for the complaint, consults with the committee, and determines whether to bring a full-blown investigation. When an investigation is warranted, the Officer sends the complaint and the COI Office’s preliminary findings to the subject of the complaint, who then has an opportunity to respond to the complaint. The Officer assigns an investigation to two Committee members who conduct the investigation, predominantly using interviews with involved parties. The COI Office does background research by obtaining relevant publications, reviewing relevant consent documents, working with the Institutional Review Board (“IRB”) when necessary to collect more information, and interviewing personnel who may have further information. The Committee members investigating a complaint have 90 days to present their findings to the Committee, which votes on whether or not a policy violation has occurred and whether any recommended corrective actions are necessary. A letter report 73 For the complete details of committee membership, see Regulations, u. utah, available at https:// regulations.utah.edu/general/rules/R1-006A.php (last visited Sept. 7, 2018).

Conflicts of interest and academic research 155 is then sent to the cognizant Senior Vice President and any relevant offices and supervisors connected to the investigation. B.

Significant Financial Interests

As noted above, in 2011 the PHS implemented a conflict of interest policy governing PHS-funded institutions that shifted the responsibility for identifying conflicts of interest from investigators to the institutions that employ them.74 In the PHS policy, now reflected in many institutional policies, including that of the University of Utah, the agency defines a financial conflict of interest as “a significant financial interest that could directly and significantly affect the design, conduct, or reporting of … research.”75 Without further guidance, it falls to the institution to identify when significant financial interests create conflicts of interest by evaluating how a significant financial interest could “directly and significantly” affect an investigator’s research. This may appear straightforward at first glance, but an institution typically makes this assessment utilizing an extensive review process. The University of Utah requires all faculty and research personnel to report outside financial income and relationships that “reasonably appear related to his/her professional responsibilities to the University” through a centralized reporting mechanism.76 The University of Utah’s Office for Sponsored Research (“OSP”) and its IRB also maintain records of research personnel associated with sponsored projects and human subjects research. When a new award is administered through OSP, a milestone report is generated by OSP, or a new application, continuing review, or amendment is submitted to the IRB, the list of investigators on the research project is checked in an automated financial disclosure system for up-to-date financial disclosures that may include a significant financial interest. Information associated with investigators who disclose a financial interest considered significant is forwarded to the university’s COI Office, which conducts a manual review of each investigator’s financial interest and the potential for the financial interest to impact the investigator’s role in the design, conduct, or reporting of the research project. To set standards and thresholds in this evaluation process, an institution may make assumptions and compromises to further the intent of its own COI policy. In some cases, the mission of the institution may be at odds with this intent. For example, a retention, promotion and tenure policy may require faculty to secure research funding to obtain tenure or promotion, or an invention policy might incentivize faculty to commercialize discoveries resulting from their research. In both cases, the incentives for faculty to engage in seeking external funding and commercialization of inventions could be undermined by a robust conflicts of interest policy tied to significant financial interests. The University of Utah policy recognizes three categories of significant financial interests: (1) direct remuneration received from a business entity in the form of consulting, speaking, honoraria, employment and the like and/or an equity interest in a publicly traded company (where equity includes stock, stock options, or an ownership interest tied to fair market value) in an aggregate of $5,000 or more in the previous 12 months; (2) any remuneration received by

74

See 76 Fed. Reg. 53256, 53288 (Aug. 25, 2011); 45 C.F.R. Part 90 (2017). 45 C.F.R. Part 93 (2017). 76 Individual Financial Conflict of Interest Policy, u. utah polICy 1-006, available at https:// regulations.utah.edu/general/1-006.php (last visited June 2, 2018). 75

156 Research handbook on intellectual property and technology transfer an individual from a non-publicly traded entity in the previous 12 months that exceeds $5,000 or when the individual holds any equity interest in a non-publicly traded entity; and (3) a filed patent, an asserted copyright, or income related to such patent or copyright rights and interests, including royalty income from the institution or others (“IP interest”). This policy explicitly excludes salary, travel reimbursements or other normal remuneration from the university if the investigator is employed by or appointed to the university, income from seminars, lectures, or teaching engagements sponsored by a federal state or local government agency, an institution of higher education, an academic teaching hospital, a medical center, or a research institute that is affiliated with an institution of higher education, income from service on advisory committees or review panels, or income from investment vehicles such as mutual funds or retirement accounts as long as the investigator does not control the funds in these vehicles.77 To apply this policy, the University of Utah’s COI Office relies on investigator self-disclosures of all financial interests reasonably related to her University responsibilities through a centralized business relationship reporting system where the questions are targeted to these specific categories. Moreover, as mentioned above, the system obtains information from OSP and IRB to link individual financial interests with research projects. The COI Office then obtains those financial disclosures from individual employees that fall into one or more of these three significant financial interest categories when the employees are listed as investigators on research project(s) in the OSP and IRB systems. Accordingly, the COI Office obtains information on all disclosed direct remuneration, equity interests, and IP interests when those interests are disclosed by an individual associated with sponsored or human subjects research.78 The following are examples of circumstances that would likely be identified as significant financial interests: (1) (2) (3) (4)

A clinical faculty member is paid more than $5000 by a pharmaceutical company to give a presentation about the benefits of one of its new drugs. A post-doctoral fellow purchases over $5000 of stock in a publicly traded company. A laboratory manager purchases any amount of stock in a non-public company. An investigator works with the institution’s technology transfer office (“TTO”) to file a patent application, where the investigator has assigned his interest in the invention to the institution but is entitled to a royalty in accordance with the institution’s IP policy.

Importantly, the categorization process allows for almost no discretion by the COI Office. Once the COI Office obtains a disclosure flagged with a significant financial interest, the office staff reviews the disclosure for new information related to an existing project already processed for conflicts and for new projects with additional investigators associated with them. The office staff also ensures that all investigators have up-to-date attestations of financial interest in the university system.

77

Id. The ability to review only those files for which OSP or IRB have active research projects is a matter of historical infrastructure and a way to manage the number of disclosures reporting to the COI Office, which employs only two full-time reviewers. 78

Conflicts of interest and academic research 157 C.

Evaluation of Significant Financial Interests for Conflicts

After flagging a significant financial interest, the COI Office makes a preliminary assessment whether a potential financial conflict of interest exists for that individual investigator, and, if so, works with the COI Committee to manage, reduce, or eliminate the conflict using a management plan. Because the University of Utah policy does not explicitly define financial conflict of interest, the COI Office’s working policy is to use the PHS definition: “a significant financial interest that could directly and significantly affect the design, conduct or reporting of research.”79 The conflict of interest review process relies on peer assessment in reviewing these standards by the standing COI Committee. Once the COI Office identifies a potential financial conflict of interest for an investigator, it assigns the review for a discussion at a Committee meeting. If the review involves human subjects research or constitutes a new significant financial interest that has not yet been reviewed, the COI Office assigns the review to a Committee member for further review. If the Committee reviews a new project with a conflict that has already been reviewed for related previous non-human subjects projects, it applies the same management plan as before and sends it to the meeting without assignment to an individual Committee member, but the Committee will modify the existing plan if necessary based on differences in study design and personnel. After completing his own review of assigned reviews, each Committee member presents to the full Committee the case for or against a financial conflict of interest. The Committee then deliberates and determines direct and significant relatedness between the investigator’s significant financial interest and the research in question. The reviewer submits a recommended management plan and the Committee votes to accept the recommendation as is or to amend the plan. Management strategies are discussed more fully below. D.

Management of Financial Conflicts of Interest

Many financial conflicts of interest can be managed, reduced, or eliminated entirely. The Committee uses multiple management strategies that focus on transparency, data integrity, protections for students and subordinates, IP reporting and recusal from commercialization activities, and enhanced human subjects research scrutiny. If, upon preliminary review, the COI Office identifies a potential conflict, its staff reaches out to the investigator and begins gathering information about the business relationship or other significant financial interest, IP and other information that could be relevant to the Committee’s review, including whether students or subordinates are participating in the research or whether the research may generate new IP. The investigator is also asked about data integrity—can the data be blinded, can a non-conflicted investigator collect the data and

79

42 C.F.R. Part 50.603. The 2011 amendments to the PHS COI policy (see supra section II.A) resulted in a paradigm shift in identifying conflicts of interest in research. Prior to the 2011 amendments, the University of Utah required researchers to self-identify and report conflicts of interest related to specific studies. The change to PHS policy required researchers to disclose to the institution all of their financial interests “reasonably related” to their institutional responsibilities. Identification of conflicts of interest became the responsibility of the institution through study-by-study review of investigators’ financial interests relative to the aims of the research.

158 Research handbook on intellectual property and technology transfer the like. The collection of this information allows for a customized management plan to be developed for every case that best reduces, eliminates, or manages the conflict of interest. The most common strategy for addressing conflicts of interest is transparency. Conflicts should be disclosed in publications and presentations given by the investigator. This is consistent with the requirements of many academic journals that also require authors to disclose conflicts of interest. The rationale for disclosing in this manner is to make a consumer of the research aware that the results could be susceptible to potential bias and better equipped to interpret the research outcomes. Other types of transparency include disclosing conflicts to the research team, disclosing to subordinates and students who may be working for the conflicted investigator (and prohibiting evaluation or review of those subordinates and students by the conflicted investigator and prohibiting the conflicted investigator from delaying publication of work by the students and subordinates that is related to the financial interest), and in the case of human subjects research, disclosing the conflict of interest to potential research participants. Here, the rationale is to educate those individuals working with the conflicted investigator to ensure that potential bias from the financial interest is recognized. Disclosure in consent documents allows human subjects to assess the costs and benefits of being involved in a research study with a conflicted investigator. Another form of disclosure required by PHS is to publicly disclose or have a means to distribute information about PHS-funded studies in which a conflict of interest has been identified. The University of Utah publishes such studies on a publicly available website.80 It has also independently decided to disclose any financial conflicts of interest involving human subjects research on that website. A second area of concern suggesting management is data integrity. The COI Committee will use one or more of several strategies including blinding the data, prohibiting conflicted individuals from data collection and analysis, requiring a non-conflicted peer to review the study and file reports with the COI Committee, having a non-conflicted individual review data and results prior to submission for publication or, in more extreme cases, having a non-conflicted peer monitor the research and submit annual reports to the COI Committee. The Committee may also notify a data and safety monitoring board, if one exists. When the Committee identifies a financial conflict of interest in a human subjects research study, the investigator receives greater scrutiny. The University of Utah COI policy includes a rebuttable presumption that the conflicted investigator cannot participate in the research at all. The Committee reviewer contacts the conflicted investigator making him aware of the rebuttable presumption, and invites the investigator to present evidence of credible, compelling circumstances to rebut the presumption and allow the investigator to continue to work on the study.81 In addition to disclosing conflicts to the potential participants, the Committee might prohibit the conflicted investigator from obtaining patient consent to participate in the study and invite a non-conflicted member of the research team to obtain research subject consent, thereby removing any type of coercion by the conflicted investigator in convincing the subject to participate in the research. Relatedly, if a PHS Operating Division funds a study

80 Conflict of Interest, u. utah, available at https://coi.utah.edu/_documents/coi_list_0415.pdf (last visited Apr. 3, 2019). 81 See Advisory Committee on Financial Conflicts of Interest in Human Subjects Research, Protecting Patients, Preserving Integrity, Advancing Health: Accelerating the Implementation of COI Policies in Human Subjects Research, AAMC-AAU, 2008, vi–vii, xii, 6–7, 15, available at https://www .aamc.org/system/files/c/2/482216-protectingpatients.pdf (last visited Oct. 17, 2019).

Conflicts of interest and academic research 159 in which an investigator receives a management plan of any variety, the COI Office must report the conflict of interest and the plan to the agency. Every conflict that the Committee manages is given, at a minimum, a mitigation plan with four basic disclosure-related components. First, the Committee will conduct an annual compliance review for each conflicted investigator to ensure that person complies with the management plan assigned by the Committee. Second, the Committee requires that the investigator disclose the financial conflict of interest in any and all publications related to the project. Third, the investigator must disclose the financial conflict of interest in all professional presentations related to the project. Fourth, the investigator must disclose the conflict of interest to the entire research team. Additionally, the Committee asks whether the conflicted investigator is working with students or subordinates on the project, and if so, the investigator must disclose his conflict to them as well. If the research project has human subject participants, the investigator must disclose his conflict of interests in the consent documents. When an inventor is investigating his own IP in a non-human subjects sponsored research project, the investigator receives a standard management plan with additional disclosures to the TTO if the project will generate new IP owned by the institution. The investigator will be prohibited from negotiations regarding the IP. If the data cannot be blinded, a non-conflicted investigator must collect and analyze the data. If this is not possible, the committee may assign a peer monitor to oversee the entire research project and to report annually to the Committee. When the conflicted investigator is not a principle investigator (“PI”), the PI can oversee the project with respect to conflicts. If the PI is conflicted, the Committee might choose a monitor not on the study or someone who is not a subordinate of the conflicted investigator. E.

Non-Research Conflicts of Interest

In addition to the measures described above to eliminate the effect of conflicts of interest on research, state law and university policy prohibit an individual from participating in any decision-making with respect to a transaction when he or she has a direct financial interest in the transaction.82 The Committee provides an additional check on these non-research conflicts of interest that may arise through a variety of business transactions by institutional employees. This includes subawards and subcontracts that derive from grants and contracts, license agreements related to IP, and general institutional procurement (for example, using research funding to buy lab supplies or other inventory or services from a company owned by the researcher). When an institutional employee submits a financial disclosure to the business relationship reporting system that includes a significant financial interest in a business entity, the system requires the employee to accept a standardized management plan. As part of the management plan the employee must refuse himself from participating on behalf of the university in the decision-making process of any business transaction between the university and the business entity in which the employee has a significant financial interest.83 Because the university

82

Utah Code §§ 67-16-1 through 67-16-10 (2019). The University of Utah uses the following standard language of management regarding business transactions: “I have disclosed a significant financial interest in the entity(ies) named above, which currently transact, or may in the future transact, business with the University of Utah. I will advise my immediate supervisor of the existence of this financial interest. In the event that the University transacts business with this entity, I will not participate in any aspect of that process (e.g., negotiating license 83

160 Research handbook on intellectual property and technology transfer offices involved in procurement, sponsored projects, and technology transfer engage in business transactions that may involve employees with conflicts, those offices have been instructed to communicate with the COI Office the names of those employees who have received the notice. When one of the employees identified has a financial interest relevant to the transaction, the COI Office confirms to the inquiring office that the individual has accepted the terms of the mandatory notice. If the conflicted employee must provide input to the transaction—for example, if the conflicted employee would provide important information regarding the negotiation of a contract—then the employee must recuse himself from the decision-making process, provide written notice to all involved individuals that a conflict exists, and, in most cases, have a non-conflicted supervisor ensure that negotiations are at arm’s length. F.

Monitoring and Non-Compliance

After a conflicted investigator receives a management plan from the Committee, the Committee continues to oversee the investigator’s compliance with the plan. The Committee also oversees compliance with the general employee disclosure policy that enables identification of significant financial interests and the institution’s polices with regard to non-research related business transactions. Questions of compliance may be brought to the COI Office or the Committee by anyone concerned, anonymously or not. The COI Office may learn about disclosure failures and underreporting of financial interests from relevant institutional actors, including internal audit offices and the like. The COI Office also may discover non-compliance on its own, including by use of the Centers for Medicare and Medicaid Services (“CMS”) Open Payments database, which tracks payments from pharmaceutical and medical device companies to physicians and teaching hospitals.84 If the COI Office suspects non-compliance by an employee who is subject to reporting requirements, by an investigator subject to a management plan, or by an employee engaging in prohibited business transactions, the COI Office opens an investigation into the alleged non-compliance, and assigns the investigation to two Committee members. The Committee members, in consultation with the COI Office, conduct the investigation and report their findings to the full Committee. The Committee then votes on recommended actions within its purview (such as recusal from the research study) and sanctions, if appropriate. Sanctions will

agreements, research proposals, subcontracts, and procurement contracts, including the preparation of specifications, evaluations or bids, committee deliberations/voting) on behalf of the University. To the extent that my University responsibilities would normally result in my participation in such transactions, I will assure that any minutes associated with that transactional process reflect my recusal from the decision-making process. If my University responsibilities require me to provide input into the selection of products or services from or to this business entity, at the time of providing such input, I will clearly disclose in writing to the University decision-makers the existence of my financial interest in the business entity. I am aware that the Utah Public Officers and Employees Ethics Act, Utah Code Ann. §§ 67-16-1 to -15 (the Ethics Act) creates additional legal obligations for me, including obligations related to my participation in a private business that enters business transactions with my public employer, and I acknowledge that I remain personally responsible for complying with the Ethics Act.” 84 CMS Open Payments, available at https://openpaymentsdata.cms.gov/ (last visisted Apr. 3, 2019), was created in accordance with section 6002 (also known as the Physician Payments Sunshine Act) of the Affordable Care Act of 2010, which required manufacturers of drugs, medical devices, and biologics to report payments to recipients. See 42 U.S.C. § 1320a-7h et seq.

Conflicts of interest and academic research 161 be recommended to the cognizant university administrator for enforcement and follow up. If the study has been funded by the PHS, the COI Office must report findings of non-compliance to the agency.85 G.

Institutional Conflicts of Interest

As discussed in section III.B above, certain interests held by an institution and its senior leadership could compromise or appear to comprise the integrity of research conducted at the institution. Therefore, the University of Utah has developed an institutional COI (“ICOI”) policy that is designed to manage, reduce, or eliminate such institutional financial conflicts of interest related to human subjects research in the same manner as the individual COI policy discussed above. As a result, the ICOI policy is very similar in form and function to the individual COI policy. For senior personnel (who may not be directly involved with research but do act in a supervisory role to the investigators conducting the research), the definition of “significant financial interest” is the same as in the individual policy. In the ICOI policy, the university financial conflicts of interest are defined as: (1) University ownership of equity in a company (stock, stock options, of at least 5% of the company), (2) University-owned technology that is currently licensed or has previously been licensed, including royalties received for such licensing agreements, and (3) significant gifts from a company to the University. Significant gifts means gifts to the university of securities, other property, or cash exceeding $100,000 over the previous four academic years. The COI Office cross-references financial disclosures by senior personnel and human subjects research projects to identify institutional conflicts. When a senior personnel-related institutional financial interest has been identified in a human subjects research study, the study goes through Committee review in the same manner as an individual financial conflict of interest. The Committee makes recommendations to the ICOI Officer, who renders the final decision on management of the conflict. To identify university financial interests related to human subjects research, the COI Office cross-references the university’s development office database and the TTO’s database of university owned IP, university company ownership and university licensing agreements with research study information from the IRB to identify potential financial interests. If one has been identified, the COI Office evaluates the financial interest to determine whether a conflict exists, using the same standard for evaluation that the Committee uses for individuals and senior personnel. If an institutional conflict of interest can be managed, the ICOI Officer unilaterally enters a management plan. In these cases, the management strategies are generally more comprehensive because data integrity and patient safety may require independent, non-university oversight. If the ICOI Officer determines that an institutional conflict cannot be managed, the research is not allowed to be conducted at the university.

85 42 C.F.R. §§ 50.604, 50.605 (2018) (describing the institution’s reporting obligations with respect to financial conflicts of interest).

162 Research handbook on intellectual property and technology transfer

V.

ACADEMIC JOURNAL POLICIES

As described above, many institutional COI policies require that investigators disclose known conflicts of interest relating to a research study in any published article reporting the results of that study. As explained by Gottlieb and Bressler, “Disclosure of a COI in a publication invites readers to consider whether the study design was objective, data were appropriately analyzed, and interpretations and conclusions accurately reflect the data.”86 There are several standardized processes for disclosing such conflicts today, the best known of which is the COI disclosure template made available by the International Committee of Medical Journal Editors (“ICMJE”).87 The ICMJE template requests that authors of works for publication disclose far more than most federal or institutional policies. Disclosed financial interests, for example, include all “financial relationships with entities in the bio-medical arena that could be perceived to influence, or that give the appearance of potentially influencing, what you wrote in the submitted work.” 88 It goes on to clarify that “if your article is about testing an epidermal growth factor receptor (‘EGFR’) antagonist in lung cancer, you should report all associations with entities pursuing diagnostic or therapeutic strategies in cancer in general, not just in the area of EGFR or lung cancer.”89 Moreover, the ICMJE template seeks disclosure of all sources of revenue paid to the author or his/her institution on his/her behalf during the prior 36 months, not only the sponsor of the particular study that is the subject of the article.90 While the transparency afforded by such disclosures is often sufficient to give reviewers and readers the information they need to assess potential bias of an author’s work, some journals impose requirements beyond simple disclosure to certain types of work. For example, some journals prefer not to publish opinion pieces (editorials, commentaries and review articles) in which an author has a conflict of interest, as a truly objective opinion may be desired in these pieces.91 In addition to authors, the editors, reviewers and publishers of scientific journals may also have conflicts of interest with respect to the subject matter of the articles appearing in them. The editors of most scientific journals are non-staff volunteers who are recognized authorities in a field. As such, these individuals may have conflicts of interest that affect their impartiality when it comes to selecting and editing articles. The same applies to peer reviewers who are invited to review and make recommendations regarding articles submitted to a particular journal. Conflicts in these cases include not only financial ties, but any significant personal, employment or co-authorship relationship or any significant scientific or ideological differ-

86 Julie D. Gottlieb & Neil M. Bressler, How Should Journals Handle the Conflict of Interest of Their Editors?, 317 J. aM. Med. ass’n 1757 (2017). 87 Int’l Comm. Med. J. Eds, Conflict of Interest Disclosure Forms, available at http://www.icmje .org/about-icmje/faqs/conflict-of-interest-disclosure-forms/ (last visited Apr. 3, 2019); see Joseph T. Thornton, Conflict of Interest and Legal Issues for Investigators and Authors, 317 J. aM. Med. ass’n 1761 (2017) (“The ICMJE form is the industry standard that many journals require”). 88 Int’l Comm. Med. J. Eds, Form for Disclosure of Potential Conflicts of Interest. 89 Id. 90 Id. 91 See Am. Med. Ass’n, AMA Manual of Style: A Guide for Authors and Editors, Ch. 5.5.1 (10th ed. 2009), Phil Fontanarosa & Howard Bauchner, Conflict of Interest and Medical Journals, 317 J. aM. Med. ass’n 1768, 1770 (2017).

Conflicts of interest and academic research 163 ence of opinion.92 In the case of editors and peer reviewers, a conflicted individual is generally required to recuse himself or herself from reviewing the article with which a conflict exists.93 And most importantly, editors and peer reviewers are expressly forbidden from using knowledge of the work they are reviewing for their own personal gain (financial or academic) before it is published.94 Journal publishers, too, may encounter conflicts of interest, particularly if they enter into financial arrangements with private firms for special issues, advertising spreads and other features.95 Ensuring that such arrangements do not compromise the editorial selection and quality of a journal’s content is critical, and few journals have robust procedures in place to do so.96 Despite the disclosure mechanisms in place at scientific journals, these processes are not perfect. In some cases, authors or others may fail, inadvertently or intentionally, to disclose a relevant COI. One recent study reported that 100 researchers who received median compensation of approximately $100,000 from surgical and medical device manufacturers reported a relevant COI in only 37% of publications.97 Another recent study found that one third of oncologists publishing the results of clinical trials failed to make full disclosure of the payments that they received from trial sponsors.98 Such failures to disclose COIs are frequently detected by journal readers. Often a reader will notice that a particular author who previously disclosed a COI in a different article or journal has failed to disclose the same COI in a more recent publication. In addition, as Thornton observes, “bloggers and other ‘COI watchers’ scour the literature for failures to disclose.”99 Most journals provide a mechanism for readers to report suspected failures to disclose COIs and will investigate questionable cases.100 In the case of Dr. Baselga of Memorial Sloan Kettering, the failure to disclose was uncovered by investigative journalists from the New York Times and ProPublica.101 If a failure to disclose a COI is discovered, a journal may require the author to issue an acknowledgement, letter of explanation, apology and/or correction.102 Journal editors caution, however, that allegations regarding a failure to disclose “need not be judged through the prism of ‘the unreasonable, uninformed, or overly zealous.’”103 These words of caution may belie the

92

See Sara Rockwell, Ethics of Peer Review: A Guide for Manuscript Reviewers 5–9, available at https://ori.hhs.gov/sites/default/files/prethics.pdf (last visited Oct. 17, 2019); Fontanarosa & Bauchner, supra note 91, at 1771. 93 See Int’l Comm. Med. J. Eds, Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals 4, Dec. 2017, available at http://www.icmje.org/ about-icmje/faqs/icmje-recommendations/ (last visited Oct. 17, 2019). 94 Id. 95 See Thomas J. Easley, Medical Journals, Publishers, and Conflict of Interest, 317 J. aM. Med. ass’n 1759, 1760 (2017). 96 Id. 97 Kasra Ziai, et al., Association of Compensation from the Surgical and Medical Device Industry to Physicians and Self-declared Conflict of Interest, JAMA surgery, Aug. 15, 2018. 98 Cole Wayant, et al., Financial Conflicts of Interest Among Oncologist Authors of Reports of Clinical Drug Trials, JAMA onCology, Aug. 30, 2018. 99 Thornton, supra note 87, at 1762. 100 See id. at 1762; Fontanarosa & Bauchner, supra note 91, at 1770. 101 Ornstein & Thomas, supra note 1. 102 See Thornton, supra note 87, at 1762; Fontanarosa & Bauchner, supra note 91, at 1770. 103 Thornton, supra note 87, at 1762.

164 Research handbook on intellectual property and technology transfer fact that, despite formal policies, few journals take punitive measures against authors who fail to make required disclosures.104

VI.

CONCERNS, CRITIQUES AND FURTHER RESEARCH

In response to the increasing support of academic research by the private sector and a few highly-publicized instances of academic misfeasance, federal funding agencies, research institutions and scientific journals have developed a complex system for identifying, assessing, reporting and mitigating conflicts of interest. Most agree that these procedures are necessary to ensure the integrity of the research enterprise and to preserve public trust. This being said, some consider the current system for reporting and managing conflicts of interest to be unduly burdensome: an over-reaction to a few egregious cases of research misconduct.105 Research institutions estimate that they spend large sums complying with federal and institutional COI rules. One survey by the Association of American Medical Colleges estimated that 71 institutions spent $23 million to comply with the revised PHS rules in the year after they were implemented.106 Many have criticized the lack of standardization among COI reporting metrics and requirements, creating unnecessary work and administrative overhead for researchers and providing inconsistent and difficult-to-understand information to the public.107 One 2015 study published in Nature observed that “institutions have vastly different standards for what constitutes a conflict—and […] they classify relatively few relationships between researchers and industry as troublesome.”108 Others have argued that COI disclosures create a barrier to effective industry-academia partnerships and collaborations, both increasing the time required to form collaborations and reducing the ability and desire of both academics and industry partners to enter into collaborations.109 Critics also argue that COI reporting requirements result in decreased educational opportunities for researchers, given reductions in corporate funding, events and paid speeches by industry experts.110 Concerns over failure to comply with COI reporting requirements could lead to over-disclosure by researchers who zealously report every possible financial and other interest in an abundance of caution. Such over-reporting could lead to an exaggerated view of the prevalence of conflicts, as well as obscuring important conflicts among the mass of trivial ones.111

104

See Ornstein & Thomas, supra note 1 (reporting that the American Assocation for Cancer Research stated that it “had never barred an author from publishing” despite failures to comply with disclosure rules). 105 IOM 2014, supra note 16, at 11. 106 Reardon, supra note 36, at 300–1. 107 IOM 2009, supra note 15, at 8–10. 108 Reardon, supra note 36, at 300. 109 IOM 2014, supra note 16, at 9. 110 Id. 111 See Allen S. Lichter, Conflict of Interest and the Integrity of the Medical Profession, 317 J. aM. Med. ass’n 1725 (2017).

Conflicts of interest and academic research 165 Yet others believe that under-reporting of conflicts is still a problem, particularly as institutions make the initial determination whether a particular COI is likely to lead to bias.112 Likewise, the transparency-based rationale that underlies most federal and institutional COI policies has been criticized as too weak, particularly in the case of human subjects research. Several groups, including the Association of American Medical Colleges, the Association of American Universities and the Institute of Medicine, have recommended that researchers be excluded from human subjects research if they have a financial interest in the outcome of the research.113 Jeffrey Botkin, former Associate Vice President for Research at the University of Utah, has recently called for more stringent penalties for researchers who violate journal or institutional COI rules.114 Responding primarily to the Baselga incident at Memorial Sloane Kettering, Dr. Botkin recommends that failures to disclose COI be considered forms of research misconduct—conduct that can lead to serious penalties that “strike[] at the basic integrity of anyone who pursues a scientific career.”115 These concerns and critiques point to the need for more research in the area of COI. Among the issues that could benefit from further research are the causal relationship between financial COI and research bias, the cost of implementing and maintaining COI reporting and management systems, the impact of COI disclosures on research subjects, and the best means for integrating disclosure mechanisms across the research enterprise.116 It is hoped that a greater, evidence-based understanding of COI and their effects on research may pave the way for continuing improvements in the COI system.

112

Reardon, supra note 36, at 300 (according to one researcher, “we’re just seeing the tip of the iceberg”). 113 IOM 2009, supra note 15, at 9 and Recommendation 4.1. 114 Jeffrey R. Botkin, Viewpoint: Should Failure to Disclose Significant Financial Conflicts of Interest be Considered Research Misconduct?, 320 J. aM. Med. ass’n 2307 (2018). Dr. Botkin does not speak on behalf of the University of Utah in this article. 115 Id. at 2308. 116 See, e.g., William W. Stead, The Complex and Multifaceted Aspects of Conflicts of Interest, 317 J. aM. Med. ass’n 1765, 1767 (2017).

8.

Modern intellectual property valuation in the academic technology transfer setting Bryce Pilz

I.

INTRODUCTION

This Chapter addresses approaches to valuing intellectual property (“IP”) in the modern university technology transfer office (“TTO”). The valuation of an IP asset that a TTO is licensing to an industry partner has long played an important role for TTOs. In the nearly forty years since passage of the Bayh-Dole Act, however, the roles and responsibilities of TTOs have drastically evolved and expanded. Most TTOs originally focused on procuring patent protection and licensing the resulting patent rights to maximize monetary returns. Today, TTOs now devote significant resources to startup support, engage in economic development, facilitate startup ecosystem building, build industry partnerships, handle a diverse set of deals beyond patent licenses, assist with entrepreneurial training for faculty and students, provide gap and translational funding for internal projects, manage seed funding for newly formed startups, and manage the equity held in their startup licensee portfolio. Not surprisingly, with the expanded mission of most TTOs, the role of IP valuation is also evolving to align with this new mission. This Chapter seeks to explain the role of various IP valuation methodologies in this modern setting. Section II of this Chapter summarizes the modern TTO office, with its expanding responsibilities and evolving mission. This section also explains how IP valuation matters in the context of the evolving TTO. Section III addresses different technology transfer deals and deal structures and how valuation relates to the terms of those deals. Section IV overviews historical valuation methodologies and how they relate to modern IP valuation by TTOs. Section V addresses IP valuation in the two specific contexts of royalty buy-outs and licenses granting the university equity in the licensee. Section VI discusses the adoption of portfolio-based valuation thinking occurring at TTOs today. Lastly, section VII offers concluding thoughts on how TTOs will continue to refine their approaches to valuation to align with their expanded missions.

II.

THE MODERN TTO AND THE ROLE OF IP VALUATION

While not all TTOs engage in the same scope of activities, generally speaking, the roles and responsibilities of TTOs have evolved significantly in the four decades since the passage of the Bayh-Dole Act. A broad discussion of the evolution of university technology transfer is beyond the scope of this Chapter. Nonetheless, a brief summary of the expanded and evolving nature of TTOs is important background for this Chapter’s discussion of IP valuation in TTOs.

166

Modern intellectual property valuation in the academic technology transfer setting 167 A.

The Expanded and Evolving Roles of the Modern TTO

TTOs today focus on much more than just pursuing patent protection and licensing the resulting patent rights. The Association of Public & Land-Grant Universities recently commissioned a “Technology Transfer Evolution Working Group” to generate a report on the evolution of technology transfer. This report explained: University leaders are increasingly responding to the needs of the innovation economy—and in particular their local economies—by including innovation, entrepreneurship, and “economic engagement” programming in their strategic planning processes. As part of this response, university technology transfer offices are evolving, and must continue to evolve, toward participation in a broader scope of efforts—with patents and licensing as one emphasis, and also connecting with and engaging in other efforts that support the learning and discovery missions of the university. In evolving toward broader participation in university economic engagement, technology transfer offices will develop deeper relationships with industry and other community partners; broaden their reach to areas such as education, technology development, and entrepreneurship; and integrate more closely with other supportive administrative functions such as industry contracting. While budget and resource threats to the university research enterprise are creating increased pressure to generate revenue from licensing and innovation activities, university leaders must recognize that successful economic engagement will not be focused on short-term income, but rather on longer-term work on relationship development and ecosystem building.1

One technology transfer veteran further explained the evolution of academic technology transfer, referring to the early days as “Technology Transfer 1.0.”2 This time period involved TTOs focusing on convincing faculty to participate in technology transfer, soliciting invention disclosures, patenting, and licensing.3 This evolved into “Technology Transfer 2.0,” with an increase of licenses to faculty startups, TTO involvement with industry sponsored research agreements, the emergence of basic gap fund programs, and an increase in material transfer agreements handled by TTOs.4 “Technology Transfer 3.0” saw an increase of funding for TTOs, more sophisticated startup support programs, an increased emphasis on marketing technologies, more education and outreach, and a greater investment in legal support for patenting activities.5 The technology transfer veteran explained the current environment as “Technology Transfer 4.0,” involving a significant focus on translational research programs, a wider variety of innovations handled by TTOs, and technology portals for online licensing.6 As just one example of how this evolution of TTOs has impacted the thinking around IP valuation, the increased focus on startups, economic development, industry partnerships, and other relationship building has placed TTOs in the middle of a complicated, multi-sided marketplace. As explained by the APLU Working Group:

1 Technology Transfer Evolution: Driving Economic Prosperity, assoCIatIon of publIC & land-grant unIversItIes, Nov. 2017, at 3, available at http://www.aplu.org/library/technology-transfer -evolution-driving-economic-prosperity/file (last visited Oct. 17, 2019) [hereinafter Technology Transfer Evolution: Driving Economic Prosperity]. 2 Arundeep S. Pradhan, The Evolution of Technology Transfer, apIoIx.CoM, 2016, available at http://apioix.com/technology-transfer-trends/ (last visited Oct. 17, 2019). 3 Id. 4 Id. 5 Id. 6 Id.

168 Research handbook on intellectual property and technology transfer [TTOs] find themselves in a dynamic environment where they must balance internal and external stakeholder expectations, including those of institutional leadership, industry, and others in the community in which the institution engages.7

This dynamic environment has dramatically impacted the manner in which TTOs address IP valuation. For example, on one side of this marketplace sit the potential and actual licensees. These licensees include companies ranging from newly formed startups (and their numerous stakeholders) to multi-national corporations. These licensees will seek economic terms that limit the total economic payout to the university and/or to structure those payments so as to significantly shift to the university the risk related to the IP being successfully commercialized. On the other side of this marketplace are the university stakeholders, including inventors and university administrators. At almost all universities, these university stakeholders share significantly in the TTO’s licensing revenues. Accordingly, these stakeholders will strive to see the TTO negotiate economic license terms that maximize the economic return to the university and shift to the licensee the risk related to the IP being successfully commercialized. A TTO seeking to build a vibrant startup ecosystem, foster longstanding, synergistic relationships with investors and serial entrepreneurs, and optimize the size and caliber of its startup and overall licensing portfolio must balance the interests of both sides of this marketplace. B.

Why Valuation Matters in the Modern TTO

In the early version of academic technology transfer, TTOs used IP valuation methodologies primarily to justify economic terms in licenses aimed at optimizing the economic return to the university. While the evolution of TTOs has placed a greater emphasis on goals other than revenue generation, IP valuation still plays an important role. Even as TTOs begin to account for a broad range of non-monetary interests in their deal-making approaches, the value of particular IP assets remains relevant for the following reasons. 1. Stakeholders directly benefit from IP revenues First, for almost all TTOs there are a variety of university stakeholders that directly financially benefit from the university licensing revenues. Most universities employ a revenue distribution policy (“RDP”) that governs the sharing of revenues among various stakeholders, such as the inventors, the inventors’ department(s), central administration, and in some schools the TTO itself. Below is the RDP used at the University of Michigan from 2007 through 2018. After recovery of University Expenses, aggregate revenues resulting from royalties and sale of equity interests shall be shared as follows. The division of revenues are subject to change through appropriate University procedures: ● Up to $200,00: ● 50% to the Inventor(s) ● 17% to the Inventor’s department ● 18% to the Inventor’s school or college ● 15% to the central University administration ● Over $200,000 (and up to $2,000,000):

7

See Technology Transfer Evolution: Driving Economic Prosperity, supra note 1, at 10.

Modern intellectual property valuation in the academic technology transfer setting 169 ● 30% to the Inventor(s) ● 20% to the Inventor’s department ● 25% to the Inventor’s school or college ● 25% to the central University administration ● Over $2,000,000: ● 30% to the Inventor(s) ● 35% to the Inventor’s school or college ● 35% to the central University administration8 As shown, inventors, departments, colleges, and central administration all receive a portion of TTO licensing revenues. Accordingly, at least through the lens of the standard TTO RDP, these stakeholders financially benefit from the monetary value received from particular deals. 2. IP revenues support additional research The revenues returned to the TTO’s institution both directly and indirectly support further research. Licensing revenues directly support research because these revenues are typically reinvested in further research. Using the RDP from the above example, the University of Michigan’s Technology Transfer Policy states “Although the University units described above shall have discretion for distributing the revenue they receive, generally it is expected that revenues will be used for research and educational purposes or for investment in further commercialization activities, such as in the laboratories of Inventors.”9 Indeed, the federal law governing the licensing of inventions created with the support of federal funds requires universities to reinvest licensing revenues in future research.10 While Bayh-Dole Act only applies to federally funded inventions, most TTOs apply the rationale of Bayh-Dole Act to all inventions they license, whether or not federally funded. Licensing revenues may also indirectly support future research by incentivizing faculty to participate in commercializing their research. Department chairs and deans frequently mention that a TTO that returns value to faculty inventors and laboratories will assist a university in attracting talented grad students and faculty.11 3. IP revenues often fund TTO’s patent budgets At many universities, licensing revenues directly support the TTO’s patent budget. In other words, TTOs often receive a percentage of licensing revenues to support patent budgets. At the University of Michigan, as licensing revenues are received, a percentage-based allocation is taken before applying the above RDP. This allocation is used to supplement the TTO’s

8

The University of Michigan Technology Transfer Policy, unIv. MICh. off. teCh. transfer, IV.2 (June 1, 2009), available at http://temp.techtransfer.umich.edu/resources/policies.php (last visited Oct. 17, 2019). 9 Id. at Section IV.4. 10 35 U.S.C. § 202(c)(7)(C) (2012) (requiring that “the balance of any royalties or income earned by the contractor with respect to subject inventions, after payment of expenses (including payments to inventors) incidental to the administration of subject inventions, be utilized for the support of scientific research or education”). 11 At the author’s university, several faculty recently cited that ability to set up consistent royalty streams to inventors in the faculty’s laboratory as a primary mechanism for recruiting the best and brightest graduate students to that laboratory.

170 Research handbook on intellectual property and technology transfer patent budget. It is this allocation that funds the filing of future patent applications and the maintenance of existing applications for which licensees are not directly reimbursing the costs. The funding of patent budgets through licensing revenue is increasingly important for TTOs. Historically, a large portion of a TTO’s patent expenditures was directly reimbursed by the licensee. Such direct reimbursement provisions were common in exclusive licenses to existing companies. Many TTOs have reported a decrease in the portion of their patent expenses that are directly reimbursed by licenses.12 Indeed, at the author’s school the percentage of gross patent expenditures directly reimbursed by licensees has noticeably and consistently dropped from 2012 to 2017. This decrease has occurred despite the number of licenses steadily increasing and the amount of gross patent expenditures remaining relatively constant and even declining slightly. Therefore, university sources of revenue to fund patent expenses are more important than ever for TTOs. Accordingly, for universities that fund the TTO’s patent budget entirely or partially from licensing revenues, IP valuation directly relates to the ability to invest in future patent protection. 4.

IP revenues may increase the likelihood of a licensee diligently commercializing research innovation An often-repeated rationale for requiring licensees to pay fair market value for IP generated from academic research is that a licensee who makes a substantive monetary investment in IP is more likely to follow through on that investment and diligently pursue commercializing that IP. The rationale is, therefore, that charging fair market value for university IP is more likely to achieve a university’s technology transfer mission. This rationale was the subject of discussion as recently as the 2018 annual conference for the Association of University Technology Managers. At that conference, a representative of one institution was discussing how that institution charges pre-set “technology fees” for its industry sponsored research. A representative from another institution provided a counterpoint, noting that a licensee with a paid-up license to IP is less likely to diligently pursue commercializing that IP. That representative continued to explain that if universities’ technology transfer mission is truly to get technologies into the marketplace, then the technology fee model of sponsored research leading to fully paid-up licenses was not the best way to fulfill a TTO’s ultimate mission. That representative concluded that a license negotiated after the identification of IP resulting from the sponsored research agreement, and negotiated to provide fair market value for the IP, was most likely to lead to products in the marketplace.

III.

TYPES OF TECHNOLOGY TRANSFER DEALS AND DEAL STRUCTURES

As discussed above, whereas prior versions of TTOs focused mostly on patent licenses, modern TTOs handle a wide variety of agreements. Because certain valuation methods and principles might be more applicable to certain types of deals, this section provides a brief overview of common TTO deals and their structures.

12

See Pradhan, supra note 2, at 4.

Modern intellectual property valuation in the academic technology transfer setting 171 A.

General Deal Structures

TTOs have long recognized the value of having multiple categories of economic consideration in a license.13 These categories include equity in the licensee and/or other change of control fees, upfront payments, milestone payments, royalties on net sales by licensee or sublicensees, a percentage of non-sales based income from sublicensees, and annual minimum payments.14 TTOs seek to have these multiple streams of economic consideration from licensees in order to diversify the mechanisms for sharing in the value the licensee captures from the licensed technology. This desire for diversification reflects at least three different principles of technology transfer licensing. First, TTOs may seek to balance the near-term revenue generation of up-front payments with the long-term potential of royalty streams or equity payouts. On one hand, near-term payments provide the TTO and its stakeholders with certain revenues and may reflect fair consideration for the value the licensee receives from accessing the university IP asset and having the opportunity to commercialize it. On the other hand, payments, such as royalties, provide the TTO the highest revenue potential. Indeed, historically the research institutions that generate the most licensing revenue do so through royalty payments.15 Second, TTOs desire to diversify their potential revenue streams from the licensee due to the unpredictability of how the licensee may generate value from the licensed IP. One extreme example is Stanford’s license to Google for Stanford’s patent on the page-rank algorithm. Despite the enormous success of Google, Google never charged its individual end users for its page-rank algorithm. Had this license merely required payment of a sales-based royalty, the university’s returns would have been limited.16 On the other hand, Stanford’s equity in Google provided for significant returns as Google’s company valuation increased significantly through multiple rounds of venture capital and then a successful initial public offering.17 Third, sharing value with the licensee in a variety of ways and at different times reflects the principle that anytime the licensee receives value from using the licensed IP, the university licensor should share in that value.18

13 Michael A. Reslinski & Bernhard S. Wu, The Value of Royalty, bIoentrepreneur, June 27, 2016, available at https://www.nature.com/bioent/2016/160601/full/bioe.2016.6.html (last visited Oct. 17, 2019). 14 For an excellent overview of the various financial terms in a TTO license agreement see: Kirsten Leute, Anatomy of a License Agreement, AUTM, 2010, 8–13, available at https://www.autm.net/ AUTMMain/media/ThirdEditionPDFs/V4/TTP_V4_AnatomyLicense.pdf (last visited Oct. 17, 2019). 15 See Reslinski & Wu, supra note 13 (citing the 2014 AUTM survey, approximately 83% of the total gross licensing income received in 2014 derived from royalty payments). 16 As amended, Stanford’s license to Google required both annual advance royalty payments and equity. See The Board of Trustees of the Leland Stanford Junior University and Google, Inc. Amended & Restated License Agreement, SEC, Oct. 13, 2003, available at https://www.sec.gov/Archives/edgar/data/ 1288776/000119312504141762/dex1010.htm (last visited Oct. 17, 2019). 17 Lisa M. Krieger, Stanford Earns $336 Million Off Google Stock, redorbIt, Dec. 1, 2005, available at https://www.redorbit.com/news/education/318480/stanford_earns_336_million_off_google _stock/ (last visited Oct. 17, 2019). 18 AUTM Technology Valuation Course, AUTM Annual Conference, Mar. 12, 2017.

172 Research handbook on intellectual property and technology transfer B.

Types of Deals

1. Patent licenses TTOs often transfer technology to companies via patent licenses. With this model, a university researcher will conceive of an invention during the course of academic research. The TTO will file a patent application to seek to obtain patent protection for the invention. Once granted, a patent will provide the patent holder the right to exclude others from making, using, selling, or offering to sell the patented invention anywhere in the territory of the patent.19 The TTO will seek out companies that might have an interest in commercializing that academic research innovation. A company desiring to commercialize an academic research innovation will desire to license the patent rights associated with that research innovation for at least two reasons. First, but for a license to the associated patent rights, a party commercializing the patented technology would infringe those patent rights upon issuance of a patent. To obtain this objective, the licensee would just need a nonexclusive license, which is a covenant not to sue from the licensor. Second, with an exclusive license, the licensee could affirmatively use the patent rights to exclude competitors from commercializing the patented technology. To obtain this objective, the licensee would need an exclusive license, which provides the licensee the exclusive right to commercialize the patented technology. Patent licenses are the common vehicle for transferring rights in academic research innovations that are therapeutics, medical diagnostics, medical devices, as well as many inventions in the physical sciences. For patent licenses, TTOs will generally seek to follow the general diversified deal structure discussed above, where the university receives economic consideration from multiple categories such as upfront fees, sales-based royalties, annual minimum payments, milestone payments, reimbursement of past and future patent expenses, and in the case of startups, equity or change of control payments. 2. Option agreements Potential licensees will often desire to vet a technology prior to formally licensing the technology from a university. An option agreement allows the potential licensee to vet the technology for a relatively short period of time. This agreement is typically a shorter, simpler contract than a full license agreement which typically extends for the life of the patent rights and therefore has to cover many more scenarios. Accordingly, an option agreement allows an interested licensee to assess a technology without incurring the upfront fees and legal expenses potentially associated with the full license. An option agreement typically grants the optionee a limited license to use the university’s IP to evaluate the commercialization potential of the technology. More importantly, the option agreement typically prohibits the university from licensing the IP to a third party during the term of the option. If the optionee “exercises” its option, it typically has the exclusive right to negotiate for an exclusive or nonexclusive license for a designated period of time. Option agreements are especially common when TTOs are considering licensing IP to a newly formed startup company. The option allows the startup or TTO to recruit management, raise initial capital, and validate a scalable repeatable business model. TTOs will typically

19

Jeffrey Schox, Not So Obvious: An Introduction to Patent Law and Strategy 8 (4th ed. 2015).

Modern intellectual property valuation in the academic technology transfer setting 173 request a single option fee in exchange for option rights. For patent options, TTOs may also require the repayment of patent costs incurred during the option term. Software distribution licenses 3. Research universities create high volumes of software-embodied inventions. For both legal and business reasons, a TTO may choose not to pursue patent protection on a particular software innovation. Even absent patent protection, IP protection still exists for software. First, copyright protection exists for the source code. A second form of IP for software is the likely know how and other technical documentation associated with the code that a licensee would desire to possess. Similar to patent licenses, software distribution licenses may be nonexclusive or exclusive, and also may be field limited. Software distribution licenses may be included with patent licenses in the case of software for which the TTO has filed patent applications. For example, the licensee would need to obtain a license under the university’s patent rights in order to make, use, sell, or offer for sale the patented technology and the licensee would need to obtain a copy of the source code and rights under the university’s copyrights in the code if the licensee desires to utilize the code created at the university. Similar to patent licenses, TTOs will generally seek to follow the general deal structure discussed above, where the university receives economic consideration from multiple categories. With software, however, the TTO’s ability to obtain sales-based royalties, and the magnitude of those royalties will vary significantly based on the commercial readiness of the licensed code. 4. End user licenses TTOs also make software innovations available to licensees via end user licenses (“EULAs”). A EULA differs from a software distribution license because the licensee typically only receives the right to use the software. The license does not provide any rights to further distribute or make derivative works of the code. TTOs will also use EULAs for non-software copyrighted content as well. For example, TTOs can provide medical questionnaires, measuring tools, or scales via a EULA model. Some TTOs also make software and copyrighted content available online through electronic EULAs.20 TTOs will typically seek an upfront and/ or annual payments for EULAs permitting commercial uses of the code. TTOs will routinely provide different tiers of EULAs for various potential uses of the software or content. For example, a TTO could offer an academic use EULA for no fee but a commercial use EULA for an upfront free and annual payments. 5. Biomaterials licenses Universities may create valuable biomaterials which industry partners will desire to obtain and license. Such biomaterials may include: cell-lines, plasmids, mouse models, antibodies, or other human-derived specimens. A biomaterials license will typically require the university to transfer the material to the licensee and then grant the licensee the right to use the material for certain purposes.21 Biomaterials licenses, as compared to material transfer agreements 20

For example MIT’s online licensing portal is here: MIT, available at https://tlo.mit.edu/portfolios/ ready-to-sign (last visisted Mar. 27, 2019). 21 An example of WARF’s biomaterials license is included on WARF’s website: WARF, available at https://www.warf.org/media.acux/c49bdad2-f8af-4062-a2db-edbbb556e3b5 (last visited Mar. 27, 2019).

174 Research handbook on intellectual property and technology transfer discussed below, typically include running royalties or other payments that are treated as licensing revenue by the TTO. Because TTOs want to ensure broad availability of these tools to the research community, the licenses granted to companies for these biomaterials are often nonexclusive. 6. Material transfer agreements Tangible materials can also be made available to industry partners through Material Transfer Agreements (“MTAs”). TTOs commonly use MTAs to transfer biologic materials such as reagents, cell lines, plasmids, or vectors,22 although MTAs can also be used for chemical materials or even devices. MTAs differ from biomaterials licenses only in that MTAs typically only seek recovery for the costs of preparing and shipping the materials, and typically do not contemplate a commercial use of the biologic material.

IV.

HISTORICAL COMMON VALUATION METHODOLOGIES USED BY TTOS

The number of valuation techniques applicable to IP assets is large and growing. This section will focus on three historically common methods: the cost method, the market-value approach, and the income approach.23 Numerous other valuation methodologies exist apart from and within these three very general examples.24 This Chapter highlights these three principles because they have been long discussed in the world of academic technology transfer and their underlying rationales reflect issues that often arise in OTT negotiations. The methods and principles discussed in this section are deal structure agnostic. There are multiple ways for an industry licensee to share value with a university licensor regardless of deal structure. These methods and principles are available to TTOs and their stakeholders to assist in thinking about the magnitude to which the university should share in the value of a licensed IP asset in the hands of a commercial partner seeking to commercialize a product or service covered by that IP asset. A.

Example of Valuation Scenario

For the following discussion of valuation methodologies, we will use the following example. A university researcher conceives of an invention from the support of approximately $2 million

22 Quick Guide to Material Transfer Agreements at UC Berkeley, u.C. berkeley sponsored prograMs offICe, 2018, available at https://spo.berkeley.edu/guide/mtaquick.html (last visited Oct. 17, 2019). 23 Doy Solomon & Miriam Bitton, Intellectual Property Securitization, 33 Cardozo arts & ent. l.J. 272 (2015), citing Robert Pitkethly, The Valuation of Patents: A Review of Patent Valuation Methods with Consideration of Option Based Methods and the Potential for Further Research 1, 10–21 (Univ. Cambridge Judge Inst. Mgmt. Studies, Working Paper No. 21/97, 1997). 24 Bruce Burton, Scott Weingurst, & Emma Bienias, Common Errors Committed When Valuing Patents: Part 2, Focus on the Income Approach, stout.CoM, available at https://www.stoutadvisory .com/insights/article/common-errors-committed-when-valuing-patents-part-2 (last visited Mar. 21, 2019) (“In addition, there are multiple valuation methodologies within each approach and many methodologies that incorporate elements of more than one approach.”).

Modern intellectual property valuation in the academic technology transfer setting 175 of federal funding. The invention generally comprises an air monitoring device for detecting the presence of various toxins in air. After reporting the invention to the university’s TTO, the TTO spent $45,000 pursuing patent protection. Now three years after the TTO received the invention report, it has pending applications in the United States and at the European Patent Office and is optimistic about its chance of receiving commercially valuable patent claims. The researcher received a $200,000 translational grant award from the university’s internal translational research fund, of which $100,000 was a matching grant from the researcher’s department. These funds were used to develop software to implement a detection algorithm for various markers. The level of software development is sufficient for proof of concept but the code will have to be entirely rewritten for any commercial implementation of the product for a medical application. The invention has applications in the medical diagnostic (patient monitoring) and environmental space. Comparable royalty rates for this product, taken from a variety of databases, show an average royalty rate of 3.5% for medical diagnostics and 4% for environmental applications. The regulatory pathway for the medical diagnostic’s most likely intended use is relatively light as a Class I exempt device. The markets in both applications are large and growing, however it is a relatively crowded patent landscape. Additionally, at least in the medical device application it is likely that the device covered by the university’s IP will be combined with other diagnostic IP and sold as a kit. The following sections demonstrate how a TTO could apply various valuation methodologies to the above scenario. B.

Cost Approach

There are at least two different valuation techniques that commentators refer to as the “cost approach.” The first such technique is the historical cost approach, which focuses on the cost the licensor incurred in developing the IP asset.25 The second is the replacement cost approach, focusing on the cost the licensee would incur to develop a replacement product for the IP asset without entering into a license.26 1. Historical cost The historical cost approach takes into account the resources invested to develop the IP asset.27 With this approach, one adds up all costs of creating the asset and then adjusts for present value where appropriate. The theory behind this approach is that a licensor should at least recoup its costs in developing the asset.28 Applying the historical cost approach to our example, one would look at the $2 million of federal funding that supported the conception of the invention. For example, this federal

25

Guido von Sheffef & Martin Zieger, Methods for Patent Valuation, InternatIonal ConferenCe Ip as an eConoMIC asset: key Issues In exploItatIon and valuatIon, July 1, 2005, at 4. 26 Id. 27 Robert Reilly, Intellectual Property Valuation Approaches and Methods, 46 les nouvelles 198, 200 (Sept. 2011). 28 Keith Witek, Intellectual Property Valuation—Cost Model Analysis, 2 Ip law & praCtICe § 22:4 (July 2018). on

176 Research handbook on intellectual property and technology transfer funding might have supported the salaries of the principle investigators, salaries of graduate student research assistants, the purchase of specialized research equipment, and the purchase of consumables.29 A university will typically charge indirect costs as a percentage of the total research budget. Indirect costs cover overhead costs such as administrative support, facility operation and maintenance, equipment, and research libraries.30 However, some research sponsors, such as philanthropic organizations, do not permit indirect cost reimbursement31 and therefore the research budget will not reflect the total cost of the research that led to an invention. The historical cost approach would likely also point to the $200,000 in translational gap funding used to develop this technology. Translational funds seek to de-risk “embryonic technologies” emerging from research labs to the point where an industry partner will license the technology or an investable startup can be created to commercialize the technology.32 The Coulter Translational Partnership and Research Awards is a long-standing program for accelerating the development of biomedical engineering innovations at several top research universities.33 These translational funds are grants internal to the university for use by or through the principle investigator and therefore are non-dilutive to the equity in any startup that might eventually commercialize the funded technology. Lastly, the $45,000 in patent costs could likely count as historical costs and be applied in valuing the IP asset being licensed by the TTO. However, most TTOs will seek direct reimbursement of these patent costs in their license.34 For the above example, the historical cost approach might look at $2.2 million of costs that went into the development of the IP asset as a measure of its value. The historical cost approach is particularly useful where university administration or faculty specifically desire to recoup the costs of developing the asset. For instance, a department might have made a specific investment in developing an asset and would like to know that it has at least broken even on its investment. 35

29

The author’s school provides a sample research budget which illustrates the types of costs typically incurred in academic research: available at http://orsp.umich.edu/sample-budget-table (last visited Oct. 17, 2019). 30 Heidi Ledford, Indirect costs: Keeping the lights on, 515 nature 327, 329 (2014), available at https://www.nature.com/news/indirect-costs-keeping-the-lights-on-1.16376 (last visited Oct. 17, 2019). 31 Id. 32 Laura Schoppe, Advice for University Translational Research to Close the Valley of Death, fuentek’s teCh transfer blog (Sept. 2017), available at https://www.fuentek.com/blog-post/ university-translational-research-to-close-valley-death-trends-advice/ (last visited Oct. 17, 2019). 33 Coulter Translational Partnership (TP) & Research Awards (CTRA), wallaCe h. Coulter, (2018), available at http://whcf.org/coulter-foundation-programs/translational-research/coulter -translational-partnership-tp-and-research-awards-ctra/ (last visited Oct. 17, 2019). 34 University Licensing: An Introduction to Licenisng Technology from Universities, fenwICk, 2005, at 4, available at https://www.fenwick.com/FenwickDocuments/university_licensing.pdf (last visited Oct. 17, 2019); see also University of Rochester Model Exclusive Patent License, AUTM, Sept. 22, 2019, at § 7.4, available at https://www.autm.net/AUTMMain/media/About/Documents/RochesterMod elExclusiveLicenseAgreement.pdf (last visited Oct. 17, 2019). 35 Harald Wirtz, Valuation of Intellectual Property: A Review of Approaches and Methods, 7 InternatIonal Journal of busIness and ManageMent 40, 48 (2012) (“In literature sometimes it is stated, that the cost approach determines a minimum value, because no rational investor would pay more for an asset than the price of a property of the same utility.”).

Modern intellectual property valuation in the academic technology transfer setting 177 While it does happen, it is rare for TTOs to license an IP asset for a one-time upfront payment. Accordingly, universities will seldom receive their full historical cost upon licensing a technology, but will instead spread that value out among a variety of repayment mechanisms, including sales-based royalties, annual minimum payments, milestone payments, and change of control fees. Because many of these repayment mechanisms are far from certain to occur, the total payments from a licensee to a university licensor in the event where a licensed product succeeds in reaching the marketplace will likely far exceed the actual historical costs. The historical cost approach provides some benefits to TTOs. First, universities have this information readily available. For inventions deriving from externally funded research, the project budgets inform the amount of investment in the creation of the invention. Second, some have criticized the historical cost approach because the application of the approach to value an invention occurs long after the costs were incurred to create the IP and it is difficult to adjust those historical costs for present value. For academic research innovations, the IP licensed to industry partners is extremely early stage. Accordingly, the costs incurred in creating the IP are often (although not always) close in time to the time of the license. Third, the historical cost approach aligns with the philosophy from the Bayh-Dole Act that universities should be reimbursed from the investment in the IP and be able to redeploy that money into future research. Lastly, use of the historical cost approach also aligns with a perceived shift in university administrators to being more strategic in deploying internal resources to support research activities. University administrators appear to be seeking to ensure that their historical costs in support of certain research activities are at least being recovered. The historical cost approach, however, presents several challenges for TTO valuing IP assets in the modern setting. First, in many cases, universities would choose to bear the historical costs that led to an invention regardless of whether the invention is licensed. While universities value their technology transfer initiatives, the universities have other significant priorities. In other words, in many (but not all) situations, universities would engage in various research activities regardless of the commercial outcome. Indeed, it is not uncommon for a licensee to point to only the TTO’s activities, namely the patent expenses, as the university’s historical costs in developing the IP asset. Licensees will argue that the other larger costs behind the underlying research are sunk costs that the university would have incurred anyway. Second, included in the historical costs a TTO uses to value an IP asset might be certain fixed costs that a university would incur anyway. A third challenge of the historical cost approach, and probably the most significant, is that in the case of inventions, the costs incurred in creating the invention may not correlate at all to the commercial value of the resulting IP asset. The historical costs may either greatly overvalue or undervalue the commercial value of the resulting IP. Accordingly, to truly understand the commercial value of an IP asset being licensed, some understanding of the market norms and the potential value of the IP to the licensee must occur. In practice, the historical cost approach can be useful in grounding the discussion with a licensee and the university stakeholders. In some situations, the historical costs might provide some evidence as to the commercial readiness of the asset. For example, significant internal investment in software development or use of translational funds to de-risk a technology should evidence a higher valuation for the IP asset.

178 Research handbook on intellectual property and technology transfer 2. Replacement cost The replacement cost approach measures the cost needed to create comparable technology in the absence of a license.36 In our example, the replacement cost approach would likely look at the following components: ● The ability of the licensee to recreate the code outside of the university; ● The ability of the licensee to design around the potential patent rights; ● In the absence of a design around the patent rights, the risk that the patent will issue with claims that cover the licensee’s product, and the university will take action against the licensee, or license the rights to a competitor who will; and ● The cost of explaining unclear chain of title in the event the licensee uses the technology provided by the faculty member but takes the position that the university does not own the IP. The author sees the replacement cost raised most commonly in two situations. The first is in negotiations involving unpatented software. In this situation, a licensee will point out that it could develop the code itself outside the university without needing a license. Where this is legitimately possible, the replacement cost will be highly relevant in acting as a ceiling for the value of the IP asset being licensed. The second situation where the replacement cost directly impacts negotiations is where the university patent rights cover features or “nice-to-have” components of the eventual commercial product but do not cover the primary ingredient or component of that product. In these situations, a licensee will appropriately balance the cost of the university license with the costs of proceeding to develop a commercial product without the licensed IP. A fair discussion of the replacement costs should also include the licensee’s cost avoidance from licensing the particular IP asset. Many TTOs can point to the following aspects of a university IP license that will benefit the license and/or save costs. In particular, university IP licenses can: ● provide clean chain of title to the licensee’s IP; ● enable access to the inventor for valuable consulting services; ● provide brand value through the licensee’s association with the university and prominent faculty inventors; and ● allow access to university resources such as university-run investment funds, accelerator space, or even use of the university brand. In practice, discussion application of the replacement cost approach to IP valuation evolves into a specific assessment as to the value of the IP being licensed. IP that covers the core ingredient of the future commercial project—such as for example, a composition of matter patent on the active ingredient in a therapeutic—has an extremely high replacement cost and a corresponding high valuation. IP that could be designed around (such as narrow patent claims), recreated (such as early stage unpatented software), or covers an optional function of the technology has relatively low replacement costs and a corresponding low valuation. In situations where replacement costs are low, TTOs should be realistic in the economic terms they seek to extract from interested licensees.

36 Id. (“[T]he replacement cost method estimates the cost of the production or purchase of a good with an equivalent benefit.”).

Modern intellectual property valuation in the academic technology transfer setting 179 C.

Income Approach

The income approach to IP valuation attempts to value an IP asset based on the future benefits it will provide the licensee. These benefits could be in the form of additional economic receipts or cost savings.37 For example, this method might attempt to estimate the future income streams from the licensee’s commercial use of the licensed IP asset and then discount those future cash flows to capture a present value of the IP asset. Applying the income approach to the above example, a TTO could look at the large and growing markets in both the medical diagnostic and environmental applications. It is not uncommon for TTO interns to return market analysis reports with top-down market assessments in the billions of dollars. On the flip side, potential licensees will talk about the small and crowded market place. Better market assessments are derived from a bottom-up market analysis.38 The TTO may formulate assumptions about the future cash flows or cost savings available to the licensee from the licensed IP asset. However, a hallmark of university licensing is that it involves extremely early stage technologies. Therefore, understanding the future value the licensee will derive from the IP asset can be difficult due to the early stage nature of most university technologies. For the typical TTO license, the licensee is many years from using it. Accordingly, a TTO may have a general sense of the future income or cost savings a licensee will enjoy from an IP asset. The extreme early stage of most university IP assets, however, will prevent the TTO from ever having a specific enough understanding of future cash flows in order to apply a discounted cash flow analysis to yield a specific IP valuation. In addition to needing to estimate future cash flows, the income approach also requires adopting a discount rate. The discount rate is an approximation of the risk involved in bringing the technology to market. Because all IP valuation involves apportioning the risk between the licensee and licensor, this risk apportionment applies to all valuation methodologies at some level. The risk associated with an early stage technology ultimately making it to market can take the form of regulatory risk, technology risk, IP risk, or market/competition risk. For example, regulatory risk directly correlates with the amount of regulatory approval required for a product to comet to market. A Class III device will have greater regulatory risk than a Class 1 exempt device and therefore a higher discount rate. It follows that a university’s effort to de-risk a technology should generally correspond with a higher valuation for that technology because the licensee’s future income stream from that asset becomes more certain. D.

Market Approach

The market approach looks to comparable transactions in an active marketplace for information about the value of a particular asset.39 A direct market approach looks at other transaction

37

Sheffef & Zeiger, supra note 25, at 5. For a comparison of top down and bottom up market sizing approaches, see Towards Data Science, Sizing Up: Market Sizing for Your Business, towards data sCI., 2017, available at https:// towardsdatascience.com/sizing-up-market-sizing-for-your-business-c569e45730ef (last visited Oct. 17, 2019). 39 Weston Anson, Want to Value Your Intellectual Property? Here are Three Approaches, Ip In brIef: trends and transforMatIons In CopyrIght & tradeMark (June 6, 2012). 38

180 Research handbook on intellectual property and technology transfer prices for the subject IP asset.40 For example, if a particular patent has been nonexclusively licensed by a university, then subsequent nonexclusive licenses might be valued based upon the prior licenses for that patent. An indirect market approach looks at similar transactions for similar IP assets. For example, for a licensor looking to value patent rights related to an oncological therapeutic in negotiating an exclusive license, the licensor might look at the average economic terms for oncological therapeutic exclusive licenses from other universities. Using our example, the direct market approach would look to see whether this IP asset had been previously licensed. Given that it had not, the licensor might look at indirect market information, such as the average royalty rate of 3.5 percent for medical diagnostic technologies and 4 percent for environmental applications. An interested party might obtain this information from a number of sources. AUTM keeps a database of economic terms from member licenses. Licensing Executives Society (“LES”) keeps a similar database. More recently, Osage University Partners maintains a database of startup licensing terms from its member universities. Other sources also exist. The market approach can be useful in grounding a discussion on royalty rates or other terms for which comparable information exists (e.g., equity amounts). It is not uncommon for a negotiation between a TTO and a potential licensee to involve a disagreement over whether a position is consistent with the “market.” Licensees tend to underestimate the amount of market data available to TTOs and the level of benchmarking that goes on between similar TTOs. When a TTO shares this information, it can assist in informing the potential licensee that there is a basis for the proposed terms. The market approach, however, is limited in the context of university IP licenses because the underlying IP asset in every deal is different. In our above example, the average royalty rate of 3.5 percent is broadly taken (hypothetically) from the category of “medical devices.” This broadly includes Class III implants facing a long regulatory pathway and relatively simple Class I exempt devices that might be close to being ready for commercial use. That average royalty rate also includes both a license with a six-figure up-front payment (and a correspondingly low running royalty rate), and a license with no up-front payment and royalty cap of $10 million (with a correspondingly higher royalty rate). Further, it includes a license for issued US and foreign patents with commercially valuable claims on the core aspect of the technology, and also a pending application in the US only that is likely to issue with extremely narrow claims if it issues at all. In other words, every deal is different and the true value of the rights being transferred in a particular deal (even within a single technology vertical) can vary significantly based on a number of factors. Even if one had full visibility to all of the terms of all other IP licenses, it still would be of questionable value. IP valuation is an imprecise science and there are nuances to each deal that impact the economic terms. Accordingly, for university IP licenses, the market approach may be useful for grounding a discussion but it is significantly limited in its ability to specify a particular value for an IP asset. As explained by one commentator: “[T]he estimation made based on the market method is largely based on the hope that people who valuated the comparable asset knew better than one self and their analysis can be applied to the [current] valuation situation. That is why the market method is not recommended to use as the only method.”41 40

Wirtz, supra note 35, at 42. Donna P. Suchy & Weston T. Anson, Fundamentals of Intellectual Property Valuation: A Primer for Identifying and Determining Value, aM. b. ass’n seCtIon of Ip l. 11 (2005). 41

Modern intellectual property valuation in the academic technology transfer setting 181 E.

Applying Common Valuation Methods

Rather than serving as rigid formulas to derive specific valuations, the above methodologies instead serve as guideposts for discussions around the appropriate economic terms in a license negotiated by a TTO. Using the above example, the following shows how the historical valuation methods may be used in a discussion about licensing terms between a TTO licensing manager and an interested licensee. Upon identifying the licensee, the TTO licensing manager will discuss the opportunity with the inventors. It is not uncommon for inventors to have an extremely high opinion of the value of their IP, perhaps pointing to the significant federal funding behind the invention (i.e., employing the historical cost approach). In many cases, translational funding programs can help the inventors develop a well-rounded assessment of the value of their IP. Many of these programs encourage or even require customer discovery that informs the inventors of alternative technologies in the marketplace and the true commercial risks associated with early stage result of academic research. A TTO licensing manager may commonly use an indirect market-based valuation approach by showing database results for average economic terms for this general class of technology. At this time, the TTO licensing manager may discuss with the inventor reasons why the TTO should propose terms above or below the market terms. For example, the IP position might be particularly strong or limited, or the regulatory pathway might be more or less complicated than the TTO believes the market averages to assume. At this point in time, the TTO licensing manager may attempt to ascertain information that will assist in applying the income-based approach to valuation. As discussed above, one challenge of the income-based approach is that university IP is typically extremely early stage so it is often difficult to understand the nature of the eventual products to be commercialized, the realistic projected sales figures, and associated profit margin. Here, again, sometimes the university’s translational funding program will have assisted in developing some useful market and sales projects. Additionally, a common tactic for TTO licensing managers is to ask the prospective licensee for a business model that include sales and market projections. The TTO can request this information as part of qualifying the licensee, but can use the information in valuing the technology under an income-based approach. The TTO licensing manager may rely heavily on information about comparables from its market-based approach and adjust accordingly based on its belief in the credibility of the income-based assessment. From this information, the TTO licensing manager will generate a term sheet that it will share with the licensee. A licensee will typically respond with a proposal for some adjustment in the economic terms. It is not uncommon for there to exist a significant gap in valuation. A potential licensee arguing for a significantly lower valuation of the IP may, for example, suggest two over-arching principles. First, the licensee may suggest that the proposed economic terms are much different than the market terms to which the licensee has access. For example, the potential licensee may suggest that the proposed economic terms are much different than those it has seen from other universities for comparable IP. Second, the licensee will suggest that the replacement costs are low. For example, the licensee may argue that the IP is suspect and unlikely to issue, covers unnecessary and unimportant features, or can easily be reproduced. Regarding the licensee’s first assertion concerning market terms, licensees often underestimate the amount of market data accessible to university TTOs. TTOs can often share some information about the access they have to market terms to address these concerns. However, as

182 Research handbook on intellectual property and technology transfer discussed above, any discussion of indirect market data is necessarily generalized and specific points about how the technology at hand differs from the assumed scenarios in the market data should be taken into account. Regarding the licensee’s second assertion about the replacement value, this assertion will typically trigger a more precise analysis of the status of the IP. In the end, this analysis typically falls into one of two categories. Either the IP is likely to cover the core aspects of the potential commercial product, or the IP is of a more questionable nature. If the former, then TTOs are justified in seeking a strong economic return for their IP. If the latter, then TTOs should be realistic in the economic terms they seek. It is often at this point when the TTO licensing manager might again look to the historical costs and the income projections as supporting a certain level of valuation and economic return. The above back and forth shows how the historical valuation approaches can serve as guideposts and talking points in the context of an IP license negotiation. These approaches rarely serve as absolute quantitative methods but instead point towards general levels of value for an IP asset that can then be apportioned in a license across the various categories of economic terms considered in the discussion on deal structures above. F.

Bridging a Valuation Gap

Even when a material valuation gap exists for an IP asset, other mechanisms exist for bridging the valuation gap in order to complete the deal. While the possibilities are nearly limitless, the following are some common approaches employed by TTOs to adjust other deal dynamics in order to finalize a license. 1. Royalty stacking or combination product language A common negotiation point from licensees attempting to support lower economic terms is that the technology exists in a crowded field and the licensee will have to license other IP. Indeed, in the hypothetical scenario used in this Chapter, the licensee will likely argue that the IP landscape for the technology is crowded and therefore the value of the university’s IP is low. While this point might have merit, it can be addressed by common deal language. For example, “royalty stacking” language can allow for a reduction in the royalties owed by the licensee if the licensee must license other third party IP in order to sell products covered by the university’s IP. This language typically includes a floor below which the royalties owed to the university will not fall. A similar, but different point, is that the IP covers just one component of an eventual product and the component protected by the university’s IP will be sold in combination with other components. Indeed, in the hypothetical used in this Chapter, the licensee will likely argue that the fact that the licensed product will be sold as part of a kit with other products should warrant a lower valuation of the IP. As with the royalty stacking principle, a well-established mechanism exists for addressing this concern. “Combination product” language provides a formula for adjusting the royalty base when the product covered by the university’s IP is sold in combination with other separately saleable products. Such “royalty stacking” and “combination product” language are well known mechanisms for addressing common points raised by licensees seeking lower economic terms.

Modern intellectual property valuation in the academic technology transfer setting 183 2. Adjusting the timing or length of payments The timing and nature of the payments can be adjusted. For example, payments can be shifted later in time, so that the university bears more of the commercialization risk. For example, instead of an upfront payment, a higher royalty rate could be adopted. Royalties could be defrayed for a period of time in order to justify a higher royalty rate. A TTO may accept a lower royalty rate in exchange for receiving post-patent expiration royalties on know-how. Alternatively, payments can be contingent upon certain events occurring, such as a commercialization milestone or an IP milestone. 3. Adjusting the scope of rights being granted The scope of rights being granted may be altered through adjusting the exclusivity, territory, or field of use terms. For example, a TTO may become comfortable with a licensee’s proposed lower economic terms if the license is converted to nonexclusive or is field-limited. Similarly, rigorous diligence milestones may de-risk the license from the university’s perspective and support the adoption of the licensee’s proposed terms. 4. Extra-license mechanisms Other aspects of the relationship between the university and the licensee may be adjusted in order to bridge a valuation gap concerning the licensed IP. For example, a licensee agreeing to support research in the inventor’s lab through sponsored research may be desirable to the licensee and permit lower licensing terms. Similarly, the licensee hiring the faculty member as a consultant may entice a reluctant faculty member to gain comfort with the licensee’s proposed deal terms. In either of these cases, care should be taken by the TTO to make sure it is receiving the proper sign-offs before sponsored research or consulting are included as part of a licensing arrangement.

V.

TWO SPECIAL CASES OF VALUATION

This section addresses two particular valuation issues that commonly arise for TTOs. One is how to address a royalty buyout proposal. The second is how to think about valuation in terms of equity in a startup licensee. A.

Royalty Buyouts

TTOs are frequently confronted with options to have their royalty streams bought out. Several universities have received lucrative buyouts of royalty streams. For example, in 2015, Emory sold its royalty rights to Emtriva to Gilead Sciences and Royalty Pharma for $525 million.42 UCLA, in 2016, sold its royalty interest in prostate cancer medication Xtandi to Royalty Pharma for $1.14 billion.43 The purchaser of the royalty stream could be a third party that spe42 Gilead Sciences and Royalty Pharma Announce $525 Million Agreement with Emory University to Purchase Royalty Interest for Emtricitabine, eMory u., 2015, available at https://www.emory.edu/ news/Releases/emtri/ (last visited Oct. 17, 2019). 43 Phil Hampton, UCLA sells royalty rights connected with cancer drug to Royalty Pharma, UCLA u. news, Mar. 4, 2016, available at http://newsroom.ucla.edu/releases/ucla-sells-royalty-rights -connected-with-cancer-drug-to-royalty-pharma (last visited Oct. 17, 2019).

184 Research handbook on intellectual property and technology transfer cializes in buying out royalty streams, as in the above examples. The purchaser could also be the licensee or the licensee’s potential acquirer. These proposals raise complexities for a TTO due to the number of internal stakeholders that benefit from licensing revenues and the typical asymmetry of information between the TTO and the potential purchaser. The typical proposal involves the purchaser offering a lump sum payment in exchange for the right to receive all or a part of the university’s royalty stream from a license. A buy-out benefits the university by accelerating and de-risking the receipt of compensation. The downside to the buy-out is that the university will potentially receive less total compensation than had it retained its royalty stream. Analysis of a buy-out option will necessarily involve analyzing the net present value of the future royalty stream. Fortunately for TTOs, consultants are available to assist in this analysis and bridge the asymmetry of market information between the potential purchaser and the university. Because numerous firms engage in royalty buy-outs, it is often possible for a university to set up a competitive bidding scenario for its royalty stream.44 Additionally, a TTO may benefit from exploring a hybrid buy-out structure. For example, a university can sell a portion of its royalty stream (e.g., 50 percent of the royalties it receives), allowing it to maintain some of the upside of future royalties.45 This will result in a lower upfront payment, but it may allow the TTO to experience the benefits of accelerating and de-risking the future payments while maintaining the potential for significant (although lowered) future royalties. B.

Equity

When a TTO licenses IP to an early-stage startup, it is common for the university to receive equity in the licensee. Equity raises unique valuation considerations for TTOs. This section explains some of the considerations for TTOs attempting to assess the amount and type of equity to seek to receive in a startup licensee. 1. Type of equity Equity can come in several different forms. TTO startup licensees typically take the form of C corporations or limited liability companies.46 Equity in a corporation is typically referred to as “shares” or “stock.”47 Equity in a limited liability company will typically be referred to as membership interests or units.48 Startups that do not foresee raising institutional capital in the short-term may start as limited liability companies in order to benefit from the flow-through taxation49 and the lower filing fees in most states.50 A startup seeking to raise institutional

44 Some firms include DRI Capital, PDL Pharma, Healthcare Royalty, Royalty Pharma, Capital Royalty Partners, Oberland Capital, and Paul Capital. 45 Reslinksi & Wu, supra note 13 (providing examples of hybrid royalty buyout scenarios). 46 Constance E. Bagley & Craig E. Dauchy, The Entrepreneur's Guide to Business Law 67–76 (4th ed. 1998). 47 Id. at 91–2. 48 Stella Lellos & Rivkin Radler, Choice and Formation of Limited Liability Companies, a praCtICal guIde to llCs, 20171117A NYCBAR 1. 49 Bagley & Dauchy, supra note 46, at 62–7. 50 Instruction No. 10, MICh. dep’t lICensIng & reg. aff. Corps., seCs. & CoM. lICensIng bureau, available at https://www.michigan.gov/documents/lara/700_08-15_528193_7.pdf (last visited Mar. 21, 2019).

Modern intellectual property valuation in the academic technology transfer setting 185 capital in the near future will likely incorporate as a C corporation because most institutional investors cannot invest in flow-through entities and will insist their portfolio companies organize as C corporations (and typically in Delaware).51 In almost all states, mechanisms exist for converting a limited liability company into a corporation (in the same state or a foreign state).52 In such conversions, membership interests of the LLC will convert into shares of the corporation according to the terms of a conversion plan.53 In general terms, stock in a C corporation can be issued as either “preferred stock” or “common stock.” The difference is that “preferred stock” will have certain preferences over the common stock.54 These preferences are listed in the Certificate of Incorporation and other organizational documents. Common preferences include a liquidation preference, special voting rights, potentially a board seat, and protective provisions against certain actions occurring without the preferred shareholders’ consent.55 Most investors insist upon receiving preferred stock in exchange for investing capital in a startup. It is commonly said that investors receive preferred stock and operators and service providers receive common stock. In the context of a TTO license to a startup, the major economic difference between having preferred and common stock occurs in the case of an exit that is less than a home-run exit. In such an exit, it is common for the liquidation preference of the preferred stock to kick-in and provide the preferred shareholders a disproportionate return as compared with the common shareholders. For example, a simple hypothetical might involve a startup that is acquired for $10 million. Assume investors have invested $5 million in the company and hold 40 percent of the equity in the company in the form of preferred stock. This preferred stock has a 1x, non-participating liquidation preference. In this case, the preferred shareholders could either convert their preferred shares to common shares and participate along with the common shareholders in sharing the $10 million acquisition price. In this case the preferred shareholders would receive 40 percent of the $10 million or $4 million. The common shareholders would receive the remaining $6 million. Because the preferred stock includes a 1x liquidation preference, however, the preferred shareholders are entitled to elect to instead receive their investment back instead of converting to common stock. In this scenario, the preferred shareholders would receive $5 million (1x of their initial investment) and the common would receive $5 million. Compare a large exit of $100 million under the above facts, where the preferred shareholders would undoubtedly convert to common and take their $40 million (40 percent of the acquisition price) rather than merely receiving their initial investment. One of the long-standing questions of academic technology transfer is whether TTOs should receive preferred or common stock when receiving equity in a startup license. On one hand, a university has funded the creation of the IP that the startup is licensing and in this sense it might view itself very much like an investor. This might be especially true of a university using a cost-based valuation approach towards its IP. This might also be tempting for a univer51

Bagley & Dauchy, supra note 46, at 67 (“An LLC is not suitable for business financed by venture capital funds because of tax restrictions on the funds’ tax-exempt partners.”); see also Brad Feld & Jason Mendelson, Venture Deals 182 (2d ed., 2016) (explaining why Delaware is a common state of incorporation for startups). 52 Converting LLC to C Corp: Everything You Need to Know, upCounse, available at https://www .upcounsel.com/converting-llc-to-c-corp (last visited Mar. 21, 2019). 53 Id. 54 Feld & Mendelson, supra note 51, at 41–7. 55 Id., at 41–72.

186 Research handbook on intellectual property and technology transfer sity that has “invested” in the development of the IP through its internal translational or gap fund programs. On the other hand, the creation of the IP also very much resembles services provided by operators of the startup. Had the inventors been founders of the startup prior to the time of invention and had the invention occurred while working for the startup, the inventors, as operators of the company, would have likely received common stock for their efforts. It is worth noting that the distinction between preferred and common does have economic consequences. It is well-recognized that preferred stock is more valuable than common stock. Startups routinely have independent third parties value their common shares for purposes of setting exercise prices on stock options. These valuations are referred to as “409A valuations” after the section of the tax code that the valuations are meant to address.56 It is common for these valuations to value the company’s common stock at 20–30% of the price percent per share of the preferred stock issued in the company’s last financing round.57 2. When to receive the equity Another common question for TTOs receiving equity in a startup license is when to receive the equity. The timing matters because of something called dilution. Dilution refers to the common situation where a company issues additional equity over time, thus diluting the existing equity holders. While an existing equity holder’s number of shares will remain the same, because the total number of shares in the company increases, the exiting equity holder’s percentage of holdings will decrease. For example, if a university is issued 500,000 shares in a startup that has issued 10 million shares, the university will hold 5 percent of the company. Let us assume the company then raises a venture capital round in which it issues 8 million new preferred shares to investors and creates a new option pool with 2 million new shares.58 After that financing, the university still holds 500,000 shares; however, now its holdings comprise 2.5 percent of the company (because it holds 500,000 of the 20 million issued and outstanding shares). Because many TTOs typically license to startups at extremely early stages, it is highly likely that the startup will issue additional equity to either raise capital or hire additional employees. Accordingly, it is highly likely that any equity issued at the time of license will experience significant dilution. It is worth noting that dilution is not necessarily a bad thing. At the end of the day, what matters most is the value of one’s equity in the startup, not the percentage of the company 56

Feld & Mendelson, supra note 51, at 184–5. Id. 58 While a detailed discussion of option pools is beyond the scope of this Chapter, it is well documented that the option pool is an economic term in venture capital financings and it is common for an option pool to be taken from the common rather than the preferred shares when allocating value during an equity financing. For example, let’s assume a company will raise $1 million of venture capital at a $4 million pre-money valuation. This means that the post-money valuation of the startup will be $5 million. Under this scenario, the venture capitalists will receive 20 percent of the company in return for their investment (taking the amount invested and dividing by the post-money valuation). In this example, the terms of the financing require the startup to create an option pool comprised of 20 percent of the company’s full-diluted post-money cap table. The is common because the option pool will be used to hire new employees. Accordingly, the post-money cap table will be divided as follows: 60 percent for the existing common shareholders, 20 percent for the option pool, and 20 percent for the preferred shareholder investors. See Babank Nivi, The Option Pool Shuffle, venture haCks, Apr.10, 2007, available at http:// venturehacks.com/articles/option-pool-shuffle (last visited Oct. 17, 2019). 57

Modern intellectual property valuation in the academic technology transfer setting 187 which that equity reflects. So, in the above example, even though the university’s equity position decreased from 5 percent to 2.5 percent, the value of that position likely increased significantly. To account for dilution, universities take one of three common approaches. The first is to receive equity at the time of license and account for the fact that it might be diluted significantly in the early organizational and financing activities of the company. The benefits of this approach are that the university is likely entitled to a higher percentage of equity. A significant disadvantage of this approach is that it is difficult to predict how this equity could be diluted in the early stages of the company. For example, if the startup issues 30 percent more equity in order to bring on a new founder, the university’s position will be diluted by that founder’s hiring even prior to the routine financing rounds that are likely to occur later. The second approach is to receive equity with some protection against dilution up to a certain milestone (commonly an early financing event). For example, a university could receive 5 percent equity in a startup licensee and receive the right that if the startup issues additional equity, then the university will receive additional shares to maintain its 5 percent position until the startup has raised a certain amount of capital. Alternatively, the university, in its license, could simply say that it receives 5 percent of the company based on the post-money cap table of the company at the conclusion of a certain financing milestone. This second approach protects the university from the early organizational equity issuances of the company, yet aligns the university with the founders and early investors at the conclusion of an early financing round once the cap table is more firmly established and there are outside directors overseeing equity issuances. A third approach is to receive a “liquidation fee” in lieu of equity. For example, a TTO could include a provision in its license that it receives 2 percent of the amount received by the company as part of a “liquidation event.”59 In this third approach, the university’s “position” of 2 percent would not be diluted in any way. This approach is often attractive to the startup founders who focus mostly on the percentage (which is presumably lower than the percentages in the first and second approach) but do not fully understand the mechanics of dilution. Many investors express discomfort with the third approach due to the fact that the university’s position is not aligned with the founders’ or early investors’ interests since they do not experience any dilution. Additionally, it is possible that the university’s liquidation fee would take a liquidation preference over the preferred stock of the investors. How much equity to receive 3. Another fundamental question facing universities receiving equity from a startup licensee is how much equity to receive. On one hand, a TTO taking a cost-based valuation approach to its IP might be tempted to look at the amount of money invested in the creation and development of the IP and ask how much equity it would have received had it invested that capital in the startup licensee prior to the date of invention. Using the above example, the TTO could look at the $2 million of federal funding and the $200,000 of translational funds and attempt to value the equity position to which it is entitled based on that $2.2 million “investment.” This approach poses several problems. First, as discussed above, the cost based valuation approach risks over-valuing the university’s investment in the IP. Second, and more importantly, there 59 Yale Startup License, yale off. CooperatIve res., 2018, available at https://ocr.yale.edu/ faculty/startup-support/yale-startup-license (last visited Oct. 17, 2019).

188 Research handbook on intellectual property and technology transfer is only so much equity that a startup should reasonably issue to non-operational entities. Conventional wisdom is that the large majority of a startup’s equity should be issued to those providing future effort to creating a successful outcome.60 Equity issued for past contributions is viewed as “dead equity” and not looked upon favorably by investors. Accordingly, universities will realize that their equity only has value if the startup experiences a successful exit, and a successful exit is only possible if enough equity is available to attract, hire, and retain the operators necessary to work towards that successful exit. This likely precludes universities from directly valuing the equity they receive in a startup based on the university’s prior investment in the IP. A third problem with this approach is that the value of the startup at the time of the license is likely low, or even incalculably low. For example, it is common for startups to make initial issuances of equity to founders at a value of $.0001 or even $.00001 per share.61 Therefore, any non-trivial dollar value that a university assigns to the current value of the equity it is seeking to receive will likely lead to a disproportionately high percentage of the company going to the university at least prior to an equity financing event that establishes a higher value for the company’s equity. Recognizing the above, most universities will seek a single-digit percentage of equity in a startup licensee. The exact amount of that equity may be impacted by the above historical valuation approaches, but universities will not attempt to directly assign a current dollar value of the equity it is receiving in a pre-financing startup. Instead, TTOs will recognize that the equity they are receiving will provide for a significant return in the event the startup has a successful exit. If the startup does not have a successful exit, the equity is worthless. 4. Participation rights It is increasingly common for universities to seek “participation rights” associated with their equity holdings in startup licensees. Participation rights are sometimes called “rights of first refusal,” “pro rata rights,” or “purchase rights.” Participation rights allow the university to purchase equity in future financing rounds. For example, some universities request the right to purchase up to 10% of the securities sold in a future financing round.62 The value of participation rights is that a university (or its designee) will have the ability to purchase additional equity in a startup licensee that is experiencing success. This allows the university to counteract the dilutive impact of future financing rounds. While the university must invest capital in order to acquire the additional equity, many universities have either established internal

60

Drew Hendricks, 6 Tips for Successfully Splitting Equity in Your Startup, Entrepreneur, Apr. 24, 2015, available at https://techmeetups.com/6-tips-for-successfully-splitting-equity-in-your-startup/ (last visited Oct. 17, 2019). 61 Ryan Roberts, Par Value for a Startup Company’s Stock, startup law., Oct. 11, 2008, available at https://startuplawyer.com/incorporation/par-value-for-a-startup-companys-stock (last visited Oct. 17, 2019); see also Yoichiro Taku, What is Par Value?, startup CoMpany law., July 18, 2008, available at http://www.startupcompanylawyer.com/2008/07/18/what-is-par-value/ (last visited Oct. 17, 2019). 62 Exclusive License Agreement, stan. off. teCh. lICensIng, § 7.2, available at https://otl.stanford .edu/sites/default/files/exclusive_03-06-2018.pdf (last visited Mar. 21, 2019).

Modern intellectual property valuation in the academic technology transfer setting 189 investment funds63 to make this investment or have entered into partnerships with external venture capital funds.64

VI.

A PORTFOLIO APPROACH TO TECHNOLOGY TRANSFER LICENSING VALUATION

The historical valuation techniques described above are all aimed at ascertaining the value of specific IP assets included in specific licenses. A supplement to this approach, if not an alternative, is to take a portfolio approach to valuation. In other words, rather than trying to specify and optimize the value received from each license, a TTO can seek to maximize the value of its license portfolio by maximizing the number of licensed assets in that portfolio. While the term “portfolio approach” has different meanings in the context of valuation, this Chapter uses the term to mean an approach to deal-making aimed at maximizing the collective value of an entire portfolio of assets rather than attempting to maximize the value received for a particular asset within that portfolio. A.

Motivations for Adopting a Portfolio Approach

Several motivations might exist for TTOs to take a portfolio approach to valuation. First, it might make economic sense for the TTO to do so. In other words, the long-term economic returns from a TTO’s IP licensing might be higher if the TTO focuses more on maximizing the value of its licensed IP portfolio rather than the economic terms of each potential deal. It might at first seem counter-intuitive that a TTO could achieve better economic returns by not attempting to maximize the economic return on each deal. However, a few points are worth remembering. First, it is well-documented that academic technology transfer is a home-run driven businesses. A recent survey from the Association for University Technology Managers found that less than 40 percent of licenses generated revenue and less than 1 percent generate more than $1 million in annual licensing revenues.65 Accordingly, a TTO is most likely to receive above-average revenues by virtue of having a home run in its portfolio, not by maximizing the economic returns on each license. A TTO’s odds of having a home run in its portfolio are increased if it maximizes the number of licensed IP assets in its portfolio.66 63

The University of Michigan established an initiative called “Michigan Investment in New Technology Startups (MINTS)” to invest in University of Michigan startup licensees. See Nicole Casal Moore, University to invest in its own startup businesses, u. MICh. reC., Oct. 5, 2011, available at http:// www.ur.umich.edu/update/archives/111005/mints (last visited Oct. 17, 2019). 64 Osage University Partners has partnered with over 80 top research universities to invest in spin-outs from those universities through the universities’ participation rights. See Osage Venture Partners, osage partners, available at https://osagepartners.com/osage-university-partners/ (last visited Aug. 31, 2018). 65 Emerging Best Practices in Translating University Research into Innovation, A Vision for the Future of Center-Based Multidisciplinary Engineering Research: Proceedings of a Symposium (2016), at 21–3 (Orin Herskowitz, director of Columbia’s University’s Columbia Technology Ventures cited the AUTM survey and also noted that 90 percent of Columbia’s more than $3 billion in licensing revenues has derived from four patents). 66 This is assuming, of course, that the license provides an uncapped upside for the university through either uncapped royalty payments or an uncapped return on equity.

190 Research handbook on intellectual property and technology transfer Second, for startups in particular, a portfolio approach lends itself to streamlined and perhaps even standardized startup licensing frameworks. This makes it easier to launch more startups through the TTO because investors and entrepreneurial faculty will be more likely to work with the TTO. Standardized deal documents are prevalent in the startup formation and financing scene. In 2003, the National Venture Capital Association (NVCA) released model documents for a Series A equity financing.67 While these documents are rarely used without modifications, they serve as a useful starting point for most venture financings.68 By 2010, multiple sets of standardized seed financing documents have been released online.69 On top of standardized deal documents, new financing mechanisms like convertible notes and SAFE agreements have become routine for early stage financings. Venture oriented law firms routinely make available standardized incorporation documents and document generators through their websites.70 Today, entrepreneurs can incorporate their startup with minimal or no legal fees, and with documents they can readily find online. Further, they can complete early stage financings with standardized documents that are readily accepted. Accordingly, for entrepreneurs that can set up a Delaware C Corp and raise millions of dollars with standardized documents, the notion of having to negotiate a customized license for IP is difficult to rationalize. Even outside the context of startups, industry will be more likely to enter licensing discussions with a TTO with a reputation for having an efficient negotiation process. Third, related to the above, a TTO with a positive reputation with investors and entrepreneurial faculty is more likely to assist the university in attracting additional entrepreneurial faculty to join the university. Attracting additional high-caliber and entrepreneurial faculty and graduate students leads to more technologies capable of being the basis for home-run startups or licenses to existing companies. B.

Examples of TTOs Adopting a Portfolio Approach

This section provides examples of situations where TTOs have adopted a portfolio approach to valuing university IP assets. The first category of examples is comprised of some of the numerous schools that have adopted fixed startup licensing terms. The second category of examples is comprised of some of the universities, such as the University of Minnesota, that have adopted fixed IP licensing terms for IP resulting from industry sponsored research agree-

67

Jonathan D. Gworek & Scott R. Bleier, The Evolution of the NVCA Documents: A Brief Description of the Changes to the Crowdsourced Gem of Venture Capital Practice, vC spotlIght, Dec. 29, 2015, available at http://www.mbbp.com/news/evolution-of-nvca-documents (last visited Oct. 17, 2019). 68 As explained by one prominent venture attorney guest lecturing in the author’s class at Michigan Law School, on the West Coast, the law firms take the NVCA documents and make them more entrepreneur-friendly. On the East Coast, the law firms take the NVCA documents and make them more investor-friendly. 69 Brad Feld, The Proliferation of Standardized Seed Financing Documents, feld thoughts, Mar. 1, 2010, available at https://www.feld.com/archives/2010/03/the-proliferation-of-standardized-seed -financing-documents.html (last visited Oct. 17, 2019). 70 See, e.g., Orrick, available at https://www.orrick.com/Total-Access/Tool-Kit/Start-Up-Forms (last visisted Mar. 21, 2019); Founders Workbench, available at https://www.foundersworkbench.com (last visisted Mar. 21, 2019); COOLEY GO, available at https://www.cooleygo.com/ (last visisted Mar. 21, 2019).

Modern intellectual property valuation in the academic technology transfer setting 191 ments. The third example is the University of Michigan’s equity-only licensing framework for software startups. Standardized startup terms 1. Several TTOs now offer and aggressively market standard startup licenses. These licenses represent a set of terms that a TTO is offering to a prospective startup licensee. TTOs often limit the applicability of these licenses to particular types of technologies or to startups in which faculty are actively involved. Nonetheless, these licenses implicitly reflect the fact that the TTO is adopting a predetermined set of economic terms regardless of the specific value of the IP asset being licensed. In other words, rather than applying one of the historical valuation approaches to try to optimize the economic return from a particular IP asset, the TTO is deciding to streamline the licensing process in order to maximize the number of IP licenses in its portfolio. The University of North Carolina launched the “Carolina Express License Agreement” with much fanfare in 2010.71 In 2013, the TechTransfer Central publication highlighted five TTOs using standardized startup licenses. More recently, Yale has joined the ranks of those offering a startup license with pre-negotiated economic terms.72 Yale’s TTO website describes the “Yale Startup License,” which it markets as a “pre-negotiated Startup License” that is “intended to greatly speed up and streamline the licensing process.”73 The license is available for any Yale technology other than a “drug or therapeutic.” Eligible licensees include startups founded by a Yale employee who invented the IP to be licensed and that are participating in Yale’s startup training programs.74 2. Pre-negotiated IP licensing terms for industry sponsored research The University of Minnesota has pioneered various options for pre-negotiated licenses to IP created during the course of industry sponsored research.75 Historically, universities have offered a time-limited option to negotiate a commercial license to IP once it is created under an industry sponsored research program. The theory behind this historical approach is that the parties must first understand the resulting IP in order to ascertain its value before negotiating commercial license terms.76 Minnesota’s approach, however, offers industry sponsors the opportunity to accept a commercial license to IP before the IP is created. Under one option, the sponsor can pay the greater of 10 percent of the sponsored research agreement or $15,000 for an exclusive license to any IP arising from the sponsored program. The license would include

71 Joseph M. DeSimone & Lesa Mitchell, Facilitating the Commercialization of University Innovation: The Carolina Express License Agreement, kauffMan foundatIon (Apr. 2010), available at https://www.kauffman.org/what-we-do/research/2012/07/facilitating-the-commercialization-of -university-innovation (last visited Oct. 17, 2019). 72 Yale Startup License, supra note 59. 73 Id. 74 Id. 75 Sponsoring Research: Minnesota Innovation Partnerships (MN-IP), u. MInn. off. teCh. CoMMerCIalIzatIon, available at https://research.umn.edu/units/techcomm/sponsoring-research-mn-ip (last visited Mar. 26, 2019). 76 There are also tax reasons why universities may have limitations on granting pre-negotiated intellectual property licenses.

192 Research handbook on intellectual property and technology transfer a 1 percent royalty on net sales when annual sales exceed $20 million. Minnesota markets this model, called “MN-IP: Minnesota Innovation Partnerships,” as follows: MN-IP improves access to University-developed technology while reducing the risk and cost associated with exclusively licensing intellectual property and sponsoring lab and clinical research. The MN-IP program has been regularly acknowledged in the industry as a leader in developing a streamlined contracting model that is flexible and provides options to suit each company’s needs.77

Minnesota further promotes its program with testimonials from industry partners talking about the ease of working with the university when negotiations are not required. In other words, Minnesota, in the case of industry sponsored research, is foregoing the specific valuation of each IP asset. It is willing to accept suboptimal economic returns on a particular deal in order to generate a larger portfolio of industry relationships and IP resulting from those relationships. The risk, of course, is that it might be agreeing to license a lucrative piece of IP for terms that are significantly below the market value. 3. An equity-only framework for software startups The University of Michigan’s TTO has adopted a startup licensing framework using a portfolio approach to valuation. This framework involves granting equity-only licenses to startups licensing software embodied inventions through the TTO. This framework is aimed at creating a more investor and founder friendly model for software startup licenses. The goal is for this framework to assist in building the local startup ecosystem, attract entrepreneurial faculty and graduate students to the university, and decrease the incentive for a software startup based on university research to work around the TTO licensing process. This framework involves three categories of equity-only licenses based on the general value of the code being licensed. The categories start with a 1 percent equity model where some concept of the software or wire-frames were developed through university research, but no significant code was developed with university resources. Where significant software development occurred at the university, then higher equity amounts are included in the license. The university requests common stock that it receives based on the post-financing cap table after the first equity round of financing (i.e., the seed round). In fiscal year 2017, the University of Michigan licensed 21 startups that were launched based on this licensed IP. Nine of those startup licenses were equity only software licenses. The vast majority of those nine licensed startups would have been unlikely to enter into a license if the TTO was insisting on a sales-based royalty model. Accordingly, by adopting a portfolio-based approach to valuation and deal-making, the university was able to vastly expand its portfolio of licensees. Under this approach, the university’s TTO has experienced a significant increase in goodwill, especially with the local seed investment community. While significant non-economic rationales for this approach drove its adoption, the approach also has an economic rationale. The valuations of those nine startups licensed under the equity only framework already total in the high hundreds of millions of dollars. While the ultimate outcomes are obviously uncertain, the university believes it now has a small equity position in a portfolio of high-potential startups that it would not have had but for this approach.

77

Sponsoring Research: Minnesota Innovation Partnerships (MN-IP), supra note 75.

Modern intellectual property valuation in the academic technology transfer setting 193 C.

Situations Still Requiring the Specific Valuation of an IP Asset Being Licensed

While TTOs will likely increasingly adopt portfolio-based approaches to valuation and deal-making in order to meet their expanded missions, there are situations where the specific valuation of an IP asset being licensed continues to make sense. One clear example is in the context of therapeutics licenses. Because of the significant research and development costs associated with bringing a therapeutic technology to market, patent protection is essential. The ability to exclude competitors makes a royalty-bearing license relatively easy to tolerate for a licensee. The exclusive power of a patent, especially one covering a pharmaceutical composition of matter, provides benefit that far exceeds a market royalty payment. A significant majority of TTO revenues derive from royalties, and a significant majority of those royalty payments derive from life sciences technologies.78 This is unlikely to change. Accordingly, for the foreseeable future, TTOs will continue to employ the historical valuation methodologies above to seek to optimize their economic terms, especially sales-based royalties and milestone payments when licensing patent rights for pharmaceutical assets.

VII.

CONCLUSION: RE-THINKING THE “VALUE” OF IP LICENSING

The above discussion focuses on “valuing” IP assets in terms of thinking about the direct monetary value of IP being licensed. As discussed in this Chapter, however, the mission of TTOs has expanded dramatically in the decades following the Bayh-Dole Act. Most TTOs no longer prioritize revenue generation over other goals, such as economic development, faculty recruitment, student and faculty education, ecosystem building, and university and faculty branding. Valuing and generating the value from particular IP assets remains important, and will likely remain important for certain situations and certain asset classes, like therapeutics. As TTOs think about capturing the value from the IP they manage, however, they will increasingly think about an expanded definition of value that aligns with their expanded goals. TTOs and university administrators will likely spend more time thinking about the following questions: ● How much is it worth to the university if superstar faculty and grad students join the university due to the university’s TTO practices? ● How much is it worth to the university if the local startup ecosystem speaks highly of the university’s TTO? ● How many jobs have been created in the local economy due to the TTO’s activities? ● How has the engagement of students and faculty with the university’s TTO inspired their future career activities? ● What is the brand recognition of the university from the IP arising from its research enterprise? As TTOs and university administrators think about and track the answers to these questions, TTOs will increasingly align their deal-making behavior with the tremendous untapped non-monetary value of their IP portfolio. Accordingly, it is likely that TTOs will continue to

78

See Reslinski & Wu, supra note 13.

194 Research handbook on intellectual property and technology transfer adopt more portfolio-based valuation approaches to their deal-making activities. Instead of seeking to maximize economic return on each individual deal, TTOs will adopt practices that optimize the collective impact of their entire portfolio to serve their expanded missions and to achieve the full potential of their cutting-edge academic research.

PART II INTELLECTUAL PROPERTY AND TECHNOLOGY TRANFSER IN THE INNOVATIVE UNIVERSITY

9.

The innovation arms race on academic campuses Todd Sherer and Liza Vertinsky

I.

INTRODUCTION

While we retain our core mission of educating the next generation and cultivating new forms of knowledge, universities must also embrace our ever-expanding role in driving innovation and catalyzing economic development.1 As tuition continues to rise, colleges and universities now increasingly market themselves to students, legislators, and donors—as storehouses of innovation and engines of the “knowledge economy.”2

Innovation has become the buzzword in university mission statements and strategic plans as universities seek to transform themselves into “innovation ecosystems” equipped to perform an ever-widening variety of entrepreneurial, innovation, and commercialization functions.3 These efforts have been fueled by federal and state government pressure on universities to incorporate innovation and entrepreneurship into their missions in ways that can produce tangible economic value.4 They have been reinforced by recent federal government initiatives designed to improve the “return on investment” from federally-funded research through acceleration of technology transfer.5 The resulting proliferation of university innovation programs has been further fueled by student demands for a campus that will train them as future 1 Farnam Jahanian, 4 Ways Universities are Driving Innovation, World Econ. Forum Annual Meeting (Jan. 17, 2018), available at https://www.weforum.org/agenda/2018/01/4-ways-universities-are -driving-innovation/ (last visited Oct. 17, 2019). 2 John P. Leary, Enough with All the Innovation, Chron. hIgher eduC., Nov. 11, 2018. 3 See, e.g., Recommendations to Facilitate University-Based Technology Commercialization, Letter to Secretary Locke, nat’l advIsory CounCIl on InnovatIon & entrepreneurshIp (Apr. 19, 2011), available at http://www.innovationamerica.us/images/stories/2011/NACIE_Letter University_ Commercialization-20110617084146-20110617215655.pdf (“Already engaged in many activities that promote innovation, entrepreneurship and the commercialization of research results, we are committed to working even more closely with industry, private foundations, venture capitalists and local, state and federal governments to enhance our efforts … many of our universities are actively building campus-wide innovation ecosystems and expanding them into regional and national networks.”); Natasha Singer, Universities Race to Nurture Start-Up Founders of the Future, N.Y. TIMES, Dec. 28, 2015, available at https://www.nytimes.com/2015/12/29/technology/universities-race-to-nurture-start-up-founders-of-thefuture.html (last visited Dec. 11, 2019); Leary, supra note 2 (discussing and critiquing trend in colleges presenting themselves “as storehouses of innovation and engines of the knowledge economy.”). 4 U.S. Dep’t of Com., The Innovative and Entrepreneurial University Report, Oct. 2013, available at https://www.eda.gov/pdf/the_innovative_and_entrepreneurial_university_report.pdf (last visited Oct. 17, 2019) (“From the i6 Proof of Concept Center at the University of Akron to the University of Wyoming’s Technology Business Center, America’s higher education institutions are embracing the importance of innovation, commercialization, entrepreneurship, and the creation of economic value for their communities.”). 5 See, e.g., Michael Horn, How Universities Should Manage Innovation, forbes, Apr. 3, 2018, available at https://www.forbes.com/sites/michaelhorn/2018/04/03/how-universities-should-manage

196

The innovation arms race on academic campuses 197 innovators.6 Universities now compete with each other for students and for funding based on innovation rankings alongside more traditional metrics,7 and innovation hubs have replaced state-of-the-art gymnasiums in university promotional materials.8 While innovation has been embraced as an integral part of the university mission at a growing number of US research universities, the term is an amorphous one that allows for many different meanings and approaches. The rush to become an innovative university, coupled with fluid and changing understandings of what this means, has resulted in a proliferation of innovation initiatives both within and across university campuses.9 These initiatives encompass, among other things, student and faculty entrepreneurship training and programing, efforts to encourage start-up companies that draw from university technologies and human capital, expanding forms of industry-university partnerships, and university engagement with regional and local economic development projects.10 The diversity of approaches is in part a reflection of variations in the nature and distribution of resources across different universities, including differences in public and private research funding, endowments, tuition dollars, and proximity to potential funders and industry partners. It is also a reflection of divergent views both within and across university campuses about the roles and functions of universities and the ways in which they should (and should not) be evolving in response to the changing economic, technological, and political pressures facing higher education. Although their approaches to innovation may vary, however, many if not all universities—major research universities in particular—are being pushed to respond in creative ways to the pressing innovation and competitiveness agendas of federal and state policymakers and the intensifying competition for students and industry partners. As the pressure on universities to demonstrate the conversion of intellectual value into economic value increases, the focus often turns to the technology transfer offices (“TTOs”) that are tasked with this conversion.11 University TTOs are being challenged to adapt to the

-innovation/#6cb2764e7c4b (last visited Oct. 17, 2019) (“Innovation has reached buzzword status inside colleges and universities.”). 6 See, e.g., Singer, supra note 3. 7 See, e.g., David Ewalt, Reuters Top 100: The World’s Most Innovative Universities, 2017, reuters, Sept. 27, 2017, available at at https://www.reuters.com/article/us-amers-reuters-ranking -innovative -univ/ reuters -top -100 -the -worlds -most -innovative -universities -2017 -idUSKCN1C209R ?hootPostID=16d7acc652ddf97912f26adda1635ff0 (last visited Oct. 17, 2019); Ross DeVol, et al, Concept to Commercialization: The Best Universities for Technology Transfer, MIlken InstItute report, Apr. 2017, available at https://assets1c.milkeninstitute.org/assets/Publication/ResearchReport/ PDF/Concept2Commercialization-MR19-WEB.pdf (last visited Oct. 17, 2019); University Innovation Fellows Ranking (by students) at http://universityinnovation.org/wiki/Category:Universities (last visited Oct. 17, 2019). 8 See, e.g., Alexandra Lange, The Innovation Campus: Building Better Ideas, n.y. tIMes, Aug. 4, 2016, available at https://www.nytimes.com/2016/08/07/education/edlife/innovation-campus -entrepreneurship-engineering-arts.html (last visited Oct. 17, 2019). 9 For a discussion of the range of different approaches to innovation on campus see, e.g., U.S. Dep’t of Com., supra note 4 (exploring variety of programs at universities designed to nurture innovation (construed broadly) and entrepreneurship among students, faculty and communities, including educational goals, fostering of startups, contributing to economic growth, among other things). 10 See, e.g., U.S. Dep’t of Com., supra note 4. 11 See, e.g., Walter D. Valdivia, University Start-ups: Critical for Improving Technology Transfer, Ctr. for teCh. InnovatIon brookIngs rep., Nov. 2013 (defining technology transfer as “the complex work done at the interface of research and productive organizations”) .

198 Research handbook on intellectual property and technology transfer ever-broadening innovation agendas of their institutions.12 In some cases TTOs have been asked to take on this broader domain of activities as part of their operations; in other cases, the initiatives have been located in separate entities within the university, and sometimes the initiatives span multiple locations within the university.13 Regardless of where the initiatives are located, they inevitably impact the TTOs that sit at the intersection of university research and the productive deployment of this research to fuel innovation. Despite this involvement, be it direct or indirect, there has been little discussion to date about the impact of the growth in innovation initiatives on university TTOs and the technology transfer functions they perform. This Chapter examines the evolving role of university TTOs amidst the proliferation of innovation and entrepreneurship initiatives in US research universities.14 It considers the implications of this arms race of innovation programs, along with the intensifying government focus on the return on investment from federal funding, for TTOs and their traditional technology transfer functions.15

II.

EVOLVING ROLE OF UNIVERSITY TECHNOLOGY TRANSFER

While universities have been engaging in technology transfer for decades, the nature and scope of academic technology transfer was transformed by a change in the law, introduced by the Bayh-Dole Act, that allowed universities to elect title to and patent their federally-funded inventions. The Bayh-Dole Act was signed into law in 1980 in the midst of debates about US competitiveness and concerns about an alleged failure of the private sector to commercialize academic research.16 Then, as now, Congress was eager to see improvements in US economic competitiveness resulting from public investments in universities, and universities were eager to take control of their faculty inventions and find new ways of demonstrating their value. The Bayh-Dole Act was designed with the goal of facilitating and accelerating the movement of early-stage inventions from university campuses and other sites of federally-funded research into the private sector for commercialization and development.17 While the mechanics of the Act focused on patenting and ownership of inventions created using federal funding, allowing

12

See, e.g., Technology Transfer Evolution: Driving Economic Prosperity, APLU, 2017. See, e.g., Brady Huggett, Reinventing Technology Transfer, nature bIoteChnology (2014) (discussing change in focus of TTOs and examples of changes in TTO domains, organization and even names among US research universities). 14 See, e.g., id. (discussing how TTOs are adapting to changing economic climate and university strategies, including shift of focus to startup companies and outreach to the private sector). 15 See, e.g., U.S. Dep’t of Com., supra note 4 (“The recent burst of entrepreneurship on campuses has greatly expanded the role of the TTOs and TLOs. Instead of merely focusing on the commercialization of individual technologies, these offices now act as a central point where students, faculty, alumni, entrepreneurs, investors, and industry can connect with each other. These offices are now focused on identifying and supporting entrepreneurship on campus, helping startups find the best opportunities and building successful business models, changing the culture of their universities, and creating companies that will be based in the communities around the university.”) 16 See, e.g., David Mowery & Bhaven Sampat, The Bayh-Dole Act of 1980 and University-Industry Technology Transfer: A Model for Other OECD Governments?, 30 J. of teCh. transfer 115 (2005). 17 Bayh-Dole Act of 1980 (Public Law 96-517, The Patent and Trademark Act Amendments of 1980). 13

The innovation arms race on academic campuses 199 universities to elect title to and license inventions developed by their faculty with the use of federal funding, the Act reflected broader ideas and expectations about the role technology transfer could play in promoting US technological innovation and competitiveness.18 Universities had been involved in technology transfer long before the Bayh-Dole Act came into effect, with some universities patenting faculty inventions as early as the 1920s and a growth in patenting and licensing activities and formation of TTOs occurring in the late 1960s and early 1970s.19 But the Bayh-Dole Act, along with other changes in intellectual property (“IP”) laws that expanded what could be patented, contributed to substantial changes in how US universities thought about, managed, and shared the discoveries made on their campuses.20 As part of this evolution, the change in law contributed to the establishment of a profession of technology transfer managers and the rapid increase in the number of TTOs in US research universities.21 The annual growth in formation of new TTOs at research universities was exponential in the decade following the enactment of the Bayh-Dole Act, growing from 25 in 1980 to 200 in 1990.22 Today most, if not all, major research universities in the US have TTOs, and many of these offices have increased in size, scope, and sophistication over time.23 Following the passage of the Bayh-Dole Act and the first wave of new TTOs, universities became viewed as a “source” of innovation to be cultivated through technology transfer practices.24 TTOs initially focused on the patenting and licensing of university discoveries, viewed as the source of innovation, to industry for commercial development. While it took some time, high hopes were eventually placed by university administrators and government policymakers alike on the ability of university technology transfer to generate sizeable economic returns on the commercialization of publicly funded research and development. Expectations about the magnitude of financial returns from technology transfer were buoyed by a few early blockbuster licensing deals, such as Stanford University’s licensing of patents covering recombinant DNA technology in the 1980s. This licensing deal generated millions of dollars in revenue for Stanford and billions of dollars in product sales.25

18

See, e.g., Wendy Schacht, The Bayh-Dole Act: Selected Issues in Patent Policy and the Commercialization of Technology, Cong. res. serv. reps. 2 (2006) (explaining the passage of the Bayh-Dole Act result of Congressional interest in facilitating US technological innovation, expectations that providing uniform national policy covering ownership of federally-funded inventions would reduce barriers and increase private sector incentives to develop and commercialize these inventions). 19 See, e.g., Mowery & Sampat, supra note 16. 20 Id. 21 Rosa Grimaldi, et al., 30 Years after Bayh-Dole: Reassessing Academic Entrepreneurship, res. pol’y (2011), available at http://desimone-group.chem.unc.edu/wp-content/uploads/2013/05/30_years _after_bayh-dole__reassessing_academic_entrepreneurship.pdf. 22 See, e.g., Valdivia, supra note 11; David Mower, et al., The Growth of Patenting and Licensing by U.S. Universities: AN Assessment of the Effects of the Bayh-Dole Act of 1980, available at https://pdfs .semanticscholar.org/c9f3/732e5793617f7cbdfcddd9ce438b805d031f.pdf (last visited Oct. 17, 2019). 23 See, e.g., Valdivia, supra note 11 (explaining that 206 US research universities are considered by the 2010 Carnegie Classification of Higher Education to have high or very high research activity, and all of them have TTOs). 24 See, e.g., Gilles Durufle, Thomas Hellmann, & Karen Wilson, Catalyzing Entrepreneurship in and around Universities, saId bus. sCh. res. paper, May 2018. 25 See, e.g., Maryann Feldman & Paige Clayton, “The American Experience in University Technology Transfer” in University Technology Transfer: The Globalization of Academic Innovation (S. Breznitz & H. Etzkowitz eds., 2015) (explaining passage of the Bayh-Dole Act coincided with the pat-

200 Research handbook on intellectual property and technology transfer Yet blockbuster deals like this one have proven to be the exception, not the norm, and licensing revenues from university technology transfer have remained modest for most institutions.26 Indeed, most TTOs are not net revenue generators. Some figures based on 2016 data suggest that as many as 73% of the TTOs are losing money and 16% are just breaking even,27 while other studies suggest that the percentage of TTOs that do not break even may be even larger.28 Out of those TTOs that are net revenue generators, the distribution of gross licensing income has proven to be highly skewed, with a small percentage of universities taking the majority of the licensing income in any one year and the composition of this group of top earners varying little over time.29 Now, almost forty years after the passage of the Bayh-Dole Act, the domain of technology transfer has expanded well beyond traditional functions of patenting and licensing university discoveries and the creation of start-up companies based on university technologies to encompass a myriad of different economic development objectives.30 As TTOs have matured they have expanded their domain to encompass additional activities such as start-up business formation and the encouragement of university-industry partnerships and collaborations.31 While the first wave of change in technology transfer focused attention on universities as sources of innovation, this second wave of change focuses on universities as “hubs” of innovation situated within an innovation ecosystem.32 Policymakers have encouraged universities to explore ways of fostering university-based entrepreneurship, construed broadly to include not only the commercialization of university-based technologies, but also the nurturing of entrepreneurial

enting and licensing of recombinant DNA technology by Stanford in what resulted in over $255 million in licensing revenue for Stanford and over $35 billion in product sales). 26 See, e.g., C. Simone Fishburn, Tables Turning for TTOs, sCIbx, Jan. 23, 2014, available at https:// www.nature.com/scibx/journal/v7/n3/full/scibx.2014.77.html (last visited Oct. 17, 2019); see also W. Valdivia, Technology Transfer: Highly Dependent on University Resources, brookIngs blog (Mar. 4, 2014), available at https://www.brookings.edu/blog/techtank/2014/03/04/technology-transfer-highly -dependent-on-university-resources/ (last visited Oct. 17, 2019). 27 See, e.g., Dave Merrill, et al., Billions at Stake in University Patent Fights, blooMberg, May 24, 2016, available at https://www.bloomberg.com/graphics/2016-university-patents/ (last visited Oct. 17, 2019). 28 See, e.g., Valdivia, supra note 11 (explaining that 84% of TTOs did not generate enough revenue to cover their operating costs based on 2012 data, an improvement over the average of 87 in the preceding 20 years). 29 See, e.g., Valdivia, supra note 11 (examining 10-year trend in licensing revenues, showing that in 2012, as a representative year, 8 universities took 50% of the licensing income and 16 universities (the top 10%) took 75% of the licensing income, and showing that only 37 universities have been able to reach the top 20 of licensing revenue any given year over a 10-year period). 30 See, e.g., U.S. Dep’t of Com., supra note 4. 31 See, e.g., Technology Transfer Evolution Working Group of APLU’s Commission on Innovation, Competitiveness and Economic Prosperity, Technology Transfer Evolution: Driving Economic Prosperity, APLU, Nov. 2017, available at http://www.aplu.org/library/technology-transfer-evolution -driving-economic-prosperity/file (last visited Oct. 17, 2019) [hereinafter APLU]; see also Brady Hugget, Reinventing Technology Transfer, bIoentrepreneur, Dec. 5, 2014, available at https://www .nature.com/bioent/2014/141201/full/bioe.2014.12.html (last visited Oct. 17, 2019). 32 See, e.g., Durufle, et al., supra note 24 (“In addition to the two traditional core pillars or teaching and research, a third role has emerged that may be broadly characterized as serving as a hub of innovation and entrepreneurship.”).

The innovation arms race on academic campuses 201 and start-up activity by students, alumni, faculty, and university partners within and around a university-supported start-up ecosystem.33 The pressure to expand and diversify the business model of technology transfer increased dramatically in the aftermath of the 2008 global financial crisis and has continued as universities struggle with reduced government funding and greater competition for tuition dollars.34 This pressure to engage more broadly in economic development comes not only from university administrators anxious about revenue flows, but also from local, state and federal governments anxious to promote entrepreneurship and to demonstrate returns from university funding in the form of jobs and new businesses. The widely discussed National Academies report, “Research Universities and the Future of America,” captures this pressure to improve university impact on economic growth and competitiveness, emphasizing the importance of supporting and expanding relationships between universities and industry to accelerate technology transfer and the need for improved efficiency and productivity to generate greater return on investment of public funds.35 A report by the US Department of Congress entitled “The Innovative and Entrepreneurial University: Higher Education, Innovation and Entrepreneurship in Focus” explores how universities can work with industry and government to advance university based innovation in support of economic value creation.36 Broader university-industry partnerships are also emphasized at a more granular level as survival strategies in reports such as “Innovation and the Independent College,” published by the Council for Independent Colleges.37 US research universities have responded to these pressures by expanding their focus beyond traditional functions of knowledge creation, education, and technology transfer to take on new innovation and entrepreneurship initiatives.38 TTOs, located as they are at the intersection of research and the productive deployment of this research, are almost inevitably involved in these initiatives. Sometimes they are tasked with managing the innovation initiatives, either alone or in partnership with other parts of the university, often having to squeeze new projects into stagnant budgets. At other times TTOs are indirectly involved in cross-campus initiatives through the impact of innovation programs on technology transfer functions. The expansion of technology transfer beyond core licensing and technology management has been reflected in the evolution of the association formed for technology managers. The board of directors of AUTM noted the dramatic change in technology transfer since its founding in 1974, with an expansion beyond patent licensing and technology management to include licensing of other types of IP, a variety of industry collaborations and partnerships,

33

See, e.g., U.S. Dep’t of Com., supra note 4 (explaining a government study celebrating the ways in which universities are influencing innovation); Durufle, et al., supra note 24 (providing a descriptive model of university start-up ecosystems). 34 Dipanjan Nag, The Changing Face of University Technology Transfer, Ip watChdog, Oct. 9, 2017, available at http://www.ipwatchdog.com/2017/10/09/the-changing-face-of-university-technology -transfer/id=88853/ (last visited Oct. 17, 2019). 35 See, e.g., National Research Council, Research Universities and the Future of America; Ten Breakthrough Actions Vital to Our Nation’s Prosperity and Security (2012) (stating recommendation to strengthen the business role in the research partnership, increase the return on investment of public money). 36 U.S. Dep’t of Com., supra note 4. 37 See, e.g., U.S. Dep’t of Com., supra note 4. 38 See, e.g., U.S. Dep’t of Com., supra note 4.

202 Research handbook on intellectual property and technology transfer increased activity in startup business formation and early-stage financial support, and entrepreneurial education and programming.39 This expansion in the domain of TTOs prompted the Board of Directors of AUTM to propose a name change, shifting from the Association of University Technology Managers (“AUTM”) to the Association for Research and Innovation Advancement (“ARIA”).40 While the change in name was ultimately rejected, an embrace of the broader scope of activities falling within the realm of technology transfer was not, and AUTM’s mission has expanded to reflect the breadth of contemporary technology transfer.41 While AUTM along with higher education associations such as the Association of American Universities (“AAU”) and the Association of Public and Land-Grant Universities (“APLU”) have “urged university leaders to embrace the role of university technology transfer in promoting innovation and economic prosperity,”42 the implications of evolving technology transfer practices have been understudied. Although the AAU and APLU have reaffirmed “that the primary goal of university technology transfer operations is to advance the public interest,”43 determinations of what that public interest consists of and how to best promote it remain open to debate. In the meantime, attention from both within the university and outside of it continues to focus on easy-to-use economic metrics such as licensing revenues, industry partnerships, and number of startup companies created. Innovation programs that promise to contribute in some way to these metrics continue to proliferate, impacting TTOs in ways that we have yet to fully understand.

III.

THE ARMS RACE OF INNOVATION PROGRAMS

University leaders are increasingly responding to the needs of the innovation economy … by including innovation, entrepreneurship, and “economic engagement” programming in their strategic planning processes. As part of this response, university TTOs are evolving, and must continue to evolve, toward participation in a broader scope of efforts …44

The expansion in scope and range of activities undertaken by university TTOs has been driven by and is reflective of broader changes taking place at universities.45 Attention to innovation is becoming institutionalized, with a growing number of senior university positions in

39

See Letter to AUTM Members, AUTM, Sept. 29, 2017, available at https://www.autm.net/ AUTMMain/media/Elections/Documents/Letter_to_Members_29SEP2017_Website_3.pdf (last visited Oct. 17, 2019). 40 See id. 41 See, e.g., AUTM Strategic Plan 2018–2020, available at https://www.autm.net/autm-info/about -autm/strategic-plan/ (last visited Oct. 17, 2019). 42 See Higher Ed Associations RFI Response—Federal Technology Transfer Authorities and Processes (Docket No. 18022019-819-01). 43 See id. 44 See, e.g., APLU, supra note 31 (“In evolving toward broader participation in university economic engagement, TTOs will develop deeper relationships with industry and other community partners; broaden their reach to areas such as education, technology development and entrepreneurship; and integrate more closely with other supportive administrative functions such as industry contracting.”). 45 See, e.g., APLU, supra note 31 (examining “how university technology transfer is evolving in the context of broader economic engagement strategies”).

The innovation arms race on academic campuses 203 “innovation” and the inclusion of innovation as part of the university mission.46 Few annual reports neglect the topic of innovation, and many lead with a discussion of the institution’s innovation success stories. Universities are being tasked by their stakeholders to show what they are “doing programmatically and strategically to nurture innovation, commercialization, and entrepreneurship among students, faculty, alumni, and within their communities … [and to] improve their ability to develop products and services with market relevance and economic value.”47 Innovation and entrepreneurship centers, ventures and programs have sprouted up on many university campuses as universities, pushed by a variety of stakeholders, compete aggressively for government funding, industry funding, students, and prestige in an arms race of innovation programs. Unfortunately, the focus on innovation as part of the university mission and as a key part of its strategic plan does not always translate into a clear set of goals and objectives. As noted earlier, “innovation” is an amorphous term that can and does encompass very different approaches and objectives. The various approaches can be organized into rough categories based on their core objectives. A joint letter from a group of major US research universities to the Department of Commerce entitled “Recommendations to Facilitate University-Based Technology Commercialization” sets out four main categories of objectives for innovation programs: (1) promoting student innovation and entrepreneurship, (2) actively supporting the university technology transfer function, (3) facilitating university-industry collaboration, and (4) engaging with regional and local economic development efforts.48 These categories are useful as an organizing framework for examining current trends in innovation programs across and within universities, and we adopt them here, although recognizing that the categories are neither exhaustive nor separate and distinct and that many universities are pursuing more than one such program. Within the first category of student innovation and entrepreneurship opportunities, at least 450 US colleges and universities now have entrepreneurship programs,49 and new degree programs focusing on varying aspects of innovation and entrepreneurship continue to emerge. The University of Colorado, for example, now offers a bachelor of innovation degree program in innovation.50 A report by the Kauffman Foundation showed a growth in the number of entrepreneurship courses on campus, rising from only about 250 courses in 1985 to over 400,000 students in 2013.51 New innovation centers of various scopes and sizes continue to

46 See, e.g., Horn, supra note 5; Jeffrey Selingo, The Rise of the Chief Innovation Officer in Higher Education, Jan. 2018, available at https://cdn2.hubspot.net/hubfs/2622320/The%20Rise%20of%20the %20CIO%20in%20HigherEd.pdf (last visited Oct. 17, 2019). 47 The Innovative and Entrepreneurial University, Executive Summary, available at https://www .eda.gov/pdf/the_innovative_and_entrepreneurial_university_report.pdf (last visited Oct. 17, 2019). 48 See Letter to Secretary Locke, supra note 3. 49 U.S. Dep’t of Com., supra note 4, at 15 (“universities are increasingly driving or involved in each of these factors: developing fertile innovation ecosystems, creating an entrepreneurial culture, and providing sustained financing for new ventures.”). 50 See, e.g., Julie Kliegman, How one college built an innovation degree program … in innovation, week, June 10, 2016, available at http://theweek.com/articles/621644/how-college-built-innovative -degree-program-innovation (last visited Oct. 17, 2019). 51 See, e.g., Singer, supra note 3; see also David Gernon, Universities, Hoping to Sway Millennials, are now Opening Innovation Hubs for Undergraduates, CNBC, July 23, 2017, available at https://www .cnbc.com/2017/07/20/universities-hoping-to-sway-millennials-are-now-opening-innovation-hubs-for -undergraduates.html (last visited Oct. 17, 2019).

204 Research handbook on intellectual property and technology transfer open on university campuses, some targeted toward students and some toward a broader range of university constituents and activities.52 By some estimates, in 2017 more than 200 colleges and universities had launched centers dedicated to innovation or entrepreneurship, as indicated by their participation in the Global Consortium of Entrepreneurship Centers.53 The second category includes programs by universities that are experimenting with ways of moving new discoveries through the “valley of death.”54 Policymakers have focused in particular on the role of universities in encouraging high value start-up formation in their scrutiny of technology transfer.55 Academic entrepreneurship, including the formation of new business ventures by faculty, students, or by affiliated parties using university technology, is looked to as an engine of economic growth.56 Policymakers are encouraging universities to develop their own entrepreneurial infrastructure, including resources and even financial support for start-ups based on university technology.57 Emerging trends in technology transfer include initiatives designed to streamline early commercialization efforts through incubators or “one-stop-shops” providing legal, business, and sometimes even financial support to faculty and student inventors.58 There are more than 32 university-related proof-of-concept centers designed to push university technologies through early development stages.59 Some universities have established seed funding programs to support faculty and student start-up activity.60 Many start-up 52

See, e.g., Brad Lukanic, The Next Hot Trend on Campuses: Creating Innovation, fastCoMpany, Feb. 2, 2015, available at https://www.fastcompany.com/3042566/the-next-hot-trend-on-campus -creating-innovation (last visited Oct. 17, 2019); Nish Acharya, You’ll Never Guess Where the New Centers of Innovation Are, forbes, Nov, 12, 2013, available at https://www.forbes.com/sites/ nishacharya/2013/11/12/youll-never-guess-where-the-new-centers-of-innovation/#6050a8386025 (last visited Oct. 17, 2019). 53 World Economic Forum Report. 54 See, e.g., University Gap Funding: Mind the Gap, available at http://www.gapfunding.org/ mindthegap/ (last visited Oct. 17, 2019); see also U.S. Dep’t of Com., Innovative & Entrepreneurial U. Rep., oCt. 2013, available at https://www.eda.gov/pdf/the_innovative_and_entrepreneurial_university _report.pdf (last visited Oct. 17, 2019) [hereinafter Innovative & Entrepreneurial U. Rep.). 55 See, e.g., Bharat Rao & Bala Mulloth, The Role of Universities in Encouraging Growth of Technology-Based New Ventures, 14 Int’l J. of InnovatIon and teCh. MgMt 1750014 (2016), available at http://batten.virginia.edu/sites/default/files/research/attachments/Mulloth_Role_of_Universities_in _Encouraging_Growth.pdf (last visited Oct. 17, 2019) (“the growing interest among universities in pursuing commercial applications of research, including new venture creation, is a clear trend of an increasing number of “entrepreneurial universities” playing an enhanced role in technological innovation.” “However, despite high expectations and significant attention given to the role of universities in encouraging the growth of technology-based new ventures, the results in most contexts are disappointing.”). 56 See, e.g., Christopher Hayter, et al., Conceptualizing Academic Entrepreneurship Ecosystem: A Review, Analysis and Extension of the Literature, available at file:///Users/lizavertinsky/Downloads/ SSRN_ID3137406_code924564.pdf. 57 See, e.g., National Research Council, Rising to the Challenge: U.S. Innovation Policy for the Global Economy, 2012, available at https://www.nap.edu/catalog/13386/rising-to-the-challenge-us -innovation-policy-for-the-global (last visited Oct. 17, 2019). 58 Innovative & Entrepreneurial U. Rep., supra note 54., at 28. 59 See, e.g., S. Bradley, et al., Proof of Concept Centers in the United States: An Exploratory Look, 38 J. of teCh. transfer 349 (2013) (discussing role of proof-of-concept centers as university technology infrastructure; proof-of-concept centers often include seed funding, incubator space, business and advisory support). 60 See, e.g., Innovosource, Mind the Gap, available at https://www.universitygapfunding.com/ (last visited Jan. 15, 2019); see also examples of programs provided in Arundeep Pradhan, The Evolution of Technology Transfer, Nov. 12, 2018 (explaining that start-up programs include PCI Ventures at

The innovation arms race on academic campuses 205 initiatives involve partnering with industry and/or with local and state government partners, blending into the third and fourth categories below. A third category of programs, one that overlaps with the others, focuses on expanding university-industry partnerships.61 Efforts have been made by universities, often with government encouragement and industry support, to expand and facilitate collaborative efforts with industry that involve a pooling of risk, cost, resources and expertise to tackle areas of shared university and industry interest.62 Efforts to encourage university-industry partnerships are not new, but the policy emphasis on expanding partnerships and budgetary pressure to seek industry support has increased in intensity in recent years.63 Universities are experimenting with ways of making university-industry engagement easier for private sector partners through strategies such as “front door policies,” web portals and streamlined licensing policies, as well as through strategies for making university facilities and infrastructure more widely available to private partners.64 Many research universities have created translational research centers and programs with the goal of promoting the advance of research into commercial applications by industry, a trend encouraged by federal government funding strategies targeting translational research.65 Research agreements between the private sector and universities have expanded in both number and, perhaps of more concern, in scope.66 Expansive research alliances such as the California Research Alliance, a framework agreement between BASF and ten California universities, have provided much needed inflows of resources to support research, while at the same time raising questions about the influence of the corporate sector on university research agendas.67 Some universities have gone as far as launching innovation campuses focused around the cohabitation of industry, academics, and students.68 The fourth category includes initiatives designed to engage with regional and local economic development programs. Universities are exploring collaborative research models such as “research parks, university corridors, start-up accelerators, shared laboratory space, incuba-

the University of Pennsylvania, the Springboard program at Oregon Health & Science University, and Health Science Entrepreneurs at Northeastern University, all involving provision of support for basic start-up activities such as legal formation and funding). 61 See, e.g., Kenneth R. Lutchen, Why Companies and Universities Should Forge Long-Term Collaborations, harv. bus. revIew, Jan. 24, 2018, available at https://hbr.org/2018/01/why-companies -and-universities-should-forge-long-term-collaborations (last visited Oct. 17, 2019). 62 See, e.g., U.S. Dep’t of Com., supra note 4. 63 See, e.g., Engineering and Medicine, Revitalizing the University-Industry-Government Partnership: Creating New Opportunities for the 21st Century: Proceedings of a Workshop- In Brief, nat’l aCadeMIes of sCI., 2018. 64 See, e.g., U.S. Dep’t of Com., supra note 4, at 31–2. 65 See, e.g., Arundeep Pradhan, The Evolution of Technology Transfer, lInkedIn, Dec. 13, 2016 (describing programs focused on translational research); see also Feature–Translational Research, burroughs-welCoMe fund, available at https://www.bwfund.org/feature-translational-research (describing evolution of translational research) (last visited Jan. 15, 2019). 66 See, e.g., Lutchen, supra note 61. 67 See, e.g., March Reisch, A New Model for Industry-Sponsored Research on University Campuses, 96 CheM. & eng’g news 34, Aug. 27, 2018, available at https://cen.acs.org/business/investment/new -model-industry-sponsored-research/96/i34 (last visited Oct. 17, 2019). 68 See, e.g., Mitch Smith, Innovation Campuses, InsIde hIgher ed, Feb. 27, 2012, available at https:// www.insidehighered.com/news/2012/02/27/midwestern-colleges-launch-campuses-built-public-private -collaboration (last visited Oct. 17, 2019).

206 Research handbook on intellectual property and technology transfer tors and innovation and manufacturing clusters.”69 “Accelerators” are re-emerging as a popular trend, with the development of facilities located either inside or in close proximity to university campuses to support and speed up the innovation and commercialization process.70 Georgia Tech’s Flashpoint offers an example of a start-up accelerator in which start-up companies that involve Georgia Tech faculty and students are provided with entrepreneurial education and access to mentors, experts and investors in a shared start-up workspace.71 Universities are also engaging in broader economic development efforts through support for research corridors that typically have a particular geographic and technology focus, such as nanotechnology, energy or advanced materials. University research corridors include the University of California, Lawrence Berkeley’s East BayGreen Corridor, Pennsylvania State University I-99 Corridor Region, and Iowa State University Research Corridor, each with a different regional and technology focus.72 Although every university is different, many now have programs that fall into most or all of the categories described above. In some cases universities promote innovation centers that attempt to do multiple things at once.73 Apart from the first category, which is most typically although not always the domain of business schools, the innovation programs all touch on the traditional domain of technology transfer in some way and require attention and support from these offices.74 Before examining some of the concerns that arise, we first consider a second trend that is intensifying the impact of the innovation arms race and the pressures that TTOs face, and that is the federal government’s focus on the return on investment from publicly funded R&D.

IV.

RETURN ON INVESTMENT TO TECHNOLOGY TRANSFER

For America to maintain its position as the leader in global innovation, bring products to market more quickly, grow the economy, and maintain a strong national security innovation base, it is essential to optimize technology transfer and support programs to increase the return on investment (ROI) from federally-funded R&D.75

69

Innovative & Entrepreneurial U. Rep., supra note 54, at 37; see also Technology Transfer Evolution: Driving Economic Prosperity, APLU, Nov. 2017. 70 See, e.g., U.S. Dep’t of Com., supra note 4, at 33–4. 71 Innovative & Entrepreneurial U. Rep., supra note 54, at 37. 72 Id., supra note 54, at 28. 73 See, e.g., Leary, supra note 2 (“Though every institution is different, innovation and entrepreneurship centers typically do a little bit of everything: incubate student and faculty start-ups (usually for a cut in royalties and licensing fees) while also addressing intractable social problems, like climate change or poverty, with market-friendly solutions. Its ability to encompass so much, while specifying so little, is the real secret sauce of ‘innovation.’”). 74 But see, e.g., Beth McMurtie, The Hope and Hype of the Academic Innovation Center, Chron. hIgher eduC., Jan. 26, 2018, A14 (“A 2015 survey found that a growing number of colleges were marrying their academic-technology units with their teaching and learning centers in hopes of igniting fundamental reforms across campus.”). 75 See President’s Management Agenda, presIdent’s ManageMent CounCIl (2018), at 47, available at https://www.whitehouse.gov/wp-content/uploads/2018/03/Presidents-Management-Agenda.pdf (last visited Oct. 17, 2019).

The innovation arms race on academic campuses 207 While TTOs are tasked with an ever broadening set of objectives in the university race to innovate, they are often evaluated on a narrow set of traditional metrics such as revenue generated, number of technologies licensed, and perhaps also number of patent applications filed.76 These metrics have become a focal point for policymakers in a political climate that has questioned the value of publicly funded research and federal spending on universities more generally, often without full consideration of well-established metrics that capture broader measures of economic benefit from university technology transfer.77 The federal government has been ramping up cross-agency efforts to reexamine and redesign policies governing the commercialization of federally-funded technologies.78 This evaluation includes revisiting the practices and policies underpinning the federal technology transfer framework with a “refocus [of] federal technology transfer on sound business principles based on private investment.”79 In the President’s Management Agenda, one of the identified goals is to improve the transfer of federally-funded technologies from lab to market as a way of increasing the return on investment (“ROI”).80 The core of this strategy is to reduce impediments (perceived or actual) to the transfer of federally-funded technology to the private sector—including the removal of conflicts of interests protections and other regulations that might limit or slow private sector access to university-generated technology. While the agenda focuses primarily on the need to improve technology transfer from federal laboratories, the underlying ROI approach informs broader perspectives on technology transfer that impact universities as well. The debate over the performance of technology transfer often involves a comparison of the total amount spent on federal research and development to a narrow set of metrics that reflect identifiable economic returns. Critics of the current performance of technology transfer suggest that the federal government investment in R&D of approximately $147 billion—$90 billion of it going to higher education institutions to support faculty research and graduate and postdoctoral student training—should have a much higher rate of ROI than the approximately $2.5 billion in licensing fees and 648 products taken to market, based on 2010 numbers.81 76

See, e.g., Committee on Science, Engineering, and Public Polocu and Global Affairs, Trends in the Innovation Ecosystem: Can Past Successes Help Inform Future Strategies?, NAP, 2013, available at https://www.nap.edu/read/18509/chapter/5 (last visited Oct. 17, 2019). 77 See, e.g., Lori Pressman, et al., Report: The Economic Contribution of University/Nonprofit Inventions in the United States: 1996–2015, AUTM, June 2017, available at https://www.autm.net/ AUTMMain/media/Advocacy/Documents/June-2017-Update-of-I-O-Economic-Impact-Model.pdf (last visited Oct. 17, 2019) (summarizing results of a model designed to estimate the economic impact of academic licensing over time); see also Best Practices in Transforming Research Into Innovation: Creative Approaches to The Bayh-Dole Act: Hearing Before the H. Subcomm. On Techn. And Innovation, 112th Cong. (2012) (Statement of Todd Sherer), available at https://www.gpo.gov/fdsys/pkg/CHRG -112hhrg74722/pdf/CHRG-112hhrg74722.pdf (last visited Oct. 17, 2019) (discussing broad impact of university technology transfer). 78 See, e.g., NIST and White House Ramping UP Effort to Redesign Technology Transfer Policies, AIP, May 1, 2018, available at https://www.aip.org/fyi/2018/nist-and-white-house-ramping-effort -redesign-technology-transfer-policies (last visited Oct. 17, 2019). 79 See, e.g., id.; see also Return on Investment (ROI) Initiative, NIST, available at https://www.nist .gov/tpo/return-investment-roi-initiative (last visited Jan. 15, 2019). 80 See President’s Management Agenda, supra note 75, at 47. 81 See, e.g., Darrell M. West, Improving University Technology Transfer and Commercialization, Brooking Report, 20 Issues teCh. InnovatIon (Dec. 2012), available at https://www.brookings.edu/wp -content/uploads/2016/06/DarrellUniversity-Tech-Transfer.pdf (last visited Oct. 17, 2019).

208 Research handbook on intellectual property and technology transfer Rather than refocusing the calculus on broader measures of return from such investments, both university reporting and current policy analysis tend to focus on a small number of performance metrics, emphasizing number of invention disclosures, patent applications filed and patents granted, licenses, university startups, and overall revenue generated from technology transfer.82 There is a danger in focusing on metrics that leave out important university functions. The goal of technology transfer as described in the Bayh-Dole Act is to promote the dissemination of university knowledge and discoveries to the public. Many discoveries may be at early stages of development and are therefore unlikely to yield large royalties or to generate immediate and/or easy-to-measure returns. In addition, pressures on universities to demonstrate their economic value have resulted in pressures on university TTOs to demonstrate results that do not fit easily with the underlying mandate to promote the dissemination of technology for the public good.83 The focus on financial returns may not always align with the potential for public benefit from access to university knowledge and discoveries, let alone the impact of innovation centers on the training of undergraduate students. Transactions that will generate higher short-term returns may be inconsistent with efforts to “build an environment for innovation that is consistent with academic goals.”84 A number of stakeholders in university technology transfer have responded to critics of technology transfer—both those concerned with too much university focus on revenue and those concerned with too little focus on revenue—with efforts to promote and measure the broader social impact of university technology transfer. 85 In 2007, a small but influential group of research universities and the Association of Medical Colleges drafted a whitepaper entitled Nine Points to Consider in Licensing University Technology,86 and in 2010 the National Research Council made some key recommendations in a report entitled Managing University Intellectual Property in the Public Interest that served to reinforce the responsibility of universities to “maximize the further development, use and beneficial social impact of their technologies.”87 The AAU formed a working group in 2014 with the task of “reaffirming that the primary goal of university technology transfer operations is to advance the public interest.”88

82

See, e.g., id. See, e.g., James K. Woodell & Tobin L. Smith, Technology Transfer for All the Right Reasons, 18 teCh. & InnovatIon 295 (2017) (“If done with the right goals in mind, technology transfer aligns with universities’ overarching research, education and service missions, helping to ensure that public investment in science is impactful, that it advances broader economic development objectives, and that it serves the public interest.”). 84 See, e.g., Goldie Blumensky, A Contrarian Approach to Technology Transfer, Chron. hIgher eduC, Mar. 12, 2004. 85 See, e.g., Woodell & Smith, supra note 83. 86 Nine Points to Consider in Licensing University Technology, AUTM, available at https://autm .net/about-tech-transfer/principles -and-guidelines/nine-points -to-consider-when-licensing-university (last visited Feb. 6, 2019). 87 See, e.g., AAU Technology Transfer Working Group, Statement on Managing University Technology Transfer in the Public Interest, AAU, Mar. 1, 2015, available at https://www.aau.edu/ key-issues/aau-technology-transfer-working-group-statement-managing-university-technology-transfer (last visited Oct. 17, 2019). 88 See, e.g., Indicators of a Successful University Technology Transfer Office, AAU, July 20, 2017, available at https://www.aau.edu/key-issues/indicators-successful-university-technology-transfer-office (last visited Oct. 17, 2019). 83

The innovation arms race on academic campuses 209 This working group recognized that a necessary part of achieving this goal was to develop better metrics and criteria for evaluating university technology transfer in ways that capture the contributions made to “discovery, learning, and the promotion of social well being.” 89 The APLU and the AAU issued recommendations in 2015 encouraging their members to reaffirm their commitment to pursuing the public interest in their management of IP.90 AUTM has also responded with metrics that demonstrate both the economic value and the social impact of university R&D and technology transfer.91 AUTM’s Better World Project provides one example of how broader measures of impact can be included in the conversation.92 Although broader measures of impact exist and continue to improve, they have not been adequately incorporated into federal government policy debates and are often ignored even at the university level when evaluating performance and deciding how to allocate resources. This deficit in the evaluation of technology transfer is likely to be compounded by the proliferation of new innovation initiatives.

THE IMPACT OF THE INNOVATION ARMS RACE ON TTOS

V.

In a knowledge-based economy, it’s more important than ever for public universities to engage their technology transfer efforts as part of their broader mission to drive prosperity … Institutions should continue to redefine expectations for their technology transfer offices, and connect them with other aspects of industry and entrepreneurial partnerships. Technology transfer must serve to help institutions’ broader strategic goals.93

In their pursuit of innovation, universities have tasked TTOs with achieving multiple, sometimes amorphous and sometimes competing, objectives. “[T]he debate over the role of TTOs in the last few years has toggled between reaping value from IP to benefit the university, to being agents of economic development to help the local economy, to providing a bridge between academia and local industry.”94 While TTOs have long served as hubs connecting internal and external parties interested in developing and commercializing new discoveries, the expansion in innovation programs has dramatically expanded the need for and role of such hubs. As noted in the Department of Commerce report focusing on university technology transfer, “[i]nstead of merely focusing on the commercialization of individual technologies, these offices now act as a central point where students, faculty, alumni, entrepreneurs, investors, and industry can connect with each other. These offices are now focused on identifying and supporting entrepreneurship on campus, helping startups find the best opportunities and 89

Managing University Intellectual Property in the Public Interest (2015); see also Indicators of a Successful University Technology Transfer Office, supra note 88. 90 See, e.g., Woodell & Smith, supra note 83. 91 See, e.g., Our Surveys: Tracking the Innovation Ecosystem, AUTM, available at https://autm.net/ surveys-and-tools/surveys/licensing-survey (last visited Jan. 15, 2019); see also Pressman, et al., supra note 77. 92 See Better World Project, http://www.betterworldproject.org/ (last visited Jan. 15, 2019) (showing the project designed to communicate the social impact of technology transfer to the public). 93 APLU President McPherson; see, e.g., Report Calls on Universities to Evolve Tech Transfer Practices to Maximize Impact, APLU, Nov. 14, 2018, available at http://www.aplu.org/news-and -media/News/aplu-report-calls-on-universities-to-evolve-tech-transfer-practices-to-maximize-impact (last visited Oct. 17, 2019). 94 Fishburn, supra note 26 (quoting Keven Cullen).

210 Research handbook on intellectual property and technology transfer building successful business models, changing the culture of their universities, and creating companies that will be based in the communities around the university.” 95 The expansion of responsibilities, including in particular pressure to contribute to economic development through technology transfer, has not always been accompanied by an expansion in training, budget, or resources for TTOs.96 Indeed, in many cases TTOs have felt intensified pressures to cover their costs. Budget constraints have necessitated trade-offs between competing demands for time and resources. In addition, while many of the innovation programs are new, there is a continued reliance on old and fairly narrow metrics for performance and outdated incentive structures for faculty and other participants in university activities. In this concluding section, we suggest the need for a critical re-evaluation of what TTOs can and should be doing, and what resources they need to do it, in the face of proliferating innovation programs and intensifying pressure to perform on relatively narrow metrics such as licensing revenue. While the impacts will vary across institutions, the proliferation of these programs and the pressures that are driving them create some shared areas of concern for TTOs. One area of concern lies in measures of impact. Universities have been increasing their spending on innovation programs even while funding in other areas has stagnated, or dwindled, and in some cases the investment has been significant. This leads to the as-of-yet unanswered question of how to evaluate the performance of innovation programs and measure their impact.97 There is a dearth of empirical work to show whether the proliferation of innovation programs will indeed lead to more innovation and there is little if any publicly available information about whether and/or which innovation programs are cost-effective ways of achieving desired objectives.98 Indeed, there is very little accessible performance data of any kind, at least in part because many of the initiatives are new and their impacts are hard to measure. TTOs continue to play a role, in some cases a central role, in these different types of initiatives, resulting in an expanding domain of effort and activity. At the same time, they face stagnating budgets and heightened scrutiny of their operating expenses from university administrators anxious to curtail expenses. Without additional resources, new projects must come at the expense of time and effort spent on existing projects. But without good metrics for measuring the impact of these new initiatives, perceptions about the performance of TTOs may fail to capture both their efforts and investments in these programs and the relative value of pursuing these programs rather than more traditional technology transfer functions. A second area of concern is the impact of the pressures discussed in this Chapter on the role of TTOs as gatekeepers of university knowledge and technology. TTOs sit at the intersection of academic research and the commercialization of that research and they are charged with the transfer of this research and the knowledge it embodies in ways that further the public interest. Where too much emphasis is placed on facilitating university-industry partnerships and expanding collaborations, or the pursuit of other market-based initiatives, the gate-keeping 95

Innovative & Entrepreneurial U. Rep., supra note 54. See, e.g., Jisun Kim, et al., Assessing University Technology Transfer: A Measure of Efficiency Patterns, 5 Int’l J. InnovatIon & teCh. MgMt. 495 (2008) (examining ways of identifying and measuring the efficiency patterns of university technology transfer). 97 See, e.g., Julian Wyllie, Colleges have Spent Big Money on Innovation Centers. Do They Work?, Chron. hIgher eduC., Apr. 13, 2018. 98 See, e.g., Leary, supra note 2 (explaining the lack of reliable statistics about the growth and number of innovation and entrepreneurship programs and the lack of clear measures of impact and points to the concern that education is now being judged in market terms). 96

The innovation arms race on academic campuses 211 role becomes harder to sustain. Moreover, the value of this gatekeeper role as a mechanism for balancing competing interests, such as trade offs between the open dissemination of knowledge and IP management to support product development, is hard to quantify in the same way as number of licenses and revenue generated. A third area of concern is the potential for both fragmentation and confusion as the number of innovation programs and innovation offices on university campuses expand. The amorphous nature of the term, along with its popularity, can lead to very different offices, initiatives, and programs on campus with similar names. The desire of different units across the university to lay claim to the term in some way can fuel this trend, creating the potential for fragmentation of resources and overlapping but disconnected agendas. In response to these concerns, and to some of the more general trends discussed in this Chapter, we conclude that more attention needs to be paid to the impact of the innovation arms race on TTOs that have long viewed the promotion of innovation through traditional technology transfer activities as their goal. We also argue for a renewed policy focus on existing well-established metrics capturing the returns on technology transfer, such as the licensing survey data that AUTM has collected for over twenty-five years, and related efforts to capture and document the broader economic benefits of technology transfer.99 We begin with the idea that “[t]he fundamental purpose of university TTOs is to ensure that federally funded and other university research outcomes serve the public interest.”100 Ideally, “[t]echnology transfer is a mechanism by which universities ensure that public investment in science is impactful, that such investments enhance economic development, and that it serves the public interest.”101 From this starting point we can explore how to evaluate the impact of innovation programs on technology transfer. A first step in evaluating the impact of innovation and entrepreneurship programs on TTOs is to identify what specific function(s) new innovation strategies and initiatives are designed to perform. In some cases the function may be an educational one, such as providing entrepreneurship training to students and even faculty, or providing environments that foster creativity and entrepreneurial thinking. Where this is the objective, university stakeholders can debate questions such as whether entrepreneurship can be taught, how it should be taught, and with what implications for other educational objectives. In other cases the goal may be to increase the number of faculty-generated startups, or to fund the development of early-stage technology through a proof-of-concept stage. A second and related step is to ensure that the programs actually address the desired goals and that in doing so they do not conflict with other goals or compromise the values central to the university mission. The mere existence of a university IP policy may be viewed as a barrier to encouraging students to engage in entrepreneurial activities, for example, even though most university IP policies already exclude undergraduate IP. Creating too many exceptions to institutional IP ownership, on the other hand, can have unintended consequences

99

See, e.g., AUTM STATT database, AUTM, available at https://autm.net/surveys-and-tools/ databases (last visited Jan. 15, 2019) (providing more than 25 years of licensing data); see also Pressman, et al., supra note 77 (summarizing results of a model designed to estimate the economic impact of academic licensing over time). 100 Woodell & Smith, supra note 83 (arguing that revenue generation is not and should not be the primary motivation for university technology transfer). 101 Id.

212 Research handbook on intellectual property and technology transfer on commercializing technology. As another example, the wholesale removal of barriers to privatization of federally-funded technologies does not necessarily translate into maximizing economic (or social) benefits from the technologies. There are likely to be different views, both within the university and outside of it, about what constitutes the public interest and how to measure it, including different views about the value of open science and the dissemination of knowledge as part of the technology transfer process. There are also likely to be different valuations placed on the fruits of technology transfer, such as the value of new products and services with significant social value but low market value. Current policy discussions about the performance of technology transfer are largely framed in terms of improving returns on investment by removing barriers to commercialization in support of economic growth. But this discussion involves important underlying assumptions about how to measure and value the benefits of technology transfer that need to be addressed in a transparent manner and with the university mission in mind. A third step is to consider whether the innovation program is best suited to the objective and, if so, what the role of the TTO should be as part of that project. It is important to make the distinction between programming aimed at creating educational opportunities for students interested in innovation versus programming aimed at increasing the translation of ideas into innovative new products and services or university companies producing the same. These are very different objectives and often require different resources and deserve different metrics. Discussions about innovation initiatives often fail to distinguish between these very different objectives and to consider the best way of pursuing them. Too often, a primary objective as well as a less-likely, but hoped-for secondary objective, are identified as program goals, creating confusion among stakeholders down the road who expect both to be achieved. Related to this step is the need for better metrics of impact. This includes both the pursuit of data that captures the full cost of developing and running innovation programs as well as finding new ways of effectively measuring the economic and social impact of these programs. It is also essential to recognize the limitations of metrics, particularly when used to forecast future performance and future impact of new programs. We caution that university innovation and entrepreneurship initiatives often promise more than they can deliver and, in doing so, generate unrealistic expectations about the immediate and direct economic returns realistically available from university technology transfer, creating beliefs about performance that may not capture underlying realities. Finally, where resources are limited, it is important to consider the tradeoffs that are made when programs multiply. We caution that by expanding the domains of university activity into sometimes amorphous realms of “innovation” without a careful analysis of both the expected cost and social impact of such efforts, universities run the risk of fragmenting resources and distorting traditional technology transfer functions of the university in ways that could limit rather than enhance the public benefits of technology transfer. In order to navigate the challenges and harness the opportunities that accompany the proliferation of innovation programs across university campuses, we suggest the need for universities to develop a comprehensive strategic plan for their innovation activities that incorporates the steps outlined above. The goals of such a plan would include reducing redundancies, mitigating confusions, identifying synergies, and enabling the effective and efficient pursuit of key opportunities in ways that are consistent with the public interest. To achieve these goals, such a plan would include a consistent process for reviewing new innovation initiatives on campus to ensure that adequate resources have been identified, key stakeholders are engaged,

The innovation arms race on academic campuses 213 and measures for success are considered up front. The review process would also provide an opportunity for the university to balance the cost and merits of competing proposals to ensure effective use of limited university resources, with reasoned explanations as to why some initiatives receive support and others do not. The plan would also provide opportunities to clarify the role and responsibilities of respective innovation programs and to coordinate the naming of different initiatives, avoiding the confusion that arises from multiple “Innovation Centers” on campus. In addition to this work at the front end when initiating new programs, the strategic plan should include mechanisms for convening appropriate offices and programs focused on innovation on a regular basis. These meetings can be used to promote cross-awareness, to ensure that programs support each other rather than compete with each other, to ensure the proper sequencing of actions that implicate university IP policies, and to find ways of providing a seamless support system for faculty, staff, and students looking to engage in innovation.

VI.

CONCLUSION

In this Chapter we have explored the potential impact of the innovation arms race on university campuses for TTOs and their core functions in ensuring public benefit from federally-funded research. Universities, including their TTOs, need to respond to changing times with new strategies and with the flexibility to harness and pursue opportunities for leveraging the knowledge they create to promote public benefit. But when expanding into the tantalizing and evolving ethos of “innovation,” universities need to carefully consider both the relative costs and benefits of new programs, the dangers of fragmentation of university resources and attention, and the alignment of alternative programs with the values of higher education. While there is much excitement about the entrepreneurial and innovative university, there is little empirical data as of yet to judge either the cost effectiveness or the economic impact of the initiatives such as academic innovation centers, incubators, venture funds, makerspace labs, and other programs that have sprung up to promote innovation.102 We conclude that while new innovation programs offer exciting opportunities, traditional technology transfer (well established as a mechanism for disseminating university discoveries to the public) remains a critical function for promoting innovation. We caution against too much fragmentation of commercialization efforts across different programs that might unintentionally cannibalize existing programs and overwhelm busy faculty researchers. In response to the increasing national focus on a ROI approach to technology transfer, we suggest the need for better metrics to capture the social impact of technology transfer functions as well as other programs that enhance their impact, and we emphasize the need to pay greater attention to the broader technology transfer metrics already in place.

102 See, e.g., Beth McMurtrie, The Hope and Hype of the Academic Innovation Center, Chron. hIgher eduC., Jan. 21, 2018; Wyllie, supra note 97 (exploring the creation of university innovation centers designed to be hubs for new business formation, argues for broad consideration of how to measure impact with focus on education); Jodi Helmer, Operation innovation: Startups on Campus, U. Bus, July 27, 2016, available at https://www.universitybusiness.com/article/operation-innovation -startups-campus.

10. Tacit knowledge and university-industry technology transfer Peter Lee

We can know more than we can tell.1

I.

INTRODUCTION

Traditional conceptions of university-industry technology transfer typically focus on patenting and licensing of academic inventions. However, effective technology transfer often requires significant knowledge exchange between academic and commercial entities in parallel to patent licensing. Although patents on university technologies nominally disclose those inventions, a significant amount of knowledge related to practicing and commercializing them remains tacit or uncodified, residing in the mind of the faculty inventor. This Chapter explores the nature of tacit knowledge and mechanisms for transferring it. It reveals that human and institutional connections play a critical role in transferring tacit knowledge between universities and commercial firms and thus facilitating technology transfer. In exploring these dynamics, this Chapter does not seek to diminish the importance of formal patenting and licensing as well as the legal and institutional framework that shapes it. Certainly, the governing federal statutory regime has been highly impactful. Particularly noteworthy is the 1980 Bayh-Dole Act, which allows and encourages recipients of federal funding (including universities) to take title to patents arising from publicly funded research.2 While university patenting was already on the rise at the time of its enactment,3 the Act greatly accelerated such activity.4 This statutory innovation dovetailed with other legal changes, such as expansive Supreme Court rulings on patentable subject matter5 and the creation of the Court of Appeals for the Federal Circuit in 1982,6 to create an environment that accelerated university patenting.7 These legal changes helped spur institutional changes, such as the establishment of

1

Michael Polanyi, The Tacit Dimension 4 (1967) [hereinafter Polanyi, The Tacit Dimension]. 35 U.S.C. §§ 200–211 (Supp. 2009). The Bayh-Dole Act and its impact are emblematic of the “triple helix” model of greater interaction among government, academia, and business. See Henry Etzkowitz & Riccardo Viale, Polyvalent Knowledge and the Entrepreneurial University: A Third Academic Revolution?, 36 CrIt. soC. 595, 600 (2010). 3 David C. Mowery et al., Ivory Tower and Industrial Innovation: University-Industry Technology Transfer Before and After the Bayh-Dole Act in the United States 104 (2004). 4 Peter Lee, Patents and the University, 63 duke l.J. 1, 34 fig.1 (2013) [hereinafter Lee, Patents and the University]. 5 See, e.g., Diamond v. Chakrabarty, 447 U.S. 303, 310 (1980); Diamond v. Diehr, 450 U.S. 175, 192 (1981). 6 The Federal Courts Improvement Act, 96 Stat. 25 (1982). 7 Additionally, scientific advances such as the discovery of recombinant DNA technology also accelerated university patenting. See Lee, Patents and the University, supra note 4, at 32–3. 2

214

Tacit knowledge and university-industry technology transfer 215 hundreds of technology transfer offices (“TTOs”) on university campuses to manage academic patenting and licensing.8 These TTOs collect invention disclosures from university scientists, determine whether to pursue patenting, coordinate prosecution, find licensees, and negotiate licenses with commercial partners. In the traditional view, technology transfer is effectuated when a TTO licenses a university patent to a commercial licensee. This Chapter, however, reveals that patent licensing alone frequently does not provide the requisite knowledge necessary to transfer and develop university inventions. While patents must disclose an invention, much invention-related knowledge remains tacit and difficult to access by a commercial licensee. Part II of this Chapter addresses the nature of tacit knowledge, which ranges from purely tacit knowledge incapable of codification to latent knowledge that is codifiable but uncodified. It further observes that tacit knowledge is highly valuable to commercial licensees both for practicing a basic invention and for commercializing it. It also explores the difficulty of transferring tacit knowledge, which usually requires direct interpersonal interaction between inventor and transferee. Part III draws on these observations to explore a variety of relational and institutional mechanisms for transferring patent-related tacit knowledge. Among other functions, networks, consulting engagements and corporate positions, sponsored research, proof of concept centers and incubators, and university spinoffs all facilitate the transfer of tacit knowledge from academic to commercial entities. Part IV explores several implications of the central role of tacit knowledge in university-industry technology transfer, including the geographically constrained nature of technology transfer and normative concerns over greater intermeshing of academic and commercial entities.

II.

TACIT KNOWLEDGE

The act of invention draws upon and generates multiple kinds of knowledge, some of which are more easily communicated than others. Even for university inventions that have been disclosed in a patent, tacit knowledge on the part of a faculty inventor can be immensely valuable for a commercial licensee seeking to adopt and commercialize a technology. This Part explores the nature of tacit knowledge, which varies in degree for different types of technologies. It further examines the commercial value of such knowledge, which is helpful for both practicing a basic invention and for commercially developing it. Finally, it discusses the costly, human-centric processes necessary to transfer tacit knowledge, thus rendering such transfer one of the central challenges of commercializing university inventions. A.

Dimensions

Numerous commentators have explored the importance of tacit knowledge in innovation and technology transfer.9 In his seminal writings,10 scientist and philosopher Michael Polanyi

8 Richard R. Nelson, Observations on the Post-Bayh-Dole Rise of Patenting at American Universities, 26 J. teCh. transfer 13, 13 (2001). 9 See Jacqueline Senker, Tacit Knowledge and Models of Innovation, 4 Indus. & Corp. Change 425, 425 (1995) (collecting sources). 10 See generally Michael Polanyi, Science, Faith and Society (1964); Michael Polanyi, Personal Knowledge: Towards a Post-critical Philosophy (1958) [hereinafter Polanyi, Personal Knowledge];

216 Research handbook on intellectual property and technology transfer famously observed: “We can know more than we can tell.”11 Tacit knowledge is that which is personal to an individual and not easily reduced to codification,12 such as the ability to recognize faces, ride a bicycle or swim.13 As Dan Burk observes, “[s]ome types of knowledge may be inherently uncodifiable because some cognitive capacities resist explicit articulation.”14 Such knowledge is experiential and comprises “rules of thumb, heuristics, and other ‘tricks of the trade’.”15 For instance, while a champion golfer can verbally describe how to hit a golf ball, much of the knowledge for actually performing this motion is tacit, residing in subconscious understanding informed by years of practice, muscle memory and athletic intuition.16 Moving beyond the sporting world, when an inventor creates a new technology, she draws upon and produces a significant amount of tacit knowledge.17 Such “non-codified, disembodied know how” arises from the internalization of learned behaviors and procedures18 and is difficult to communicate. For example, tacit knowledge is highly relevant to biotechnology,19 an area of significant university innovation.20 Modern biotechnology involves the use of recombinant DNA technology and monoclonal antibodies to produce biologic compounds based on living material.21 According to the Office of Technology Assessment, “Because of their complex and unknown

Polanyi, The Tacit Dimension, supra note 1; see also Richard R. Nelson & Sidney G. Winter, An Evolutionary Theory of Economic Change 76–82 (1982); Robin Cowan et al., The Explicit Economics of Knowledge Codification and Tacitness, 9 Indus. & Corp. Change 211, 211 (2000). 11 Polanyi, The Tacit Dimension, supra note 1, at 4; see also Partha Dasgupta & Paul A. David, Toward a New Economics of Science, 23 res. pol’y 487, 493 (1994). 12 Lynne G. Zucker et al., Commercializing Knowledge: University Science, Knowledge Capture, and Firm Performance in Biotechnology, 48 MgMt. sCI. 138, 140 (2002) [hereinafter Zucker et al., Commercializing Knowledge]. 13 Senker, supra note 9, at 426. 14 Dan L. Burk, The Role of Patent Law in Knowledge Codification, 23 berkeley teCh. l.J. 1009, 1014 (2008). 15 Ashish Arora, Contracting for Tacit Knowledge: The Provision of Technical Services in Technology Licensing Contracts, 50 J. dev. eCon. 233, 234 (1996) [hereinafter Arora, Contracting for Tacit Knowledge]; see also Érica Gorga & Michael Halberstam, Knowledge Inputs, Legal Institutions and Firm Structure: Towards a Knowledge-Based Theory of the Firm, 101 nw. u. l. rev. 1123, 1144 (2007). 16 Burk, supra note 14, at 1014. 17 Polanyi, Personal Knowledge, supra note 10, at 52; Joanne E. Oxley, Appropriability Hazards and Governance in Strategic Alliances: A Transaction Cost Approach, 13 J. l. eCon. & org. 387, 393 (1997); see also Gary P. Pisano, Can Science Be a Business? Lessons from Biotech, harv. bus. rev., Oct. 2006, at 10 [hereinafter Pisano, Can Science Be a Business?]. 18 Jeremy Howells, Tacit Knowledge, Innovation and Technology Transfer, 8 teCh. analysIs & strategIC MgMt. 91, 92 (1996). Economics and management scholarship typically uses the term “tacit knowledge,” while the applied science and research literature typically refers to “know-how.” Margaret Chon, Sticky Knowledge and Copyright, 2011 wIs. l. rev. 177, 187. This Chapter generally uses the term “tacit knowledge.” 19 Senker, supra note 9, at 430. 20 See generally Zucker et al., Commercializing Knowledge, supra note 12; Gary P. Pisano et al., “Joint Ventures and Collaboration in the Biotechnology Industry” in International Collaborative Ventures in U.S. Manufacturing 183, 187 (David C. Mowery ed., 1988) (“Universities and nonprofit research organizations have shaped the scientific foundations for commercial biotechnological development.”). 21 See Pisano et al., supra note 20, at 184 (offering a concise overview of biotechnology, including modern techniques based on recombinant DNA technology and monoclonal antibodies).

Tacit knowledge and university-industry technology transfer 217 nature, many biological inventions, especially organisms, cannot be sufficiently described in writing to allow their predictable reproducibility on the basis of that description alone.”22 Thus, even if a university scientist describes a biotechnological invention in an academic journal or a patent, much knowledge related to that invention remains tacit.23 The centrality of tacit knowledge to comprehending an invention, moreover, complicates efforts to transfer that invention to private parties for further development and commercialization. At this point, it is useful to address how patented inventions, including university technologies, may not disclose valuable tacit knowledge. After all, patent law requires that applicants disclose their inventions “in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same.”24 Furthermore, patent law requires that applicants “shall set forth the best mode contemplated by the inventor of carrying out his invention.”25 Indeed, the essential quid pro quo of the patent system comprises an exchange of exclusive rights for technical disclosure.26 While the disclosure requirements of patentability suggest that patented inventions should have little remaining tacit content,27 that is often not the case;28 the enablement and written description requirements still permit many knowledge gaps.29 Patent law does not require applicants to provide detailed production specifications30 or to disclose their inventions thoroughly enough to eliminate the need for technical artisans to engage in some experimentation to practice them.31 Patents applicants have strong incentives to disclose as little as possible,32 and they often employ unwieldy jargon and formalism33 that obfuscates rather than illuminates. Furthermore, patent disclosure occurs early in the application process and is generally fixed, thus discouraging the applicant from disclosing newly discovered information during prosecution.34 Additionally, the “best mode” requirement—which mandates disclosure of specific techniques and instrumentalities known to the applicant as the best way of practicing an invention—is largely a dead letter after statutory reforms in 2011.35 According to one

22

Office of Technology Assessment, Commercial Biotechnology—An International Analysis 368 (1984). 23 Peter Lee, Innovation and the Firm: A New Synthesis, 70 stan. l. rev. 1431, 1446 (2017) [hereinafter Lee, Innovation and the Firm]. 24 35 U.S.C. § 112(a) (2012). 25 Id. 26 Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 484 (1974). 27 Burk, supra note 14, at 1012. 28 Douglas Lichtman, How the Law Responds to Self-Help, 1 J.l. eCon. & pol’y 215, 255 (2005) (“The information available on the face of a patent document is rarely revealing.”). 29 See, e.g., Timothy R. Holbrook, Possession in Patent Law, 59 s.M.u. l. rev. 123, 132–46 (2006); James Bessen, Patents and the Diffusion of Technical Information, 86 eCon. letters 121, 122 (2005); Jeanne C. Fromer, Dynamic Patent Disclosure, 69 vand. l. rev. 1715, 1718 n.13 (2016) (collecting sources). 30 In re Gay, 309 F.2d 769, 774 (C.C.P.A. 1962). 31 See In re Wands, 858 F.2d 731 (Fed. Cir. 1988) (exploring the “undue experimentation” standard for assessing compliance with the enablement requirement). 32 See Brenner v. Manson, 383 U.S. 519, 534 (1966) (critiquing patent drafting techniques that disclose little while claiming much). 33 Sean B. Seymore, The Teaching Function of Patents, 85 notre daMe l. rev. 621, 626 (2010). 34 Fromer, supra note 29, at 1719–20. 35 See 35 U.S.C. § 282(b)(3)(A) (2012) (establishing that the failure to disclose a best mode is not a permissible ground for invalidating a patent or rendering it unenforceable).

218 Research handbook on intellectual property and technology transfer observer, “even if a university invention is highly codified in terms of a pending patent, some knowledge that is relevant for commercialization is still tacit and has to be made available by human interactions.”36 In exploring the nature of tacit knowledge, several distinctions are helpful. First, tacitness is not a binary designation but a question of the degree.37 At one end of the spectrum lies purely tacit knowledge that is not capable of expression in explicit form.38 Toward the other end of the spectrum is “latent” knowledge that is technically codifiable but uncodified.39 For example, a university scientist who creates a new recombinant DNA may not reveal certain details of the best way for producing it, even though those details are capable of codification. Latent knowledge may remain uncodified for a variety of reasons, including the high cost of codification,40 a lack of incentive to codify, or an affirmative desire not to codify to advance self-interest.41 For instance, while a scientist may report research findings in an academic article, she may not feel it is worthwhile to disclose a novel measuring instrument that she developed in the course of her research; this research “byproduct” may be highly valuable yet remain in tacit form.42 Toward the far end of the spectrum lies explicitly codified knowledge, perhaps as disclosed in a patent or article. Even explicit knowledge, however, requires tacit knowledge to be understood and acted upon.43 Second, some commentators distinguish tacit knowledge, which rests upon understanding, from skills, which encompass the ability to make something happen, such as through manual dexterity or sensory ability.44 Other commentators, however, group tacit knowledge and skills together.45 Third, utilizing confusingly similar terminology, commentators also observe that different types of tacit knowledge manifest in different types of entities. “Skills,” such as those necessary to perform a difficult experimental

36

Fritjof Karnani, The University’s Unknown Knowledge: Tacit Knowledge, Technology Transfer and University Spin-Offs Findings from an Empirical Study Based on the Theory of Knowledge, 38 J. teCh. transfer 235, 238 (2012). 37 Michael D. Santoro & Shanthi Gopalakrishnan, Assimilating External Knowledge: A Look at University-Industry Alliances, proC. of pICMet 15 227, 229 (2015) (describing a continuum between tacit and explicit knowledge). 38 Cf. Nelson & Winter, supra note 10, at 73 (characterizing the knowledge underlying serving a tennis ball as largely tacit). 39 See Ajay Agrawal, Engaging the Inventor: Exploring Licensing Strategies for University Inventions and the Role of Latent Knowledge, 27 strategIC MgMt. J. 63 (2006) [hereinafter Agrawal, Engaging the Inventor]. 40 James E. Bessen, From Knowledge to Ideas: The Two Faces of Innovation 3 (Mar. 1, 2011) (unpublished manuscript), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1698802 (last visited Apr. 24, 2019) [hereinafter Bessen, From Knowledge to Ideas]. 41 See Chon, supra note 18, at 191; cf. H.M. Collins, The TEA Set: Tacit Knowledge and Scientific Networks, 4 sCI. stud. 165, 180 (1974) (“Let’s say I’ve always told the truth, nothing but the truth, but not the whole truth.”) (quoting a scientist involved in transferring the TEA laser to another laboratory). 42 Cf. Karnani, supra note 36, at 236. 43 Collins, supra note 41, at 167 (“All types of knowledge, however pure, consist, in part, of tacit rules which may be impossible to formulate in principle.”); Miriam Knockaert et al., The Relationship Between Knowledge Transfer, Top Management Team Composition, and Performance: The Case of Science-Based Entrepreneurial Firms, 35 entrepreneurshIp theory & praC. 777, 780 (2010). 44 Senker, supra note 9, at 427. 45 Ikujiro Nonaka, A Dynamic Theory of Organizational Knowledge Creation, 5 org. sCI. 14, 16 (1992).

Tacit knowledge and university-industry technology transfer 219 protocol, reside in individuals, while “routines,” such as laboratory workflow processes, reside in organizations.46 Continuing the theme of distinctions, the tacit dimension of any given technology is highly context specific.47 Some inventions are easily comprehensible even absent any formal codification. Many patented commercial products are “self-disclosing” in that they intrinsically reveal how they are made and used.48 Similarly, even early-stage university inventions may self-disclose their mode of manufacture and function. For example, an early-stage mechanical device developed in a university engineering department may be easily comprehensible even absent explicit codification. For such straightforward inventions, the tacit dimension is relatively low, thus easing technology transfer.49 For other inventions, codification through patenting, publishing or other means is sufficient to adequately disclose and transfer an invention. At the far end of the spectrum, however, are inventions in more “unpredictable” arts like chemistry and experimental sciences50 that may possess a significant tacit dimension. These technologies do not self-disclose how they operate, and codification cannot adequately describe how to make and use them. As we will see, early-stage university inventions encompassing cutting-edge technology tend to have a high tacit dimension.51 On a related note, tacitness is also a function of time. New information tends to emerge in tacit form but become codified over time52 as “recipe knowledge.”53 In particular, knowledge pertaining to a pioneering scientific discovery may initially arise in tacit form, as only the discoverer herself can understand and exploit it. Over time, however, knowledge tends to lose its tacit dimension as pioneering discoveries become widely shared background knowledge. For instance, early biotechnology innovations maintained a significant tacit dimension for the first ten or fifteen years after discovery, after which they became more readily accessible by the broader scientific community.54 B.

Commercial Value

The tacit knowledge retained by a faculty inventor, even when the inventor has disclosed the invention in a patent, is highly valuable. “Utilizing technology requires a great deal

46

See Nelson & Winter, supra note 10, at 14, 72–73. See, e.g., Senker, supra note 9, at 439 tbl.3 (differentiating between different sources of knowledge in biotechnology, ceramics and parallel computing). 48 Katherine J. Strandburg, What Does the Public Get? Experimental Use and the Patent Bargain, wIs. l. rev. 81, 83 (2004) (defining self-disclosing inventions as those “that are easily copied from their commercial embodiments”); Frank H. Easterbrook, Intellectual Property Is Still Property, 13 harv. J.l. & pub. pol’y 108, 109 (1990) (“The product itself, not the patent papers, usually discloses things.”). 49 Raymond W. Smilor & David V. Gibson, Technology Transfer in Multi-Organizational Environments: The Case of R&D Consortia, 38 Ieee transaCtIons on engIneerIng MgMt. 3 (1991). 50 See Sean B. Seymore, Heightened Enablement in the Unpredictable Arts, 56 uCla l. rev. 127, 137–9 (2008). 51 See infra notes 62–8 and accompanying text. 52 Zucker et al., Commercializing Knowledge, supra note 12, at 140. 53 Id. at 141; see also Senker, supra note 9, at 431. 54 See Lynne G. Zucker et al., Intellectual Human Capital and the Birth of U.S. Biotechnology Enterprises, 88 aM. eCon. rev. 290, 291 (1998) [hereinafter Zucker et al., Intellectual Human Capital]. 47

220 Research handbook on intellectual property and technology transfer of know-how,”55 and tacit knowledge is often necessary to exploit codified knowledge.56 Accordingly, tacit knowledge is critical for effectively transferring many patented inventions.57 Patent licensees frequently recognize that “the aggregate value of all the ‘minor’ improvements, tweaks, and accumulated operational wisdom [related to a patent] often exceeds the value of the basic invention itself.”58 Not surprisingly, companies licensing patents often seek to license invention-related tacit knowledge as well.59 Indeed, “acquisition of tacit knowledge to support innovation is a purposive activity of much industrial development.”60 Within science-based fields typical of university-industry technology transfer, tacit knowledge exchange from a discovering scientist is positively correlated with a firm’s success in commercializing a technology.61 While tacit knowledge is valuable for many kinds of technology transfer, it is particularly critical for transferring academic inventions. As noted, new information tends to arise in tacit form and become codified over time.62 Relatedly, inventions that are more novel and complex are likely to have a higher tacit dimension.63 Given the pioneering, early-stage nature of academic inventions, tacit knowledge is particularly valuable for effective university-industry technology transfer.64 One empirical study revealed that over 75% of university inventions were merely proofs of concept or lab-scale prototypes at the time of licensing.65 Similarly, 55

Ashish Arora, Licensing Tacit Knowledge: Intellectual Property Rights and the Market for Know-How, 4 eCon. of InnovatIon & new teCh. 41, 42 and 54 (2006) [hereinafter Arora, Licensing Tacit Knowledge]. 56 Robin Cowan & Dominique Foray, The Economics of Codification and the Diffusion of Knowledge, 6 Indus. & Corp. Change 595, 599 (1997). 57 See Agrawal, Engaging the Inventor, supra note 39, at 65 (emphasizing the value of latent knowledge in technology transfer); Jonathan M. Barnett, Intellectual Property as a Law of Organization, 84 s. Cal. l. rev. 785, 801–2 (2011) (discussing “information opacity” as a barrier to technology transfer); Barry Bozeman, Technology Transfer and Public Policy: A Review of Theory and Research, 29 res. pol’y 627, 642 (2000) (“[T]he extent of transfer of tacit knowledge often has a major impact on the effectiveness of manufacturing technology transfer.”); Howells, supra note 18, at 97 (“[T]acit knowledge has been a key barrier in the diffusion of technological innovation.”). 58 Robert P. Merges, A Transactional View of Property Rights, 20 berkeley teCh. l.J. 1477, 1501 (2005). 59 See Arora, Licensing Tacit Knowledge, supra note 55, at 43. 60 Senker, supra note 9, at 428. 61 Zucker et al., Commercializing Knowledge, supra note 12, at 143. 62 See supra note 52 and accompanying text. 63 Zucker et al., Commercializing Knowledge, supra note 12, at 140–1. 64 Ajay Agrawal, University-to-Industry Knowledge Transfer: Literature Review and Unanswered Questions, 3 Int’l J. of MgMt. rev. 285, 293 (2001) [hereinafter Agrawal, Knowledge Transfer]; Kwanghui Lim, The Many Faces of Absorptive Capacity: Spillovers of Copper Interconnect Technology for Semiconductor Chips, 18 Indus. & Corp. Change 1249, 1252 (2009); Jerry G. Thursby & Marie C. Thursby, Are Faculty Critical? Their Role in University-Industry Licensing, 22 ConteMp. eCon. pol’y 162, 170 (2004); see also Fiona Murray, The Role of Academic Inventors in Entrepreneurial Firms: Sharing the Laboratory Life, 33 res. pol’y 643, 650 (2004); cf. Bruce Kogut & Udo Zander, Knowledge of the Firm, Combinative Capabilities, and the Replication of Technology, 3 org. sCI. 383, 389 (1992); Janet Bercovitz & Maryann Feldmann, Entrepreneurial Universities and Technology Transfer: A Conceptual Framework for Understanding Knowledge-Based Economic Development, 31 J. teCh. trans. 175, 181 (2006) (“In general, early stage technologies such as those originating at universities require more extensive research investment to reach commercial viability.”). 65 Richard Jensen & Marie Thursby, Proofs and Prototypes for Sale: The Licensing of University Inventions, 91 aM. eCon. rev. 240, 243 (2001).

Tacit knowledge and university-industry technology transfer 221 only 12% were ready for commercial use, and 8% had known manufacturing feasibility.66 Other research revealed that only 7% of universities technologies were ready for practical or commercial use at the time of licensing, and 40% were proofs of concept.67 Given the embryonic nature of many university technologies, the need to transfer tacit knowledge to facilitate commercialization is particularly acute.68 While tacit knowledge is useful for practicing a basic invention, it is particularly important for developing a basic invention into a commercial product.69 As noted, faculty inventors need not disclose detailed manufacturing specifications in order to obtain a patent,70 and universities often license their inventions at a relatively early stage of development.71 These patents typically cover basic versions of inventions that operate in a laboratory, which may be very different from commercial embodiments of a technology ready for market. For instance, producing a (patented) biologic compound in a laboratory is significantly different from industrially manufacturing it in mass quantities.72 A host of factors, including cell culture medium, oxygen levels, and temperature can affect the quality and effectiveness of such processes.73 While there is already a gap between what a patent discloses and the knowledge needed to fully practice a basic invention, there is an even larger gap between patent disclosure and the knowledge needed to produce a commercial product. The inventor’s tacit knowledge, which constitutes action-oriented practical intelligence,74 is highly valuable to commercial licensees. According to one entrepreneur, So much of what we call technology transfer is information transfer, knowledge transfer. It’s not something that could immediately be put into a product. It might be something that is a tidbit of knowledge that will help somebody in their development efforts at one of our companies.75

These “tidbits” of tacit knowledge are highly valuable for firms licensing university patents, and effective innovation often depends on transferring such know-how from the laboratory to production facilities.76

66

Id. Thursby & Thursby, supra note 64, at 167. 68 See Zucker et al, Commercializing Knowledge, supra note 12, at 141; Cowan & Foray, supra note 56, at 595; cf. Bessen, From Knowledge to Ideas, supra note 40, at 3–4 (noting the importance of personal interactions in transferring early-stage technologies). 69 See Agrawal, Engaging the Inventor, supra note 39, at 68. 70 See, e.g., DSL Dynamic Sciences v. Union Switch & Signal, 928 F.2d 1122 (Fed. Cir. 1991) (holding that a coupler mount assembly that was not commercial-grade was nonetheless reduced to practice for the purposes of patent law). 71 See supra notes 62–8 and accompanying text. 72 See Gary P. Pisano, The Governance of Innovation: Vertical Integration and Collaborative Arrangements in the Biotechnology Industry, 20 res. pol’y 237, 244 (1991) (describing the technical challenges of “develop[ing] a large-scale process which preserves the desirable characteristics of the product produced in the lab”) [hereinafter Pisano, Governance of Innovation]. 73 Id. 74 See generally Robert J. Sternberg et al., Practical Intelligence in Everyday Life (2000). 75 Donald S. Siegel et al., Toward a Model of the Effective Transfer of Scientific Knowledge from Academicians to Practitioners: Qualitative Evidence from the Commercialization of University Technologies, 21 J. engIneerIng teCh. MgMt. 115, 130 (2004). 76 Pisano et al., supra note 20, at 186. 67

222 Research handbook on intellectual property and technology transfer Extrapolating beyond this point, what commercial licensees may seek is not just the static tacit knowledge of a faculty inventor but also her dynamic problem-solving skills in an area of unique technical expertise, which is also a tacit quality. Faculty patentees may not have considered the challenges of commercialization when obtaining their patents, but they are uniquely suited to working on those challenges given their intimate understanding of the underlying technology.77 For example, in the field of biotechnology, “[i]n the absence of well-defined and well-understood scale-up recipes, ensuring product integrity requires extensive interaction between the scientists who designed a cell in the laboratory and bioprocessing engineers charged with developing the production process.”78 Tacit knowledge has a “dynamic” quality in that it encompasses the inventor’s ability to extrapolate from present understanding to solve problems and extend a technology beyond its basic form. C.

Challenges of Transfer

The immense value of tacit knowledge to firms commercializing academic technologies raises the question of how faculty inventors can transfer it.79 By definition, tacit knowledge is “sticky”80 and very difficult to communicate.81 Codification allows for relatively easy knowledge transfer,82 but tacit knowledge resides not in texts but in “people, institutions or routines.”83 Several factors complicate the transfer of tacit knowledge. First, transferring tacit knowledge is a deeply human endeavor that requires direct interpersonal interaction.84 In a classic account of tacit knowledge, Collins describes how laboratories building a TEA laser relied on personal visits, telephone calls or personnel transfer from a source laboratory that

77

Cf. Lee, Innovation and the Firm, supra note 23, at 1448–9 (discussing innovative human capital). Pisano, Governance of Innovation, supra note 72, at 244. Additionally, sometimes the downstream commercializing entity has tacit knowledge related to scaling up industrial manufacturing that the inventor herself lacks. Pisano et al., supra note 20, at 191; id. at 213 (“[T]acit knowledge from actual production experience often plays a key role in the development and application of process innovations.”). 79 Cf. Thursby & Thursby, supra note 64, at 162. 80 See Chon, supra note 18, at 177; see Cowan & Foray, supra note 56, at 598. 81 See Gorga & Halberstam, supra note 15, at 1142; cf. Burk, supra note 14, at 1015; Dasgupta & David, supra note 11, at 494. Difficulties of transferring tacit knowledge increase with its complexity. See Kogut & Zander, supra note 64, at 388; Agrawal, Engaging the Inventor, supra note 39, at 64. 82 See Bessen, From Knowledge to Ideas, supra note 40, at 2. 83 Cowan & Foray, supra note 56, at 596. 84 Senker, supra note 9, at 445 (“Personal interaction is required with the source of the new scientific or technological knowledge in order to capture fully the tacit dimension.”); Arora, Contracting for Tacit Knowledge, supra note 15, at 235 (“[Tacit knowledge transfer] requires, almost by definition, face-to-face contact.”); Howells, supra note 18, at 93 (“Above all, there are no clear market mechanisms which facilitate the transfer of tacit knowledge directly or by which it can be adequately measured.”); Scott Shane, Selling University Technology: Patterns from MIT, 48 MgMt. sCI. 122, 124 (2002) (“[W]hen information is tacit, it must be transferred through interpersonal contact, and economic actors must develop relationship-specific assets to facilitate that transfer.”); Oxley, supra note 17, at 393 (“[Tacit knowledge] is extremely difficult to transfer without intimate personal contact, involving teaching, demonstration, and participation.”); David B. Audretsch & Paula E. Stephan, Company-Scientist Locational Links: The Case of Biotechnology, 86 aM. eCon. rev. 641, 647 (1996); see Collins, supra note 41, at 177; Cowan et al., supra note 10, at 215 (describing the development of the TEA laser and observing that “none of the research teams which succeeded in building a working laser had done so without the participation of someone from another laboratory where a device of this type already had been put into operation”). 78

Tacit knowledge and university-industry technology transfer 223 had already built one.85 In the corporate context, tacit knowledge transfer “requires personal interaction, for example through secondment or training.”86 Focusing on university-industry technology transfer, tacit knowledge is best exchanged through direct interaction between academic inventors and firms that license their patents.87 In biotechnology, the complexity of underlying inventions “and their strong tacit character make direct contact between basic (university) and applied (commercial) researchers necessary for successful technology transfer.”88 In short, technology transfer typically unfolds as a “contact sport.”89 Compared to low-cost dissemination mechanisms such as written disclosure, the need to interact with an inventor imposes significant temporal and spatial constraints on transferring tacit knowledge. Second, tacit knowledge transfer arises through experience and practice.90 Simply listening to a faculty inventor describe a technology may not be enough; direct interaction and practical application is necessary to transfer and cultivate such knowledge. In similar fashion, scientists gain tacit knowledge not through lectures and manuals but through apprenticeship.91 Senker notes that tacit knowledge is “heuristic, subjective and internalized” and “is learned through practical examples, experience, and practice.”92 For this reason, companies operating in industries where know-how is critical “must be expert at both in-house research and cooperative research” with external partners such as academic scientists.93 Empirical analysis suggests that joint work at the lab bench is an important mode of transferring tacit knowledge.94 Studies have shown that joint laboratory work by star bioscientists (primarily faculty members) and corporate scientists has a consistently positive effect on the performance of biotech firms.95 Third and relatedly, tacit knowledge transfer takes time and repeated interactions.96 The practical, experiential nature of tacit knowledge acquisition takes significant time, which further raises the cost of transfer. Furthermore, there is a diffusionary element to tacit knowledge transfer wherein more contacts over time are likely to transfer more knowledge. Tacit knowledge exchange is not only informal, but it is also in some ways capricious, “a symptom of the lack of organization of inarticulated knowledge into visible, discrete, and measurable units.”97 Informal conversations and happenstance encounters can be important vehicles of tacit 85

Collins, supra note 41, at 177. Senker, supra note 9, at 428. 87 Agrawal, Engaging the Inventor, supra note 39, at 65; see Edwin Mansfield, Academic Research Underlying Industrial Innovations: Sources, Characteristics, and Financing, 77 rev. eCon. & stat. 55, 64 (1995). 88 Pisano et al., supra note 20, at 187. 89 Agrawal, Engaging the Inventor, supra note 39, at 65 (quoting the Director of the MIT Technology Licensing Office); see id. at 78 (“[P]rivate contracts (or at least relationships) between the acquiring firm and the inventor are required in order to perform the necessary knowledge transfer.”). 90 Senker, supra note 9, at 429. 91 Id. at 427. 92 Id. at 426. 93 Walter W. Powell et al., Interorganizational Collaboration and the Locus of Innovation: Networks of Learning in Biotechnology, 41 adMIn. sCI. Q. 116, 117–19 (1996) (citing Eric Von Hippel, The Sources of Innovation (1988)). 94 Zucker et al., Commercializing Knowledge, supra note 12, at 151. 95 Lynne G. Zucker et al., Geographically Localized Knowledge: Spillovers or Markets?, 36 eCon. InQ. 65 (1998). 96 See Bercovitz & Feldmann, supra note 64, at 181 (noting the importance of “close and ongoing interactions” to transfer early-stage technologies); Knockaert et al., supra note 43, at 780. 97 Collins, supra note 41, at 171. 86

224 Research handbook on intellectual property and technology transfer knowledge transmission. Unlike one-off market transactions, effective technology transfer— which often encompasses tacit knowledge exchange—unfolds in extended relationships.98 Empirical studies underscore the importance of direct participation of faculty inventors in successful university-industry technology transfer. A survey of TTOs at US research universities revealed that 71% of licensed inventions required faculty cooperation for successful commercialization.99 Other research revealed that roughly 40% of all university licenses required faculty involvement.100 From the perspective of companies commercializing academic discoveries, 70% of entrepreneurs in one survey recognized “informal transfer of know how” as an output of university technology transfer.101 In sum, personal interactions are critical to transferring tacit knowledge, which is highly valuable for commercially developing university inventions.

III.

MECHANISMS FOR TRANSFERRING TACIT KNOWLEDGE

Given the value and difficulty of transferring tacit knowledge, mechanisms are often necessary to effectuate such transfers in parallel to patent licensing. These mechanisms span informal vehicles such as professional and social networks to more formal arrangements such as consulting and sponsored research agreements. At the most formalized end of the spectrum, institutional integration between academic and commercial entities facilitates tacit knowledge exchange and technology transfer more broadly. While varied in their characteristics, all of these vehicles facilitate the interpersonal, practice-oriented and repeated interactions necessary for transmitting tacit knowledge.102 While this Part focuses on how faculty inventors, universities and firms exploit knowledge-exchange mechanisms to supplement patent licensing, these mechanisms perform this function outside of formal technology transfer as well.103 A.

Networks

At the most informal end of the spectrum, professional and social networks facilitate tacit knowledge transfer and provide introductions leading to more intensive knowledge

98

Peter Lee, Transcending the Tacit Dimension: Patents, Relationships, and Organizational Integration in Technology Transfer, 100 CalIf. l. rev. 1403, 1539 (2012) [hereinafter Lee, Transcending]; Bercovitz & Feldmann, supra note 64, at 182 (emphasizing the role of relationships in university-industry technology transfer); see generally Eric Von Hippel, Sources of Innovation (1988); cf. Powell et al., supra note 93, at 119–20 (noting how long-term relationships can approximate the economies of team learning within an integrated firm). 99 Jensen & Thursby, supra note 65, at 243. 100 Thursby & Thursby, supra note 64, at 170; see also Mowery et al., supra note 3, at 159–66, 169–76 (2004) (describing the importance of faculty inventor involvement in the development of gallium nitride, a wide band gap semiconductor, and the Ames II test). 101 Siegel et al., supra note 75, at 126, 130. 102 This is not an exhaustive list of transfer mechanisms. For instance, knowledge exchange also occurs through hiring research students and graduates. Bercovitz & Feldmann, supra note 64, at 177. 103 Cf. id. at 176–7 (“Universities’ relationships with industry are formed through a series of sequential transactions such as sponsored research, licenses, spin-off firms and the hiring of students.”).

Tacit knowledge and university-industry technology transfer 225 exchange.104 According to Barry Bozeman, “much of scientific and technical human capital is embedded in social and professional networks of technological communities.”105 Even before any patenting and licensing, companies utilize networks spanning academic communities to access the background knowledge necessary to exploit new innovations. Networks are particularly valuable when relevant knowledge is widely distributed among different sources, such as in the biotechnology industry.106 Technical personnel directly transfer tacit knowledge through personal and professional networks.107 More broadly, participation in such networks enhances the “absorptive capacity”108 of industrial scientists seeking to appropriate new academic discoveries.109 Furthermore, networks provide companies access to a different type of uncodified knowledge: information about promising new technologies under development in university laboratories.110 While casual encounters within networks may be too fleeting to transfer some tacit knowledge, these relationships can set the stage for more intensive knowledge exchange. As sociologist Walter Powell and his colleagues have observed, “Beneath most formal ties … lies a sea of informal relations.”111 Not surprisingly, academic engagement with industry is more common among well-connected faculty members with significant social capital.112 Interestingly, however, weak connections that traverse significant sociological distance may be the most important for innovation networks,113 as they open up new and unexpected sources of information. Empirical work by Powell and others on the biotechnology industry demonstrates the importance of networks in knowledge exchange between academic and commercial entities. Empirical data from the late 1980s and early 1990s demonstrate the close relationships between academic communities and leading biotech firms like Genentech and Chiron.114 “[B]iotech firms grow by being connected to benefit-rich networks,”115 and collaborative

104 See generally Laurel Smith-Doerr, “Network Analysis” in International Encyclopedia of Economic Sociology 472 (Jens Beckert & Milan Zafirovski eds., 2006); see Bercovitz & Feldmann, supra note 64, at 176 (“[A] growing literature documents the importance of social interaction, local networks, and personal communication in knowledge transmission.”). 105 Barry Bozeman et al., The Evolving State-of-the-Art in Technology Transfer Research: Revisiting the Contingent Effectiveness Model, 44 res. pol’y 34, 40 (2015). 106 Powell et al., supra note 93, at 119 (“[W]hen knowledge is broadly distributed and brings a competitive advantage, the locus of innovation is found in a network of interorganizational relationships.”). 107 Cf. Eric von Hippel, Cooperation Between Rivals: Informal Know-how Trading, 16 res. pol’y 291, 292 (1987). 108 Wesley M. Cohen & Daniel A. Levinthal, Absorptive Capacity: A New Perspective on Learning and Innovation, 35 adMIn. sCI. Q. 128 (1990). Absorptive capacity is a multidimensional phenomenon encompassing learning from external sources, transferring external knowledge, and assimilating such knowledge to develop new products. Santoro & Gopalakrishnan, supra note 37, at 227. 109 Powell et al., supra note 93, at 120. 110 Audretsch & Stephan, supra note 84, at 646; cf. Powell et al., supra note 93, at 187 (“Linkages with a major university or research institute (including teaching hospitals) are viewed by industry as necessary to track and exploit a rapidly expanding technological frontier.”). 111 Powell et al., supra note 93, at 120. 112 Markus Perkmann et al., Academic Engagement and Commercialisation: A Review of the Literature on University-Industry Relations, 42 res. pol’y 423, 429 (2013). 113 Collins, supra note 41, at 169. 114 Powell et al., supra note 93, at 141. 115 Id. at 139.

226 Research handbook on intellectual property and technology transfer networks represent a critical mechanism for sharing knowledge.116 Academic and industrial scientists form a common technological community in which professors spend sabbaticals at biotech firms, postdocs move between universities and companies, and universities hire leading commercial scientists.117 The quasi-academic nature of some biotechnology firms, which reward academic publishing, helps attract university researchers.118 In a nod to the theory of the firm, Powell describes the upstream-downstream linkages between universities and biotech firms as “virtual integration.”119 Among other functions, such networks serve to transmit tacit knowledge about academic discoveries to commercial partners. While much of the empirical work on networks has focused on biotechnology, the importance of networks likely extends to other fields where knowledge sources are dispersed, such as ceramics and software.120 B.

Consulting Engagements and Corporate Positions

While networks can directly facilitate tacit knowledge transfer, they also provide introductions that can lead to more intensive knowledge exchange. For instance, private companies frequently hire academic scientists as consultants,121 which further facilitates tacit knowledge transfer. Consulting relationships involving academic faculty have become quite common,122 and many of these relationships arise outside of the context of formal patenting and licensing. Indeed, empirical research indicates that companies value “academic engagement”—which spans consulting, collaborative research, contract research, ad hoc advice, and networking— significantly more than licensing university patents.123 However, companies also hire faculty inventors as consultants while licensing those inventors’ patents from a university.124 Not surprisingly, formal patenting and licensing of a university invention heightens firms’ interest in interacting directly with academic inventors.125 One study of licenses from the Mechanical Engineering as well as the Electrical Engineering and Computer Science departments at MIT found that involvement of the academic inventor increased the likelihood of commercialization and the amount of royalties generated.126 Several qualitative case studies of university-industry licensing highlight the importance of

116 See Walter W. Powell, Inter-organizational Collaboration in the Biotechnology Industry, 152 J. InstItutIonal & theoretICal eCon. 197, 208 (1996); Powell et al., supra note 93, at 138–9 (“[N]etworks of collaboration provide entry to a field in which the relevant knowledge is widely distributed and not easily produced inside the boundaries of a firm or obtained through market transactions.”). 117 Powell et al., supra note 93, at 123. 118 Id. 119 Powell, supra note 116, at 209. 120 Powell et al., supra note 93, at 143. 121 Bercovitz & Feldmann, supra note 64, at 178. 122 Karen Seashore Louis et al., Entrepreneurs in Academe: An Exploration of Behaviors Among Life Scientists, 34 adMIn. sCI. Q. 110, 113 (1989); see also Martin Kenney, Biotechnology: The University-Industrial Complex 91–3 (1986); Mansfield, supra note 87, at 64. 123 Wesley Cohen et al., Links and Impacts: The Influence of Public Research on Industrial R&D, 48 MgMt. sCI. 1 (2002). 124 See Kenney, supra note 122, at 104. 125 Albert N. Link et al., An Empirical Analysis of the Propensity of Academics to Engage in Informal University Technology Transfer, 16 Indus. & Corp. Change 641, 642 (2007). 126 Agrawal, Engaging the Inventor, supra note 39, at 16.

Tacit knowledge and university-industry technology transfer 227 ongoing interactions between the faculty inventor and the licensee during commercialization.127 Similarly, empirical research reveals that ineffective or nonexistent involvement by faculty inventors contributed to 18% of commercialization failures for technologies licensed from universities.128 In the technology transfer context, surveys indicate that consulting is the most frequently utilized mechanism for transferring inventor knowledge to a licensee firm.129 For their part, faculty members may be motivated by financial incentives, technical curiosity or a desire to help commercialize their inventions to work as consultants with companies licensing their inventions.130 The importance of faculty involvement in commercializing academic technologies is evident in the early biotechnology industry.131 The participation of “star” bioscientists significantly and positively influenced the number of products in development, the number of products released, and employment growth.132 The centrality of academic scientists to the early biotechnology industry illustrates that “university-firm technology transfer for breakthrough discoveries generally involves detectable joint research between top professors and firms that they own or are compensated by.”133 Beyond serving as consultants for finite engagements, academic inventors also serve as permanent technical advisers and may become directors of companies licensing university patents.134 In the 1970s Herbert Boyer of UCSF and Stanley Cohen of Stanford invented and patented the basic techniques of recombinant DNA technology. Stanford licensed this technology to numerous biotech companies, including Genentech and Cetus; at the time, Boyer served on Genentech’s board of directors and Cohen was a scientific advisor to Cetus.135 While hiring academic scientists serves several functions—such

127

Jeannette Colyvas et al., How Do University Inventions Get into Practice?, 48 MgMt. sCI. 61, 64 (2002) (finding that faculty involvement was important to seven out of eleven cases studies of technology transfer). 128 Thursby & Thursby, supra note 64, at 167. 129 Id. at 170; Nicholas S. Argyres & Julia Porter Liebeskind, Privatizing the Intellectual Commons: Universities and the Commercialization of Biotechnology, 35 J. eCon. behav. & org. 427, 450 (1998); see Thomas Hellmann, The Role of Patents for Bridging the Science to Market Gap, 63 J. eCon. behav. & org. 624, 627 (2007); see also Howells, supra note 18, at 95. 130 Faculty members may be particularly motivated to consult on projects related to their patents to supplement rather limited patent-related royalties. See Bercovitz & Feldmann, supra note 64, at 179 (noting the relatively small after-tax returns on royalties compared to consulting). 131 See Pisano, Can Science Be a Business?, supra note 17, at 2. 132 Lynne G. Zucker & Michael R. Darby, Star Scientists and Institutional Transformation: Patterns of Invention and Innovation in the Formation of the Biotechnology Industry, 93 proC. nat’l aCad. sCI. u.s. aM. 12,709, 12,712 (1996). 133 Zucker et al., Commercializing Knowledge, supra note 12, at 149; see Zucker et al., Intellectual Human Capital, supra note 54; Zucker & Darby, supra note 132, at 12,709. 134 Siegel et al., supra note 75, at 118; see Kenney, supra note 122, at 149–54 (discussing biotechnology Scientific Advisory Boards). 135 Tamar Lewin, The Patent Race in Gene-Splicing, n.y. tIMes, Aug. 29, 1982, available at http:// www.nytimes.com/1982/08/29/business/the-patent-race-in-gene-splicing.html (last visited Oct. 17, 2019); see Sally Smith Hughes, Making Dollars Out of DNA: The First Major Patent in Biotechnology and the Commercialization of Molecular Biology, 1974–1980, 92 IsIs 541, 562 (2001) (“What raised concern in Cohen’s case was the fact that he was an inventor on a Stanford patent application and at the same time a paid consultant for a company seeking a license on the invention being patented.”); Kenney, supra note 122, at 94 (“Genentech was presumably started on Boyer’s consulting time while he was a professor at the UCSF Medical Center.”).

228 Research handbook on intellectual property and technology transfer as attracting venture capital and shareholder investment136—it also facilitates the transfer of tacit knowledge. The importance of engaging faculty inventors in commercializing university discoveries is not unique to the biotechnology industry and extends to other technical fields, like semiconductors.137 In sum, the participation of faculty scientists plays an important role in bringing academic discoveries to market.138 C.

Sponsored Research

Another mechanism by which tacit technical knowledge flows from academic to industrial scientists is sponsored research.139 In such arrangements, companies fund research at universities and exercise varying levels of control over the research and its outputs, which may include patented inventions. For example, as far back as 1974, Monsanto entered into a 12-year agreement to provide $23 million in research and endowment support to Harvard Medical School in exchange for the rights to any patents that arose from research on a substance related to cancer growth.140 Declining federal support has led universities to rely more heavily on sponsored research, which has also led them to “focus more on knowledge transfer issues that provide value to the sponsoring firm.”141 Sponsored research serves several objectives, some of which directly implicate tacit knowledge. Close ties between a corporate sponsor and a university can allow industry personnel to become aware of commercially promising research even before a university begins marketing it.142 More generally, funding university research exposes a corporate sponsor to the latest science, which enhances its absorptive capacity.143 Furthermore, it allows companies to exploit the technical expertise and infrastructure of universities to pursue research with potential commercial applications. Relatedly, as with the Monsanto-Harvard Medical School deal, sponsors may obtain patent rights on valuable research outputs.144 Along the way, sponsored research often involves joint research between academic and corporate scientists, thus facilitating tacit knowledge transfer.145 Sponsored research is particularly well-suited for providing access to university inventions that are too embryonic to patent and license.146 In sum, sponsored

136 In the early biotech industry, IPO prospectuses often highlighted the participation of notable academic scientists. Zucker et al., Commercializing Knowledge, supra note 12, at 143. 137 Zucker & Darby, supra note 132, at 12,715. 138 Id. 139 Bercovitz & Feldmann, supra note 64, at 177. 140 Barbara J. Culliton, Harvard and Monsanto: The $23-Million Alliance, 4280 sCIenCe 759, 759 (1977). 141 Michael D. Santoro & Paul E. Bierly, III, Facilitators of Knowledge Transfer in University-Industry Collaborations: A Knowledge-Based Perspective, 53 Ieee transaCtIons engIneerIng MgMt. 495, 495 (2006). 142 Colyvas et al., supra note 127, at 66–67. 143 See supra note 108 and accompanying text. 144 See Kenney, supra note 122, at 58. 145 Iain Cockburn & Rebecca Henderson, Absorptive Capacity, Coauthoring Behavior and the Organization of Research in Drug Discovery, 46 J. Indus. eCon. 157, 166 (1998); Link et al., supra note 125, at 645; David Blumenthal, Academic-Industrial Relationships in the Life Sciences, 349 new eng. J. Med. 2452, 2454 (2003). 146 Thursby & Thursby, supra note 64, at 163.

Tacit knowledge and university-industry technology transfer 229 research allows corporations to maintain relationships with faculty inventors whose inventions they are licensing or seek to license.147 Sponsored research often blurs the institutional boundaries between the corporate sponsor and the university it funds.148 For example, when Hoechst sponsored research at Massachusetts General Hospital, one commentator observed that scientists at the hospital “might well come to believe that they work for Hoechst.”149 Additionally, sponsored research may involve embedding corporate scientists in university laboratories.150 Sponsored research has led to new institutions that bridge academic and industrial partners. For instance, the Whitehead Institute for Biomedical Research at MIT, which is financed by biotech venture capitalist and major Revlon shareholder Edwin Whitehead,151 reflects “an attempt to create an inter-penetrating system of public and private research within a university setting.”152 Further blurring institutional boundaries, sponsored research tends to lead to more sponsored research. At UC Irvine, corporate research sponsors can receive not just exclusive licenses to patented inventions but also the right to sponsor further research in related areas.153 Additionally, about a third of university licenses require compensation by the licensee in the form of sponsored research,154 and licensing for sponsored research represents the predominant licensing strategy for 11% of TTOs.155 Sponsored research helps maintain interactions—and, in some cases, institutional meshing—between universities and corporate entities, which promotes tacit knowledge exchange. D.

Proof of Concept Centers and Incubators

In addition to direct contacts between university inventors and licensees, universities often create more enduring institutional connections with commercial entities to serve several functions, including transferring tacit knowledge. For example, many universities have established proof of concept centers, which “are becoming an important vehicle for advancing technology commercialization.”156 These centers nurture commercialization relatively early in the development chain, earlier than traditional business incubators.157 Proof of concept centers provide

147

Id. at 175. Cf. Etzkowitz & Viale, supra note 2, at 602 (discussing permeable boundaries between academic and private entities). 149 Nicholas Wade, The Science Business: Report of the Twentieth Century Fund Task Force on the Commercialization of Scientific Research 48 (1984). 150 Murray, supra note 64, at 654. 151 Argyres & Liebeskind, supra note 129, at 448. 152 Id. 153 Id. at 444. 154 Jensen & Thursby, supra note 65, at 246. 155 Gideon D. Markman et al., Entrepreneurship and University-based Technology Transfer, 20 J. bus. venturIng 241, 251 (2005). 156 Christopher S. Hayter & Albert N. Link, On the Economic Impact of University Proof of Concept Centers, 40 J. teCh. transfer 178, 179 (2015) [hereinafter Hayter & Link, Economic Impact]. 157 Bob Tedeschi, The Idea Incubator Goes to Campus, n.y. tIMes, June 27, 2010, at BU1; see Ewing Marion Kauffman Foundation, Proof of Concept Centers: Accelerating the Commercialization of University Innovation (2008) (discussing the Deshpande Center at MIT and the von Liebig Center at the University of California San Diego). 148

230 Research handbook on intellectual property and technology transfer a range of resources, services and networks to help university startups.158 While the primary function of proof of concept centers is to provide funding, they also promote knowledge exchange between university innovators and industry via various mentors.159 For example, the von Liebig Center at UCSD provides technical advisors to participants, and the Deshpande Center at MIT provides volunteer advisors to serve as catalysts with industry.160 Interestingly, in addition to helping to transfer tacit technical knowledge to industry, such proof of concept centers allow advisors to transfer tacit business knowledge to faculty inventors. By 2012, over 30 proof of concept centers had been established in the United States.161 Further downstream in the development chain, universities have also created incubators to facilitate the commercialization of academic technologies.162 Such entities “are property-based organizations with identifiable administrative centers focused on the mission of business acceleration through knowledge agglomeration and resource sharing.”163 They provide social, technological, organizational and managerial resources to help transform technology-based business ideas into sustainable enterprises.164 Many universities are establishing incubators to promote commercialization of academic technologies;165 in one study, 62% of universities surveyed report that they are establishing incubators and research parks.166 One of the functions of incubators is to promote interactions and working relationships between academic scientists and entrepreneurs,167 which can facilitate tacit knowledge transfer. Closely related to incubators are university research parks, agglomerations of resources and businesses that foster the creation of startups based on university technologies.168 Due to the institutional and often geographic proximity of research parks and universities, firms located in such parks have greater opportunities to obtain tacit knowledge from academic scientists.169 These parks “enhance the two-way flow of knowledge between firms and universities,”170 thus promoting commercialization.

158

Hayter & Link, Economic Impact, supra note 156, at 179. Christine A. Gulbranson & David B. Audretsch, Proof of Concept Centers: Accelerating the Commercialization of University Innovation, 33 J. teCh. transfer 249, 250 (2008). 160 Id. at 252–3; Hayter & Link, Economic Impact, supra note 156, at 179. 161 Christopher S. Hayter & Albert N. Link, University Proof of Concept Centers: Empowering Faculty to Capitalize on Their Research, Issues In sCI. & teCh., Winter 2005, at 35. 162 Phillip H. Pan et al., Science Parks and Incubators: Observations, Synthesis and Future Research, 20 J. bus. venturIng 165, 167–8 (2005). 163 Id., at 166. 164 Id. at 170–71; Rosa Grimaldi et al., 30 Years after Bayh-Dole, Reassessing Academic Entrepreneurship, 40 res. pol’y 1045, 1048 (2011). 165 See Annetine C. Gelijns & Samuel O. Thier, Medial Innovation and Institutional Interdependence, 287 JAMA 72, 75 (2002); Markman et al., supra note 155, at 244–5 (describing the Rensselaer Polytechnic Institute incubator); id. at 252 (noting that 62% of TTOs devoted significant resources to building business incubators); Tedeschi, supra note 157. 166 Markman et al., supra note 155, at 242. 167 Dante Di Gregorio & Scott Shane, Why Do Some Universities Generate More Start-Ups Than Others?, 32 res. pol’y 209, 213 (2003); Grimaldi et al., supra note 164, at 1048. 168 Albert N. Link & John T. Scott, The Economics of University Research Parks, 23 oxford rev. eCon. pol’y 661, 661 (2007). 169 Id. at 664. 170 Id. at 670. 159

Tacit knowledge and university-industry technology transfer 231 E.

University Spinoffs

Another mechanism to transfer tacit knowledge from universities to commercial entities is to integrate academic and commercial functions in a university spinoff. In the 1990s, Stanford faculty began starting companies, and Stanford granted a significant number of licenses to “university spinoffs or startups where Stanford inventors hold key executive positions or serve on scientific advisory boards.”171 In a common pattern, a faculty inventor will assign her patents to a university (based on her employment agreement), and the university will then license those patents to a startup founded by the faculty inventor. Several universities extend preferential treatment to university inventors seeking to license their own inventions to start companies.172 More broadly, “[u]niversity spin-offs have become a favored mechanism by which universities transfer technology to the commercial realm.”173 The organizational form of a university spinoff is well suited to transfer tacit knowledge. According to one observer, “[s]tart-ups internalize the full set of interactions between research and productive organizations that lead to the successful development of a commercial product; the full set includes many more things transacted other than patents,” including tacit knowledge.174 Due to the highly tacit nature of much cutting-edge scientific knowledge, university scientists are best situated to exploit it commercially, and they are perhaps the only persons capable of doing so.175 Indeed, one theoretical model posits that university spinouts established by faculty inventors will be the sole vehicle for commercializing inventions with a significant tacit dimension.176 In many spinoffs, academic expertise and commercial functions are integrated when a faculty inventor establishes a new company.177 Entrepreneurial university scientists who participate in commercialization help minimize gaps in tacit knowledge.178 Furthermore, empirical studies reveal that star academic scientists frequently played a critical role in establishing new biotechnology companies.179 Just as university spinoffs are well-positioned to exploit tacit knowledge, such knowledge is often critical to their success.180 An empirical examination of spinoffs from a leading technical

171

Jeannette A. Colyvas & Walter W. Powell, Roads to Institutionalization: The Remaking of Boundaries between Public and Private Science, 27 res. org. behav. 305, 343 (2018); see also Etzkowitz & Viale, supra note 2, at 597. 172 Grimaldi et al., supra note 164, at 1049. For instance, the University of North Carolina at Chapel Hill has created an Express Licensing Agreement that expedites licensing to faculty inventors seeking to commercialize their inventions. Id. 173 Bercovitz & Feldmann, supra note 64, at 179; see also Etzkowitz & Viale, supra note 2, at 600 (describing the emergence of the “entrepreneurial university” that generates spinoffs). 174 Walter D. Valdivia, University Start-Ups: Critical for Improving Technology Transfer 18 (2013). 175 Di Gregorio & Shane, supra note 167, at 212; see Robert A. Lowe, Who Develops a University Invention? The Impact of Tacit Knowledge and Licensing Policies, 31 J. teCh. trans. 420 (2006). 176 Lowe, supra note 175, at 415. 177 Cf. Andy Lockett et al., Technology Transfer and Universities’ Spin-Out Strategies, 20 sMall bus. eCon. 185, 186, 196 (2003). 178 Etzkowitz & Viale, supra note 2, at 600. 179 Zucker & Darby, supra note 132, at 12,713; Powell, supra note 116, at 200. According to one observer, “The pervasive role of professors in managing and directing [biotechnology] startups is unique in the annals of business history.” Kenney, supra note 122, at 4; see id. at 94 (“All of the earliest genetic engineering companies were founded by professors.”); see id. at 98 (listing the involvement of founding professors with various companies). 180 Knockaert et al., supra note 43, at 780.

232 Research handbook on intellectual property and technology transfer university in Belgium showed that “effective tacit knowledge transfer was crucial” for the performance of science-based entrepreneurial firms.181 For spinoffs where the faculty inventor is not the principal entrepreneur, the inventor’s participation in commercialization is often critical to success.182 Empirical research has found that “[s]uccessful knowledge transfer is more likely if the original scientists who worked on developing the technology are also involved in the venture.”183 Furthermore, it found that a critical mass of tacit knowledge was necessary, which included employing the majority of initial researchers.184

IV.

RETHEORIZING UNIVERSITY LICENSING AND TECHNOLOGY TRANSFER

The centrality of tacit knowledge to effective commercialization of academic technologies raises a host of implications for university-industry technology transfer. First, it underscores that technology transfer is much more than simply patenting and licensing university inventions.185 As this Chapter has demonstrated, a rich set of human and institutional connections is often necessary in parallel to licensing to effectively transfer and commercialize a patented academic invention. Relatedly, the centrality of tacit knowledge also calls into question traditional metrics for measuring the effectiveness of technology transfer. The so-called “out-the-door” criterion focuses on whether a technology has been converted to some kind of transfer mechanism, such as a license, which another party has acquired.186 This criterion, however, does not inquire into what the receiving entity does with the technology, and in the absence of technology-related tacit knowledge, it may not be able to do much at all. Second, the importance of tacit knowledge flows helps explain a distinctive characteristic of university-industry technology transfer: its highly localized nature. We have seen that tacit knowledge transmission often requires direct interpersonal interaction; not surprisingly, tacit knowledge tends to be geographically concentrated around the location of a discovery.187 Similarly, empirical studies illustrate the importance of proximity in capturing knowledge spillovers from universities,188 and commercialization of university inventions tends to be localized around the place of invention.189 Even though universities can license their patents nationally (or globally), university-based entrepreneurship tends to be decidedly local.190 Of course, several factors contribute to this geographic proximity, such as connections between TTO personnel and local businesses as well as universities’ commitments to foster local economic development. Nonetheless, the need to transfer tacit knowledge and interact directly

181

Id. at 784. Lockett et al., supra note 177, at 187. 183 Knockaert et al., supra note 43, at 788. 184 Id. at 789. 185 Valdivia, supra note 174, at 5 (“Technology transfer is thus not a matter of patents alone; rather it is the complex work that takes place at the interface of research and productive organizations.”). 186 Bozeman et al., supra note 105, at 37 187 Zucker et al., Commercializing Knowledge, supra note 12, at 142. 188 Adam B. Jaffe et al., Geographical Localization of Knowledge Spillovers as Evidenced by Patent Citations, Q.J. eCon. 577 (1993). 189 Agrawal, Knowledge Transfer, supra note 64, at 301. 190 Bercovitz & Feldmann, supra note 64, at 179. 182

Tacit knowledge and university-industry technology transfer 233 with faculty inventors plays a role as well: “When university-based scientists are actively involved in knowledge transfer, their knowledge is more easily tapped if the firm is located in the same region as the scientist.”191 Third, the importance of tacit knowledge to practicing university inventions challenges traditional rationales for patenting such technologies in the first place. This Chapter has focused on tacit knowledge transfers in parallel to formal patenting and licensing. The central narrative of patents justifies exclusive rights to protect against costless appropriation of an otherwise nonrival, nonexcludable resource.192 However, tacit knowledge is naturally excludable,193 and the highly tacit nature of some embryonic university inventions lessens the need to patent them to safeguard against unauthorized appropriation. As Dan Burk observes, “codified knowledge, having been separated from human memory, may be more readily moved about, but the uncodified knowledge that supports this codified knowledge moves only with the humans who carry it, or sometimes not at all.”194 Ultimately, the centrality of tacit knowledge calls into question the rationale for patenting certain cutting-edge university inventions. Fourth, while the need to transfer tacit knowledge fuels greater connections between individual scientists and companies, it also motivates broader institutional integration between universities and companies, thus blurring the boundaries between these historically distinct institutions. In understanding such organizational integration, the knowledge-based theory of the firm provides useful insight. In classic articulations of the theory of the firm, integrated firms emerge as a solution to transaction costs and opportunism that plague contractual relationships between independent entities.195 In the context of technology transactions, patents can ameliorate these contractual hazards, thus facilitating contracting between separate parties.196 However, patents do not disclose and licenses do not convey critical tacit knowledge necessary to commercialize many patented technologies. In the realm of university-industry technology transfer, this knowledge deficit has motivated a host of institutional connections spanning sponsored research agreements, proof of concept centers, incubators and spinoffs. Such integration implicates the knowledge-based theory of the firm, which characterizes integrated organizations not as solutions to contractual hazards and opportunistic behavior197 but as solutions for the challenges of accessing, collecting and exploiting knowledge.198 After all, “[c]ommunication between R&D and manufacturing flows more easily when the organizations share common experiences and information systems.”199 Ultimately, organizational integration between academic and commercial entities may be the most effective means of transferring tacit knowledge and promoting commercialization. This finding, moreover, is rather startling given the stark normative and cultural differences between those types of institutions.200

191

Audretsch & Stephan, supra note 84, at 644. Lee, Transcending, supra note 98, at 1570. 193 Zucker et al., Commercializing Knowledge, supra note 12, at 141. 194 Burk, supra note 14, at 1017. 195 See, e.g., Benjamin Klein et al., Vertical Integration, Appropriable Rents, and the Competitive Contracting Process, 21 J.l. & eCon. 297 (1978). 196 Lee, Innovation and the Firm, supra note 23, at 1439–42. 197 Powell et al., supra note 93, at 117–18. 198 Knockaert et al., supra note 43, at 780; Powell et al., supra note 93, at 118. 199 Pisano et al., supra note 20, at 199 (citation omitted). 200 Lee, Transcending, supra note 98, at 1511. 192

234 Research handbook on intellectual property and technology transfer Fifth and relatedly, tighter connections between academic and commercial entities raise normative concerns about compromising traditional academic norms. The naturally excludable character of tacit knowledge enhances the commercial value of faculty inventors (and their tacit knowledge) and may encourage faculty members to pursue more commercially lucrative rather than scientifically meritorious lines of research. In biotechnology, for example, increases in the quality and commercial relevance of the work of star scientists are correlated with an increased probability of conducting joint research with or joining a commercial entity.201 Close collaborations with companies raise concerns of “increased opportunity for conflicts of interest, redirection of research, less openness in sharing of scientific discovery, and a greater emphasis on applied rather than basic research.”202 At a broader institutional level, organizational integration between universities and profit-maximizing firms calls into question universities’ commitment to disinterested academic inquiry and basic research with no immediate commercial applications. While research suggests that traditional scientific norms have not eroded and that universities are not performing less basic research,203 greater proximity of academic and commercial entities may change the normative identity of scientists and universities over time.

V.

CONCLUSION

This Chapter has explored the importance of tacit knowledge in university-industry technology transfer. Discussions of technology transfer often focus on patenting and licensing, which is certainly important to commercializing academic technologies. However, this Chapter has highlighted the significance of noncodified, experiential, tacit knowledge on the part of university inventors in transferring technologies to the private sector. Patents do not disclose tacit knowledge, which can run the gamut from latent knowledge that is codifiable but not codified to pure tacit knowledge, which is not capable of codification. The tacit dimension of an invention depends on the nature and complexity of a technology, and it tends to decrease over time as personal knowledge diffuses and becomes more accessible. Tacit knowledge is valuable not only for practicing some basic invention but also for adapting it to industrial use and commercial production. Transferring such knowledge is very difficult and often requires direct interpersonal interaction between a faculty inventor and a commercial licensee. Accordingly, this Chapter has also explored several relational and institutional mechanisms for transferring tacit knowledge. Networks, consulting agreements and corporate positions, sponsored research, proof of concept centers and incubators, and university spinoffs all facilitate greater engagement between faculty inventors and commercial partners, which promotes tacit knowledge exchange. These observations simply underscore the reality that “[u]niversity patents represent only one mechanism by which academic research results can be transferred 201 Lynn G. Zucker et al., Labor Mobility from Academe to Commerce, 20 J. labor. eCon. 629 (2001). 202 Wendy H. Schacht, The Bayh-Dole Act: Selected Issues in Patent Policy and Commercialization of Technology, 13 Cong. res. serv. (2012); see Perkmann et al., supra note 112, at 428. 203 Grimaldi et al., supra note 164, at 1046; see also Kira R. Fabrizio & Alberto Di Minin, Commercializing the Laboratory: Faculty Patenting and the Open Science Environment, 37 res. pol’y 914, 916 (2008) (finding a positive correlation between patenting and publishing that declines with the increasing number of cumulative patents attributed to a faculty inventor).

Tacit knowledge and university-industry technology transfer 235 to the market place.”204 Furthermore, the importance of direct personal interactions with inventors helps explain the geographically localized nature of university-industry technology transfer. Greater engagement between faculty members and university with commercial partners can accelerate tacit knowledge exchange and technology transfer, but it raises concerns over altering traditional academic research norms and priorities. Ultimately, conveying tacit knowledge represents a central challenge of university-industry technology transfer.

204

Grimaldi et al., supra note 164, at 1047.

11. Technology transfer and the public good Brian L Frye and Christopher J Ryan, Jr

I.

INTRODUCTION

Something is rotten in university patent policy. Universities and patents are both supposed to promote the public good. But sometimes, patents may encourage universities to pursue goals inconsistent with the public good. In 1980, the Bayh-Dole Act amended the Patent Act to allow universities to patent inventions and discoveries funded by federal grants.1 In response, universities began creating “technology transfer” offices in order to help researchers file patent applications and license university patents. Some university technology transfer offices (“TTOs”) are successful, generating substantial revenue for the university, but most are not and operate at a loss.2 Even more troubling, university patents and TTOs may be economically inefficient. In theory, universities should patent inventions and discoveries that are socially beneficial and should fund socially valuable research, irrespective of its likelihood of generating patentable inventions and discoveries. But currently, universities have an incentive to patent all of the patentable research they generate, even if doing so reduces the public benefit generated by that research. They also have an incentive to patent inventions and discoveries that have no commercial value, because the patent may still have litigation value. And they may even have an incentive to invest more heavily in research that is likely to generate patentable inventions and discoveries than research that is not, irrespective of the social value of the research. While universities should resist those incentives, the evidence suggests that many cannot. Maybe the government can help? Obviously, the government could largely eliminate the incentive for universities to patent research by repealing the Bayh-Dole Act and making most university research unpatentable. But sometimes, university patents may be justified. If a university generates a commercially valuable invention or discovery, why should a university not be able to claim some of the value of that invention or discovery from the private companies that ultimately commercialize it? After all, universities can use that additional revenue to fund more research and benefit the public in other ways. However, the government could make university patents more efficient by clarifying the standards for patentability and reducing the bias in favor of patentable research. While universities are hardly alone in filing some weak patent applications, reducing uncertainty about patentability would enable them to streamline their patenting efforts. In addition, the government could increase the efficiency of university patents by instructing public grantmakers to prioritize funding research that is unlikely to produce patentable inventions or discoveries.

1

Patent and Trademark Laws, 96 P.L. 517, 94 Stat. 3015 (1980). See David Mowery, et al., Ivory Tower and Industrial Innovation: University-Industry Technology Transfer before and after the Bayh-Dole Act (2015); see also, Joseph Friedman & Jonathan Silberman, University Technology Transfer: Do Incentives, Management, and Location Matter?, 28 J. teCh. transfer 17 (2003). 2

236

Technology transfer and the public good 237 Additionally, universities could better align their patent policies with their public purpose and mitigate the risk associated with investing in patents by pooling them and sharing the profits.

II.

UNIVERSITIES, PATENTS AND PUBLIC GOODS

While universities and patents are functionally unrelated, they share the goal of increasing social welfare by producing public goods. Universities are charitable organizations that rely primarily on altruism, and patents are property rights that rely primarily on profit. But both are intended to encourage innovation. Innovation is like a “public good,” because it is non-rival: the use of an innovation does not reduce the supply. But neoclassical economics predicts that “free riding” will cause “market failures” in such public goods.3 In other words, people will use innovations without paying the marginal cost of production. In theory, the government can use direct subsidies to solve market failures in public goods, including innovation. But information costs often prevent the government from solving market failures, causing “government failures,” because the government does not always know which innovations to subsidize. Universities use altruism to solve market and government failures in innovation. They receive government grants and charitable contributions, which they use to subsidize innovation. They solve market failures by relying on altruism, rather than profit. Universities encourage free riding on the innovations they produce, and they solve government failures by reducing transaction costs. Universities have more and better information than the government about which innovations are likely to have social value. By contrast, patents use profits to solve market and government failures in innovation.4 They solve market failures by preventing free riding: patents make innovation partially excludable, enabling patent owners to recover the cost of innovation by commercializing their inventions and discoveries. And they solve government failures by reducing transaction costs. Innovators have more and better information than the government about which innovations are likely to have commercial value. In other words, both universities and patents are intended to increase the efficiency of investment in innovation, but in different ways. Universities make public investment more efficient and patents make private investment more efficient. In theory, universities and patents not only overlap but also complement each other. Universities produce innovations that companies commercialize and also train the innovators whom companies hire. Patents encourage companies to collaborate with universities, fund their research and support their students. The problem is how to maximize mutual efficiency and avoid inefficient incentives.

3 See Kenneth J. Arrow, “Economic Welfare and the Allocation of Resources for Invention” in Readings in Industrial Economics, 219 (1972); Francis M. Bator, The Anatomy of Market Failure, 72 Q. J. eCon. 351, 377 (1958). 4 See Richard Posner, Economic Analysis of Law § 3.3, 48–50 (8th ed., 2011).

238 Research handbook on intellectual property and technology transfer

III.

A BRIEF HISTORY OF UNIVERSITY PATENTS

Historically, universities rarely filed patent applications or pursued research likely to produce patentable inventions or discoveries.5 Prior to the Second World War, only a handful of universities applied for patents. Universities focused on producing basic research, and businesses focused on developing practical applications. However, in the second half of the 20th Century, universities began to engage more robustly in research and development that could produce a patent, in large part because of the rights that were granted to them under the Bayh-Dole Act. A.

Government Funded University Research and Government Patents

During the Second World War, the government mobilized university researchers as part of the war effort, culminating in programs like the Manhattan Project to develop the atomic bomb. After the war, the government continued to invest in university research, initially through the Department of Defense, but eventually through an alphabet soup of federal agencies.6 The government viewed science as a new “endless frontier,” and university researchers as the pioneers who would populate it, with government grants as their grubstakes, and government patents as the result.7 It was the beginning of a new era of government investment in university research, as the government became even more involved in university research, and acquired an ever-increasing number of patents.8 During the Cold War, the government retained the right to patent any inventions or discoveries generated by government-funded research, with limited exceptions. So, if a government agency funded university research, the agency typically owned any patent rights in the results of that research.9 In theory, the government would license those patents to private companies, which would in turn commercialize the innovations they embodied. However, university research funded by the government during the Cold War was rarely commercialized, and some policymakers began to wonder whether it was an efficient use of public funds.10 While the government accumulated more than 28,000 patents during the 1960s,

5 American University Patent Policies: A Brief History, aMerICan assoCIatIon of unIversIty professors, available at https://www.aaup.org/sites/default/files/ files/ShortHistory.pdf (last visited Sept. 23, 2018). 6 See Stuart W. Leslie, The Cold War and American Science (1993); see also, aMerICan assoCIatIon of unIversIty professors, supra note 5. 7 Vannevar Bush, Science: The Endless Frontier (1945), available at https://doi.org/10.1002/sce .3730290419 (last visited Oct. 17, 2019). 8 Vanessa Bell, The State Giveth and the State Taketh Away: Patent Rights Under the Bayh-Dole Act, 24 S. Cal. InterdIsC. l.J. 491, 503 (2015). 9 If a university received research funding from the Department of Health, Education, and Welfare, it could retain the patent rights in its research, under certain circumstances. Gary Pulsinelli, Share and Share Alike: Increasing Access to Government-Funded Inventions Under the Bayh-Dole Act, 7 MInn. J.l. sCI. & teCh. 393, 400 (2006). The Department of Defense (DoD) also allowed patent recipients to retain title so long as they had an established commercial position in the field. Universities were able to make use of an exception to the rule under “special situations” that let them retain title without approval. Id. 10 For example, an early 1960s study found low usage rates of government funded inventions. Memorandum and Statement of Government Patent Policy, 28 Fed. Reg. 10,943 (Oct. 10, 1963).

Technology transfer and the public good 239 it only licensed about 5 percent of them to private parties for commercial development.11 Of course, the low rate of commercial utilization reflected in part the fact that many government patents covered military technology and were classified as a matter of national security.12 It is unclear what percentage of unclassified government patents were licensed for commercial use and what percentage facilitated other socially beneficial innovation. Congress created the Commission on Government Procurement in 1969, to study the licensing of government patents and encourage their commercialization. The Commission suggested the government adopt a uniform policy of allowing research grant recipients to retain patent rights, with certain exceptions, on the ground that grant recipients would pursue commercialization more efficiently than the government. Specifically, it recommended that government-funded researchers retain patent rights, subject to reserved government “march-in” rights.13 The Commission’s report initiated a shift in government policy that eventually led to the privatization of patents in government-funded research. In 1976, President Ford created the Office of Science and Technology Policy in order to encourage universities and private businesses to cooperate on research.14 But the Office was largely ineffective, at least in part because the government retained ownership of any patents derived from government-funded research. Accordingly, universities had little incentive to develop commercial applications for the research they produced. Eventually, Congress decided to solve the problem by getting out of the patent business entirely. In 1980, Congress passed the Bayh-Dole Act, which amended the Patent Act to provide that universities could patent inventions and discoveries derived from government-funded research and license those patents to private companies.15 The Bayh-Dole Act was intended to promote innovation by encouraging universities to invest in patentable research that private companies could commercialize.16 The Bayh-Dole Act was supposed to make government grants more efficient by encouraging universities to invest in the production of patentable innovations and license those patents to private companies for commercial use. In theory, universities would start licensing patents that had previously lain fallow and invest in the production of ever more commercially valuable innovations. But in practice, results were mixed. While universities certainly acquired more patents, many had little or no commercial value.17

11

H.R. REP. 109-409, 2. This figure is considered debatable, as it includes patents held by the federal government as matters of national defense that would not reach commercialization even under the Bayh-Dole Act. See Robert M. Yeh, The Public Paid for the Invention: Who Owns It?, 27 berkeley teCh. l.J. 453, 471 (2012). 13 Pulsinelli, supra note 9. 14 Dov Greenbaum, Academia to Industry Technology Transfer: An Alternative to the Bayh-Dole System for Both Developed and Developing Nations, 19 fordhaM Intell. prop. MedIa & ent. l.J. 311, 350 (2009). 15 Patent and Trademark Laws, 96 P.L. 517, 94 Stat. 3015 (1980); 35 U.S.C. § 200–212 (2012). 16 Among other things, the Bayh-Dole Act replaced 26 distinct federal policies with one uniform policy. See Llewellyn Joseph Gibbons, Tech Transfer: Everything (Patent) Is Never Quite Enough, 48 u. louIsvIlle l. rev. 843, 848 (2010). 17 See U.S. Patent statistics Chart by Calendar Years 1963–2015, uspto, 2016, available at https:// www.uspto.gov/web/offices/ac/ido/oeip/taf/us_stat.htm (last visited Oct. 17, 2019). 12

240 Research handbook on intellectual property and technology transfer B.

The Bayh-Dole Act and University Patents

For better or worse, when Congress passed the Bayh-Dole Act in 1980, it sparked a dramatic increase in university patent activity.18 In 1979, the Patent Office granted 267 university patents, but in 1980, it granted 394. And the number quickly rose. In 1985, it granted 594, and in 1989, it granted 1,261. By 2000, it was granting about 3,000 per year, and today, it grants almost 5,000 per year.19 Additionally, between 1990 and 2000, invention disclosures by university TTOs increased by 79 percent, patent applications increased by 253 percent, patent grants increased by 131 percent, and start-up companies evolving out of university research increased by 92 percent— all indications of a robust response by universities to the incentives to universities to enter the patent sector.20 While the Bayh-Dole Act clearly achieved its goal of encouraging universities to pursue patents, it also had unintended consequences. The Bayh-Dole Act was a response to the concern that universities were underinvesting in research with commercial potential. It

Figure 11.1

University utility patents granted 1980–2012

Source: USPTO, US Patent Count and Expenditures, https://developer.uspto.gov/sites/default/files/viz/university -patent-count-expenditures.xlsx

18 Christopher J. Ryan, Jr. & Brian L. Frye, An Empirical Analysis of University Patent Activity, 7 N.y.u. J. Intel. prop., MedIa, & ent. l.J. 51 (2018). 19 University Patents and Expenditures, uspto, available at https://developer.uspto.gov/ visualization/university-patent-count-expenditures (last visited Aug. 31, 2018); see also, Ass’n of Univ. Tech. Managers, AUTM U.S. Licensing Survey: FY 2004, at 2 (2004). 20 Greenbaum, supra note 14.

Technology transfer and the public good 241 relied on the assumption that patentable research is more likely to be commercially valuable research. But that is not always the case, especially because university researchers have conflicting incentives about what research to pursue and do not know what kinds of patents will be most valuable to commercial firms. As a consequence, while universities have pursued and obtained more patents, universities have not necessarily received commercially valuable patents. While a few university patents are extremely valuable, the vast majority are effectively worthless. Indeed, university patents have been likened to lottery tickets: not an investment but a gamble.21 Moreover, many scholars and policymakers are concerned about the effect of the Bayh-Dole Act on consumers. Before the Bayh-Dole Act, most publicly funded university research was in the public domain and could be used by anyone to produce commercial products. But the effect of the Bayh-Dole Act is to make the public pay for innovation twice: once, when the government funds university research that generates patents, and again, when universities license those patents to commercial firms, which use them to charge monopoly prices on the resulting products. This is especially troubling because universities are charitable organizations, created for the purpose of promoting the public benefit.22 Why should charitable organizations use public funds to obtain patents that ultimately benefit private firms?23 A cynic might observe that the practical effect of the Bayh-Dole Act was to enable private firms to offload some of their research and development costs onto universities. It is not obvious that this generated much public benefit. Any efficiency gains were enjoyed primarily by private firms, which could license commercially valuable university patents, without paying the full cost of the research required to generate the patent. C.

Public Investment

Since the Second World War, the government has invested enormous amounts of money in university research. For most of that time, the government funded the overwhelming majority of university research. But that is changing, as private organizations play an increasingly important role. The government continues to provide an enormous amount of funding for university research. In 2017, National Institutes for Health (NIH) awarded more than $18.7 billion in university research grants.24 Similarly, the National Science Foundation (NSF) awarded more than $5.1 billion in university research grants.25 However, while the government is still the largest source of university research grants, it no longer provides the majority of research funding. Throughout the 1960s, the federal govern-

21

Id. Jacob H. Rooksby, University Initiation of Patent Infringement Litigation, 10 John Marshall rev. Intell. prop. l. 623, 634 (2011). 23 Brian K. Krumm, University Technology Transfer – Profit Centers or Black Holes: Moving Toward A More Productive University Innovation Ecosystem Policy, 14 nw. J. teCh. & Intell. prop. 171, 184 (2016). 24 NIH Awards by Location & Organization, nIh report, available at https://report.nih.gov/award/ index.cfm (last updated June 25, 2018). 25 nat’l sCI. fndn., Award Summary: by Top Institutions, budget InforMatIon systeM, available at https://dellweb.bfa.nsf.gov/Top50Inst2/default.asp (last visited June 27, 2018). 22

242 Research handbook on intellectual property and technology transfer ment funded about 70 percent of university basic research.26 But the percentage of university research funding derived from government grants gradually declined over time. In 2015, the government funded only 44 percent of the $86 billion spent on basic research.27 For the first time since the Second World War, the government funded only a plurality of university research, and the majority of the funding came from an assortment of philanthropic and private sources.28 Notably, even as federal funding declined in absolute and relative terms, total university spending on basic research increased.29 Universities replaced and augmented federal funding with funding from other sources. Additionally, universities created TTOs to patent and license potentially commercial applications of the results of basic research. D.

The Gamble of Patents

For some universities, wild success resulting from university-sponsored research produced patents that generated windfall revenues. A paradigmatic example is Northwestern University’s patent on what became the drug Lyrica, a nerve-pain treatment for fibromyalgia—among other conditions—which has generated over $1.3 billion in licensing revenues for the university.30 But the most successful example of the blockbuster patent is Gatorade, which originated in 1965, by the request of the coaching staff of the University of Florida football team that researchers at the university’s medical school look into developing a drink to help athletes replace body fluid lost during exercise. The team went from mediocre to the Sugar Bowl in the year following its first use, and its popularity only grew in the intervening years. Since 1973, it has earned the University of Florida more than $281 million. The University has been able to use the money from Gatorade to fund various campus research buildings and provide seed money for various other projects totaling over $250 million to research.31 However, because this invention occurred before the passage of the Bayh-Dole Act, and the principal inventor of Gatorade had been funded by the federal government for research involving salt and water metabolism between 1962 and 1967, the federal government originally

26

Basic research is “activity aimed at acquiring new knowledge or understanding without specific immediate application or use.” Basic research is differentiated from applied research, which is designed to solve a direct problem or provide a specific commercial application: Data Check: U.S. Government Share of Basic Research Funding Falls below 50%, sCI. Mag., Mar. 9, 2017, 1:15 PM, available at http://www.sciencemag.org/news/2017/03/data-check-us-government-share-basic-research-funding -falls-below-50 (last visited Oct. 17, 2019). 27 Id. 28 Id. 29 See, id.; see also University Patent Count & Expenditures, uspto, available at https://developer .uspto.gov/visualization/university-patent-count-expenditures (last visited June 26, 2018). 30 Dave Merrill, Blacki Migliozzi, & Susan Decker, Billions at Stake in University Patent Fights, blooMberg, May 24, 2016, available at https://www.bloomberg.com/graphics/2016-university-patents/ (last visited Oct. 17, 2019). 31 April F. Lacey, UF Celebrates 50 Years of Gatorade, uf news, Sept. 28, 2015, available at http://news.ufl.edu/articles/2015/09/uf-celebrates-50-years-of-gatorade.php (last visited Oct. 17, 2019). However, it should be noted that a significant portion of this licensing revenue is generated by the “Gatorade” trademark, rather than the patent. See id.

Technology transfer and the public good 243 sought to take away these benefits accruing to the university and the inventors of Gatorade.32 The university also tried to acquire full rights to the invention, but the principal inventor never signed the standard licensing agreement, which would have granted them rights in the invention; thus, the university eventually settled for 20 percent of the royalties.33 This illustrative example highlights the complexities and qualifications of successful patents launched through university research and development. While commercially successful, the public benefits that accrued from this invention rest almost exclusively with the university and, apart from access to the product, not the larger public. Moreover, the university had to litigate with the primary inventor and the licensee of the patent to even receive the benefits of the invention it currently enjoys, despite providing the inventors with jobs, labs and research assistance in the development of the invention. This example should also be viewed as the exception and not the rule. University research and development has, of late, become more of a speculative enterprise than a sure bet.34 This example also serves to underscore another point. The success of Gatorade depended in large part on the licensees of the patent, Stokely-Van Camp, and later Quaker Oats and PepsiCo, all of which successfully produced and marketed the invention to a public audience. While some universities have successfully brought their inventions to market for public consumption,35 many university inventions never realize this goal, because universities do not have the capacity to manufacture the inventions and must rely on intermediaries to commercialize the inventions. This reliance results in a reality in which many university TTOs find themselves: most universities do not earn enough from licensing revenue to cover the cost of their TTOs.36 This fact further strains the notion that higher education institutions, as recipients of public financial support, are serving a public interest through their research and development. Additionally, many university-produced and university-held patents lack practical applications that would make them commercially successful. Current estimates put commercially licensed patents at about 5 percent of the typical university patent portfolio.37 This significant under-commercialization of the average American university patent portfolio has created a cottage industry for patent aggregators in which non-practicing entities—like Intellectual

32 David E. Greenbaum, U.S. Sues Gatorade for Its Profits, Saying Grant Financed Developer, n.y. tIMes, Aug. 12, 1971, at 21. 33 Lacey, supra note 31. 34 See Scott Andes, Technology Transfer 2.0: Finding Economic Value in University R&D, brookIngs, June 7, 2016, available at https://www.brookings.edu/blog/metropolitan-revolution/2016/ 06/07/technology-transfer-2-0-finding-economic-value-in-university-rd/ (last visited Oct. 17, 2019); Walter Valdivia, University Start-ups: Critical for Improving Technology Transfer, brookIngs, November 20, 2013, available at https://www.brookings.edu/research/university-start-ups-critical-for -improving-technology-transfer/ (last visited Oct. 17, 2019). 35 See, e.g., Chris Nicholson, Maximizing the ROI of Intellectual Property, u. bus., Sept. 29, 2014, available at https://www.universitybusiness.com/article/maximizing-roi-intellectual-property. 36 Andes, supra note 34. 37 Heidi Ledford, Universities Struggle to Make Patents Pay: Surfeit of Unlicensed Intellectual Property Pushes Research Institutions into Unseemly Partnerships, nature, Sept. 24, 2013, available at https://www.nature.com/news/universities-struggle-to-make-patents-pay-1.13811 (last visited Oct. 17, 2019).

244 Research handbook on intellectual property and technology transfer Ventures, which holds thousands of university patents—thrive.38 The California Institute of Technology (CalTech), one of the largest university patent owners, licensed over 50 patents to Intellectual Ventures in 2008, as a means of recouping research and development investments.39 Patent auctions have become increasingly popular means for extracting licensing royalties, often at a significant loss on the investment in research and development. For example, a recent patent offered for auction, one of 73 being offered for auction by Penn State University, could be purchased for as little as $5,000.40 Such a low price for acquiring a license to a university-produced invention devalues the market for legitimate patents and suggests that some patents are only useful as a litigation weapon in the hands of patent assertion entities.41 This use of patents to generate insignificant amounts of short-term revenue raises concerns about the efficiency of university patents and how they are licensed. Blockbuster patents are the holy grail of university TTOs, but blockbuster patents are very rare. Many investments in university research and development never result in a useful invention. While a few universities strike it rich, most do not. Only 11 percent of TTOs are profitable, and a much smaller percentage generate significant revenue. Essentially, investing in technology transfer is the university equivalent of gambling. While probably a reasonable investment in moderation, it should not become a primary focus, and universities should be wary of allowing technology transfer to determine research investment decisions. After all, the purpose of universities is not licensing patents, but contributing to the public good.

IV.

TECHNOLOGY TRANSFER OFFICES

The rapid growth of university patent activity was driven, at least in part, by the widespread creation of university TTOs, which apply for and license university patents. The Bayh-Dole Act is credited with driving the creation and expansion of TTOs at universities. Before the Bayh-Dole Act, very few universities had TTOs. But immediately afterward, universities began creating them, and they have become commonplace. Today, more than 700 universities have TTOs. TTOs are responsible for the growth of university patent portfolios and the revenue they generate. While most university patents and TTOs are unprofitable, a small minority generate enormous profits. For example, in 2006, 189 universities generated a total of more than $1.5 billion from their intellectual property portfolios, the vast majority of which came from patent licensing royalties.42 But the overwhelming majority of that revenue went to only a few of

38

See Jacob H. Rooksby, The Branding of the American Mind: How Universities Capture, Manage, and Monetize Intellectual Property and Why It Matters 162 (2016). 39 Ledford, supra note 37. 40 Joe Mullin, November’s “Stupid Patent of the Month,” brought to you by Penn State, ars teChnICa, Nov. 25, 2014, 9:55 AM, available at https://arstechnica.com/tech-policy/2014/11/novembers -stupid-patent-of-the-month-brought-to-you-by-penn-state/ (last visited Oct. 17, 2019). 41 Colleen Chien, Presentation to the DOJ/FTC Hearing on Patent Assertion Entities: Patent Assertion Entities, at slide 23 (Dec. 10, 2012), available at http://papers.ssrn.com/sol3/papers.cfm ?abstract_id=2187314 (last visited Oct. 17, 2019) (reporting that 61% of patent lawsuits filed from January 1 to December 1, 2012, were filed by patent assertion entities). 42 Gibbons, supra note 16.

Technology transfer and the public good 245 those universities. As is so often the case in higher education, return on investment is highly stratified. The top twenty universities by patent revenue received 83 percent of the $1.5 billion.43 Of course, when it comes to university patents, that stratification is leavened by chance. There is always the possibility that a lesser-known university will produce an especially valuable patent. Regardless of the prestige of the university at which a patent originates, however, commercialization can present a challenge. The prevailing theory of the Bayh-Dole Act holds that university patents are justified because private firms need exclusive patent rights in order to commercialize innovations.44 Indeed, one of the explicit goals of the Bayh-Dole Act was to facilitate the commercial exploitation of university research.45 But scholars have questioned whether exclusivity is always necessary for effective commercialization.46 While it is probably true that exclusivity is necessary in some cases—for example, pharmaceuticals, where regulatory costs are high—it may not be necessary in other cases.47 And it is inefficient to create exclusive rights when they are unnecessary. The nearly 40 years that have passed since the Bayh-Dole Act was signed into law provide insight into the fact that the central theory on which the law was based—promoting the commercialization of university research—has not come to fruition as envisioned by the Act’s sponsors. Yet, even though the vast majority of the revenue from licensing university patents goes to a tiny minority of universities, many consider the Bayh-Dole Act an unmitigated success. This view is not entirely unfounded. Without the Act, and the ownership interests it created for universities, most of those universities could not have secured patents at all. Surely, $1.5 billion or more in revenue is an enormous benefit to universities, even if it is distributed unequally among them.48 But this view also considers only the benefits and ignores the costs. It is true that, if universities cannot get patents, universities cannot generate licensing revenues. But if otherwise patentable innovations are placed in the public domain, that act increases competition to commercialize the innovation and presumably reduces the market price of the ultimate product. Moreover, when universities have an incentive to pursue patents, they end up with a lot of worthless patents, adding to the problem of patent thickets, especially because universities have limited information about the needs of commercial firms. Finally, the pursuit of patents may distract universities from their core mission of promoting the public good. A.

Advantages and Disadvantages of TTOs

TTOs enable universities to identify potentially patentable research, obtain patents, and license those patents for commercial use, ideally generating revenue in the process. But the overwhelming majority of TTOs lose money. Between 1993 and 2013, 87 percent of university TTOs operated at a loss, primarily because most universities collect only a fraction of 43

Id. Ian Ayres & Lisa Larrimore Ouellette, A Market Test for Bayh-Dole Patents, 102 Cornell l. rev. 271, 286 (2017). 45 See 35 U.S.C. § 200 (2012) (including among the goals of the statute “to promote the utilization of inventions arising from federally supported research or development,” and to “protect the public against nonuse … of inventions”). 46 Ayres & Ouellette, supra note 44. 47 Id. at 289–90. 48 Id. 44

246 Research handbook on intellectual property and technology transfer the licensing revenue from their patent portfolios, while paying all of the overhead for their TTOs.49 Most university TTOs would need to generate nearly $1 billion in product sales from their intellectual property to justify their average $5 million budget, given an average 2 percent chance of expected royalty return from university research and development investments and the average 25 percent share of royalties that TTOs usually claim.50 University TTOs that operate at a deficit may do so because of their business model. While different universities have different policies, typically, one third of patent licensing revenue goes to the inventors, one third goes to the lab producing the patent, and one third goes to the university.51 But this system works only if a university’s patent portfolio includes commercially valuable patents that generate substantial licensing revenues. If a university owns a blockbuster patent, its TTO may be rolling in cash, but if it does not, it may be bleeding cash instead. Well-funded universities with engineering programs and medical schools have a considerably higher chance of landing a blockbuster patent, only exacerbating the disparity in the patent sector. Elite universities that file many patents have about a 30 percent chance of landing a revenue-generating patent, while their plebeian cousins have only a 5 percent chance or less.52 Of course, even TTOs that operate at a loss may generate social value. For example, university patents contribute to the introduction of about two new commercial products every day, and TTOs facilitate the launch of about two new businesses every day. Not all of these generate substantial revenue for the university, but they may nevertheless generate substantial social value. Accordingly, it is hard to judge the success of TTOs on purely quantitative measures, because relying exclusively on income is necessarily reductive and does not reflect the breadth of their mission. B.

TTOs and the Public Good

For all of their positive public good externalities, scholars have identified ways in which university TTOs impede the public good. For example, TTOs may be poorly aligned with the mission of the universities with which they are affiliated. For example, the University of Minnesota rightly views itself as an institution with global reach, but its patent to the anti-HIV/AIDS drug, Ziagen—a lifesaving technology—is mostly beyond the reach of the poverty-stricken countries most affected by HIV and AIDS.53 Instead, scholars have characterized Ziagen as being used for revenue for the university and not for public benefit. Also, “golden rice”—a food source designed to address Vitamin-A deficiencies that could result in child blindness in third-world countries and that was first developed by the Rockefeller Foundation and clinically-tested at Louisiana State University—was delayed in its marketa-

49 Walter D. Valdivia, University Start-Ups: Critical for Improving Technology Transfer, Ctr. for teCh. Innov. at brookIngs, Nov. 2013, at 6. 50 This of course assumes that royalty income alone is the sole source of revenue for a TTO. It is not, but it accounts for an overwhelming share of university revenue. Dipanjan Nag, The Changing Face of University Technology Transfer, Ip watChdog, Oct. 9, 2017, available at http://www.ipwatchdog.com/ 2017/10/09/the-changing-face-of-university-technology-transfer/id=88853/ (last visited Oct. 17, 2019). 51 Id. 52 Id. 53 Ronald L. Phillips, Intellectual Property Rights for the Public Good: Obligations of U.S. Universities to Developing Countries, 6 MInn. J.l. sCI. & teCh. 177, 183 (2004).

Technology transfer and the public good 247 bility by almost 2 years due to the technology being tied up in 70 patents held by 32 organizations.54 These examples illustrate how hard it is to reconcile a university’s duty to society to further knowledge with a university’s self-interested use of public funding for research that results in profits realized only by the university.55 This tension is further complicated by the language in documents formally creating and authorizing TTOs to act: their mission statements. A study of 165 American university TTO mission statements indicated that, in terms of their mission statement, many TTOs were created in service of the public good and not private returns.56 For example, the most frequently included terminology in the mission statements of these TTOs is “commercialization,” at 35 percent, followed by “public good,” at 22 percent.57 While mission statements may not be a perfect means of assessing the purpose of university TTOs, they are useful in determining the extent to which disclosure—in service of the public good—is an important element of university research. Moreover, disclosure—as a goal of patent—may be impaired by the presence of TTOs. Prior to the establishment of TTOs, universities used to immediately publish their findings.58 Today, researchers are encouraged by TTOs to hold off on prompt publication of work until a patent is filed or even longer.59 To put academics’ declining willingness to disclose in quantitative perspective, in 1966, 50 percent of biologists felt safe sharing current research with others, but this number dropped to 26 percent by 1998.60 Thus, the regulations imposed by TTOs may have had a chilling effect on sharing research results between universities, slowing the advancement of knowledge. There is also a concern that certain academic disciplines may be preferred by the boards controlling university TTOs. A study conducted in 2000 found that “28% of life sciences faculty received private sponsor funding, 15% held equity in private sponsor, 33% were engaged in paid consulting arrangements, and 32% held board consulting positions.”61 This example highlights the conflicts of interest that may exist at the faculty level and could evince the reality that TTOs’ priorities might lead to funding being preferentially given to certain academic departments, regardless of their economic viability.

54

Id. See, e.g., Mark A. Lemley, Are Universities Patent Trolls?, 18 fordhaM Intell. prop. MedIa & ent. l.J. 611 (2008). 56 Seok-Ho Kim & Alan S. Paau, Bridging the Gap for Public Good: A Study on the Mission Statements of U.S. University Technology Transfer Programs, 46 les nouvelles 244, 245 (2011). 57 Id. 58 Margo A. Bagley, Academic Discourse and Proprietary Rights: Putting Patents in Their Proper Place, 47 b.C. l. rev. 217, 218 (2006). 59 Id. 60 Id. 61 TTOs might in fact be the beachhead of the corporate world in universities, concerned with the bottom line, which conflicts with the purpose of universities: to promote the dissemination of academic knowledge. See, e.g., Ryan & Frye, supra note 18. Lemley calls universities “‘crack addicts’ directed by ‘small minded tech transfer offices’.” See, e.g., Lemley, supra note 55. Universities also give exclusive licenses for the higher price tag (over 60% of the licenses in 2005); see also Mark L. Gordon, University Controlled or Owned Technology: The State of Commercialization and Recommendations, 30 J.C. & u.l. 641, 647 (2004) (examining how university technology transfer programs are seen by industry and if they act as patent trolls). 55

248 Research handbook on intellectual property and technology transfer Additionally, it has been argued that while prior regimes may have stifled universities’ entry into the patent sector, the present regime may over-incentivize patent-seeking among universities, even when those patents may be acquired by non-practicing entities. For example, in 2012, universities filed 14,000 patent applications, received about 5,000 patents, and formed about 5,100 new patent licensing agreements.62 Because those licensing agreements are largely held by third parties—practicing and mostly non-practicing entities—universities rarely go to court themselves63 to litigate possible infringements and instead “hide behind patent trolls,” or non-practicing entities, which cost an estimated $29 billion per year.64 This recent development has many implications for the public dissemination of knowledge, whether funded by public or private investment.65 On one hand, universities should reap the benefits of their research, perhaps even when this benefit may only be realized through litigation—especially since most universities reinvest proceeds from their inventions back into the university for many virtuous purposes. However, when the universities—whether on their own or through an intermediary—initiate legal actions regarding their patent portfolio, this assumption may be placed in tension with the idea that universities should promote the free spread of knowledge or, more fundamentally, that higher education should work in the public interest.

V.

SOLVING UNIVERSITY PATENT FAILURES

While universities and patents are both intended to increase net public welfare, university patents can create incentives for universities to make inefficient investments. The uncertainty of patentability encourages universities to invest in weak patents as well as strong ones. The possibility of licensing revenue encourages universities to invest in patentable research rather than valuable research. And the unpredictability of patent value encourages universities to invest in patenting everything they can, rather than asking whether a patent is likely to have commercial value. Potential licensing revenue encourages universities to invest in research. But maybe universities and the government can solve those inefficiencies, or at least mitigate them.

62 Daniel Engber, In Pursuit of Knowledge, and Profit: How Universities Aid and Abet Patent Trolls, slate, May 7, 2014, 11:49 PM, available at http://www.slate.com/articles/technology/history _of_innovation/2014/05/patent_trolls_universities_sometimes_look_a_lot_like_trolls.html (last visited Oct. 17, 2019). 63 Universities themselves only participate in 1 to 2 percent of all litigation involving their patents. Id. Moreover, many universities are hesitant to publicly assert their patent rights and instead prefer to have intermediaries, such as non-practicing entities, do so for them. See Rooksby, supra note 38. 64 Id. For example, California Institute of Technology sold exclusive rights to 51 of its patents in 2008 to the notorious patent aggregator Intellectual Ventures, which has acquired rights from at least 60 separate American universities. Id. However, licensing to non-practicing entities is a perfectly rational response to the inherent incentives in the present patent regime and should not necessarily be viewed as improper. See, e.g., Robert E. Thomas, Vanquishing Copyright Pirates and Patent Trolls: The Divergent Evolution of Copyright and Patent Laws, 43 aM. bus. l.J. 689, 705 (2006). 65 Liza Vertinsky, Universities as Guardians of Their Inventions, 2012 utah l. rev. 1949 (2012).

Technology transfer and the public good 249 A.

Clarity

Patent policy inspires a lot of disagreement, but almost everyone agrees that the requirements for patentability is unclear. Under the Patent Act, an invention or discovery is eligible for patent protection only if it comprises patentable subject matter that is novel, useful and non-obvious.66 But none of those requirements for patentability are clearly defined, especially patentable subject matter. As a consequence, universities have no way of knowing which patents to pursue and which to ignore. The government could improve the efficiency of university patents—and probably all patents—by increasing the clarity of the requirements for patentability. Of course, that is easier said than done. The concept of “novelty” is inherently context-specific, the concept of “usefulness” has already been defined largely out of existence, and the concept of “non-obviousness” is intentionally discretionary. However, the government could easily clarify the patentable subject matter requirement by making it categorical. Currently, courts must determine whether a patent claims patentable subject matter by asking whether it has certain qualities: does it claim a “law of nature,” is it “naturally occurring,” is it “too abstract,” and so on. This creates considerable uncertainty about whether any particular claim is patentable, creating substantial information costs borne by universities. Instead, the government could simply redefine the patentable subject matter requirement to exclude certain categories of inventions and discoveries from patentability, like software and business methods. Several justices have already suggested this change, and Congress could easily adopt their suggestion. A categorical approach to patentable subject matter would almost certainly improve the efficiency of the patent system as a whole by excluding many weak claims from patentability, with minimal effect on the production of valuable inventions and discoveries. For example, most software and business method patents are already invalid, and few of the claims they disclose require a patent incentive. The primary value of software and business method patents is litigation, not licensing.67 Eliminating them would simply prevent rent-seeking. Likewise, a categorical approach to patentable subject matter would especially benefit universities by helping them allocate their resources more efficiently. Universities want to invest in valid patents. But as it stands, they cannot know for certain whether software and business method patents are valid. Accordingly, university TTOs often feel obligated to file applications for software and business method patents, even though the Patent Office is likely to deny the applications, and the patents are likely to be invalid anyway. Excluding software and business method claims from patentability would enable TTOs to focus their resources on claims that are more likely to be valid. One problem with university patents, and patents in general, is that the standards for validity are unclear. TTOs invest university resources in weak patents because it is unclear whether

66

35 U.S.C. §§ 101-03 (2012). Patents can have both licensing value and litigation value. A patent has licensing value if its claims are valid and have commercial value. But a claim can have litigation value, even if its claims are invalid, so long as someone wants to practice them. A rational defendant will settle for anything less than the predicted cost of litigation, even if the claims are invalid. Accordingly, patent owners can benefit from invalid patents, so long as their litigation value is higher than the cost of prosecution. If the predicted settlement is higher than the cost of obtaining it, the patent is economically valuable. 67

250 Research handbook on intellectual property and technology transfer patents are valid. Thus, TTOs have an unfortunate incentive to apply for patents, irrespective of the strength of the patent claim or the likely commercial value of the claim. Once the university owns the patent, it has to license it to someone in order to recover its investment, which includes maintenance fees following issuance of the patent. Thus, in the case of low-value patents, universities have incentive to and often do license to patent assertion entities. But universities typically do not want to apply for or own invalid patents, and do not want to file patent infringement actions or be associated with patent infringement actions, especially for dubious patents. This is not only because they do not want to invest resources in litigating patent validity, but also because they are disincentivized to pursue patent litigation at the expense of their public goodwill. Naturally, universities want to own valid and valuable patents, but because present incentives do not always move universities and their TTOs to pursue a purely profitable patent portfolio, the solution to this issue may present itself outside of the university infrastructure. Thus, the USPTO should more thoroughly vet patent applications and thereby remove the incentive for universities to auction non-profitable university produced technologies by refusing to grant patents to inherently unmarketable inventions, technologies that are so remotely useful or economically viable, or those likely to only end up in the hands of non-practicing entities for the purpose of being used as a sword against technological progress.68 B.

Discretionary Patents

University patents are discretionary. While the Bayh-Dole Act enables universities to patent inventions produced by federally-funded research, they are not obligated to patent anything. Universities should ask themselves when patenting is appropriate and when it is not. And universities should provide clear guidelines to their TTOs, explaining when to pursue a patent and when to place an invention or discovery in the public domain. Universities are charitable organizations created to promote the public good. Accordingly, they have a legal duty to consider whether their actions are consistent with their charitable mission. Sometimes, university patents are the best way for universities to promote the public good. But sometimes they are not. For example, if university research produces an innovation that improves a luxury good, then patenting is probably appropriate, as the high-income consumers of the good will happily pay for the improvement, and the university can use the additional revenue to fund additional research. But if university research produces an innovation that leads to a life-saving drug, then patenting may not be appropriate; even though a patent would generate revenue, some low-income consumers may die—who would not have if the university had not patented the drug in the first place—because they are priced out of the market for the drug. From a distributive justice perspective, it may be justifiable to use patents on luxury goods to help fund university research, but unjustifiable to use patents on life-saving

68

See, e.g., Tracie L. Bryant, The America Invents Act: Slaying Trolls, Limiting Joinder, 25 harv. J.l. & teCh. 697 (2011). This approach is actually codified by statute. For example, the America Invents Act works to lower the number of suits patent trolls can bring by limiting the number of defendants that can be joined in a suit. This reduces litigation cost and time. Leahy-Smith America Invents Act, Pub. L. No. 112-29, 125 Stat. 284 (2011).

Technology transfer and the public good 251 drugs. It is hard to see how funding university research could outweigh saving lives. As Jonas Salk famously observed of the polio vaccine: “There is no patent. Could you patent the sun?”69 Accordingly, before a university applies for a patent, it should consider whether patenting is in the public interest, and before a university licenses a patent, it should consider whether licensing is in the public interest. Universities should regularly review their patent portfolios, in order to identify patents that are no longer in the public interest. Universities should consider using open patent licenses when appropriate. And universities should release patents to the public domain if they lack significant commercial value or if it is otherwise in the public interest to do so. C.

Discretionary Grants

Government grants are also discretionary and intended to fund research that will benefit the public interest. Accordingly, grantmakers endeavor to allocate grant funds in order to maximize the public interest. Among other things, grantmakers focus on funding basic research, rather than applied research, because basic research tends to generate more social value than market value.70 But grantmakers can do better. Whenever the social value of research exceeds its market value, grants can play an important role. For example, applied research to develop treatments for medical diseases of the poor or innovations that cannot be patented are ideal targets for grant funding.71 Likewise, grantmakers should look for innovations with high social value but low consumer salience, like vaccines.72 Private firms are unlikely to solve social welfare problems like these, because they lack an economic incentive.73 Grants can play an important role in solving these “patent failures.”74 Accordingly, grantmakers should focus on funding research that is likely to generate more social value than market value. Of course, that is easier said than done. Most research generates very little value of any kind. And that is as it should be. We learn from failure as well as success. But it is also hard to know whether research is likely to generate market value. If predicting demand were easy, we would not need markets. Luckily, grantmakers can apply a simple heuristic: if private firms are funding a research project, the government probably should not. Private firms have more information than the government about whether research is likely to have market value. If private firms are interested in a particular research project, it is likely to have potential market value. The government should probably avoid funding this type of project and focus instead on promising projects that private firms are ignoring. When allocating funds, grantmakers should consider whether research is likely to attract private investment. Grantmakers should try to fund university research that is likely to benefit 69

Jane S. Smith, Patenting the Sun: Polio and The Salk Vaccine (1990). See generally W. Nicholson Price, II, Grants, 34 berkeley teCh. l. J. 1. 71 See id. at 43. 72 See Rachel E. Sachs, Prizing Insurance: Prescription Drug Insurance as Innovation Incentive, 30 harv. J.l. & teCh. 153, 168–9 (2016). 73 See Price, supra note 70, at 46 (“This type of social welfare problem—social value that exceeds market price signals—is exactly the type of problem that market actors with private knowledge are ill-suited to fix.”). 74 A “patent failure” is an inefficiency in investment in innovation that patents cannot solve. 70

252 Research handbook on intellectual property and technology transfer the public, but unlikely to attract private investment. For example, grantmakers should try to fund research that is unlikely to have commercial applications and should avoid funding research that is likely to have commercial applications. Of course, it will still be hard for grantmakers to identify research that is likely to generate social value, just like it is still hard for private firms to identify research that is likely to generate commercial value. But grantmakers can fund socially-valuable research that private firms will ignore, because it lacks commercial value.75 Superficially, this approach to grantmaking seems to be in tension with the goals of the Bayh-Dole Act, which was intended to encourage universities to develop commercial applications for university research by allowing them to claim university patents. If grantmakers decline to fund university research with commercial applications, then universities will have less to patent, and the patents are likely to be less commercial. But that reflects a crabbed understanding of the purpose of the Bayh-Dole Act, which was only intended to encourage the commercialization of government-funded university research with commercial applications, not to encourage government funding of university research with commercial applications. After all, grantmakers focus on basic research precisely because it is unlikely to have commercial applications; therefore, it is unlikely to be pursued by private firms. If university research is likely to have commercial applications, it is more likely to attract private investment and less likely to require government grants. In other words, the proper purpose of the Bayh-Dole Act is to encourage universities to commercialize research that turns out to have commercial value. After all, commercial value is hard to predict, and sometimes basic research will generate a commercially valuable innovation. In that case, the Bayh-Dole Act enables the university to patent the innovation and internalize some of the commercial value. Ultimately, grantmakers should strive to fund university research that is likely to generate social value and unlikely to be funded by private firms. D.

University Patent Policy

As Fritz Machlup famously observed in his 1958 study of United States patent system: If we did not have a patent system, it would be irresponsible, on the basis of our present knowledge of its economic consequences, to recommend instituting one. But since we have had a patent system for a long time, it would be irresponsible, on the basis of our present knowledge, to recommend abolishing it.76

The same is probably true of the Bayh-Dole Act. In retrospect, many of its effects were neither intended nor expected. Indeed, it may be quite inefficient, but it is hard to imagine Congress revisiting its decision to allow universities to patent innovations generated by grant-funded research, especially because it would cause enormous disruption. However, the government may be able to increase the efficiency of the Bayh-Dole Act on the margins. For one thing, the government could begin exercising its “march-in” rights to license and use university patents. Under the Bayh-Dole Act, the government retains the right to license 75

See Price, supra note 70. See Subcomm. on Patents, Trademarks, and Copyrights of S. Comm. on the Judiciary, 85th Cong., An economic Review of the Patent System, Study No. 15, at 32 (Comm. Print 1958) (prepared by Fritz Machlup). 76

Technology transfer and the public good 253 a university patent to a new licensee, if it is not made “reasonably available” by the original licensee, as well as the right to use university patents for government purposes.77 In other words, if a university patent is licensed by a patent assertion entity or “patent troll,” the government can license it to a practicing entity. And if a university patent is not being used, because it lacks market value, the government can use the patent itself and realize its social value. In that way, the government could solve many of the inefficiencies associated with university patents. Unfortunately, the government has never exercised its “march-in” rights and it seems to have little appetite to do so.78 If Congress is willing to revise the Bayh-Dole Act, small changes could dramatically improve its efficiency. For example, Ayres and Ouellete have proposed the adoption of a “market test” for licensing university patents. Under their test, before licensing a patent at prevailing market rates, a university would be required to determine whether private firms would be willing to commercialize the patent in exchange for a nonexclusive license at a nominal fee. If so, the university would be precluded from licensing the patent at prevailing market rates, because it would be against the public interest.79 Presumably, universities and their TTOs would be appalled by such a test, as it would almost certainly prevent them from maximizing the economic value of their patent portfolios. But it would also make university patents more economically efficient by ensuring that universities grant exclusive licenses only when they are necessary to provide an adequate incentive for commercialization and do not create deadweight losses by pricing potential competitors out of the market.80 E.

University Patent Pools

University patents are intended to encourage universities to invest in the production of socially valuable research by enabling them to internalize some of the positive externalities associated with that research. Unfortunately, they do not accomplish that goal very efficiently because they direct almost all of the benefits to a very small number of universities in a largely arbitrary fashion. Ideally, those benefits would be distributed more broadly among universities. While the Bayh-Dole Act establishes an “eat what you kill” system, in which universities benefit only from their own patents, nothing prevents universities from establishing alternative methods of distributing revenue. One potential alternative is “university patent pools,” which would aggregate university patents and share profits among all of the members of the pool. In fact, a version of a university patent pool already exists. Historically, many university researchers have assigned their patents to the Research Corporation for Science Advancement.81 The Research Corporation is a charitable organization that uses all of the profits from the donated patents in its portfolio to fund the advancement of scientific research. 77

28 U.S.C. § 1498 (2012). See Ayres & Ouellette, supra note 44; see also, Hannah M. Brennan, et al., A Prescription for Excessive Drug Pricing: Leveraging Government Patent Use for Health, 18 yale J. l. & teCh. 275 (2016) (arguing that the government should use its march-in rights to purchase low-cost generic versions of high-cost drugs). 79 See, generally, Ayres & Ouellette, supra note 44. 80 See id.; see also, Daniel Jacob Hemel & Lisa Larrimore Ouellette, Innovation Policy Pluralism, 128 yale l. J. (forthcoming 2018). 81 Res. Corp. for Sci. Advancement, available at http://rescorp.org/ (last visited Sept. 23, 2018). 78

254 Research handbook on intellectual property and technology transfer While the Research Corporation is an admirable organization, ensuring that university patents donated to it are used in service of the public good, few universities choose to donate their patents. Presumably, universities prefer to use the proceeds from their patent portfolios to fund their own research, rather than research in general. The principal problem with this scheme is that there is considerable risk associated with university patents because universities cannot know ex ante whether their patent portfolios will contain any commercially valuable patents. A small minority of universities were fortunate enough to obtain some patents with substantial commercial value, and a few ended up with a blockbuster patent that generated windfall profits. However, the overwhelming majority merely break even, or even lose money. But university patent pools could mitigate the risk associated with university patents through profit-sharing. For example, a group of universities could create a charitable organization similar to the Research Corporation and pledge to give all of their patents to the organization, in exchange for an equal share of any profits generated by the organization.82 Ideally, universities would collectively create one such organization, but there could also be multiple organizations comprising a subset of universities. Such an organization could also facilitate some of the other efficiency-increasing approaches to university patents, like placing clearly uncommercial patents in the public domain. Obviously, university patent pools would appeal primarily to universities looking to pledge future patents. Universities that already own commercially valuable patents would be unlikely to give them to a pool. But many universities might also give to a pool currently non-performing patents, in order to reduce administrative costs, given the economies of scale associated with managing a patent portfolio. In any case, university patent pools would enable universities to hedge against the risk of a non-performing patent portfolio, and make university patents more efficient by distributing the profits from commercially valuable patents more broadly. If university patents are like an academic lottery, bestowing windfall profits randomly on lucky universities, there is no reason to believe that lightning will strike twice in the same place. University patent pools recognize that commercially valuable university patents are unpredictable and can surface anywhere, so it is more efficient to distribute the surplus as broadly as possible.

VI.

CONCLUSION

In sum, university patenting practices have become misaligned with the public-facing purposes of the university. Under the US patent law and policy regime, universities have tremendous incentive to act in their own self-interest, which results in market inefficiencies. We suggest five changes to remedy the current state of affairs in technology transfer. First, the requirements for patentability should be clarified. We recommend making patentable subject matter categorical to resolve issues of what is patentable and what is not, benefitting all patent filers and especially universities, given their strong interest in procuring valid patents. Second, universities should regularly review their research and development portfolios, considering whether seeking patents, licensing them, or alternative actions best serve the public interest. 82 Of course, the pool could also offer a premium to the universities that generated commercially valuable patents.

Technology transfer and the public good 255 Our next two suggestions address inefficiencies in the university patent market. Funders of university research—particularly federal funding agencies—should fund the types of research projects that the private sector is unlikely to support. And as a related matter, the federal government should use its march-in rights to take ownership of federally-funded research when the university patents are unused or unprofitable. Finally, universities should consider adopting a patent pool model, sharing profits from successful patents and distributing the losses among them equitably. These reforms should result not only in greater market efficiencies within the technology transfer sector but also a greater alignment between the public purpose and technology transfer practices of universities.

12. US patent sales by universities and research institutes Brian J Love, Erik Oliver and Michael Costa

I.

INTRODUCTION

In addition to licensing or litigating patent rights, universities have the option to sell their patents outright to third parties. In this Chapter, we explore the extent to which universities and other nonprofit research institutes currently participate in the secondary market for patents. We find that, while research-focused entities do participate in the patent marketplace, they do so relatively rarely. During the period 2012–2017, we were able to document 220 assignments, involving a total of 544 US patent assets, that appear to be the result of arm’s-length patent transactions negotiated with universities and other nonprofit research institutes.1 By comparison, the entire US patent market has been estimated to exceed 100,000 assets in the same period,2 with individual companies like Alcatel Lucent, AT&T, and IBM selling over one hundred assets in 2017 alone.3 In addition to identifying sales, we present data on the entities and assets involved, as well as the publicly available circumstances underlying each sale. Among other findings, we show that foreign universities are the most active market participants. Overall, US universities and labs account for less than one quarter of sales. We also observe that few patent transactions bear the hallmarks of technology transfer. Just 11 percent of assets appear to have been purchased for commercialization. Virtually all other purchases appear to have been either defensive acquisitions by operating technology companies or purchases by non-practicing entities. The remainder of this Chapter is organized as follows. First, we provide a brief description of the marketplace for patent rights and introduce the most common reasons why academic institutions transfer their rights. Next, we summarize our data collection methodology and present descriptive statistics on patent sales by universities and research institutes. Finally, we analyze our findings and consider what conclusions policymakers and university administrators may draw from our data.

1

Here and throughout, “assets” includes both issued patents and pending applications. Between 2012 and 2017, approximately 10,000 US patent assets were sold through patent brokers. See Brian J. Love, Kent Richardson, Erik Oliver, & Michael Costa, An Empirical Look at the “Brokered” Market for Patents, 83 Mo. l. rev. 359 (2018); Kent Richardson, Erik Oliver, & Michael Costa, The 2018 Brokered Patent Market, Intell. asset MgMt., Jan./Feb. 2019, at 24. Though it is unclear how many additional patents were transferred in private sales, we previously estimated “[e]xtrapolating from overall PTO assignment data … that the private market is roughly ten times larger than the brokered market.” Love, et al., supra, at 368, n 42. 3 For a ranking of top sellers, see Richardson, Oliver & Costa, supra note 2. 2

256

US patent sales by universities and research institutes 257

II.

THE PATENT MARKETPLACE

Broadly speaking, the market for patent rights can be divided into deals that license the right to use patented inventions and deals that lead to the outright sale of one or more patents.4 The market for patent sales can, in turn, be subdivided into two segments: first, a quasi-public “brokered” market in which patent assets are shopped by patent brokers to multiple potential buyers and, second, a private market in which specific parties negotiate transactions on an ad hoc basis and generally in secret.5 Patents offered for sale on the “brokered” market are typically offered to multiple prospective buyers and thus are generally observable to interested market participants—though often subject to confidentiality agreements that render the market unobservable to the public at large.6 Patents offered for sale in this manner are virtually always shopped by patent “brokers,”7 who act as the intellectual property equivalent of real estate agents. Brokers market patents on sellers’ behalf, negotiate deals with potential buyers,8 and in return take a fee of roughly 20 percent of deals that they close.9 Patent purchases that take place outside the brokered market are both harder to observe and harder to categorize. Sales in this “private” market are often large deals negotiated directly between buyers and sellers with broader considerations in mind. Beyond the transferred patents themselves, private buyers are often motivated by general concerns about freedom to operate in their industry, risks posed by specific competitors, and vulnerabilities that accompany public announcements of funding rounds or public stock offerings.10 Most patents sold on the secondary market come from operating technology companies, though often from companies that are more established and sometimes during periods of financial distress.11 Patent buyers, by contrast, are more diverse. In addition to operating companies and “non-practicing entities” (NPEs)—entities that specialize in monetizing patent rights12— patents are frequently purchased today by patent “aggregators” that facilitate coordination

4 In reality, of course, many deals include both licenses and sales. In addition, as we discuss infra, long-term exclusive licenses can effectively function as sales. 5 See Love, et al., supra note 2, at 364–5. 6 Id. at 364. 7 Fewer than two percent of patents shopped on the brokered market between 2012 and 2017 were, to borrow a phrase from the real estate market, “for sale by owner.” Id. 8 Brokers often perform tasks such as selecting which patent assets to sell, setting asking prices, identifying potential infringement, identifying and contacting potential buyers, and establishing a procedure for prospective buyers’ diligence and bidding. See Kent Richardson, Erik Oliver, & Michael Costa, Inside the 2016 Brokered Patent Market, Intell. asset MgMt., Jan./Feb. 2017, at 35. 9 See Kent Richardson & Erik Oliver, Turning the Spotlight on the Brokered Patent Market, Intell. asset MgMt., Jan./Feb. 2013, at 11, 16 (reporting an average commission rate of 22 percent). 10 Consider for example one deal that became public in 2014. That year, Twitter purchased 900 patents from IBM in a $36 million deal. However, the parties’ agreement followed a threat from IBM to file suit against Twitter shortly before its initial public offering, and the deal included cross-licensing terms in addition to the transfer of patents See, e.g., Klint Finley, Twitter Pays $36 Million to Avoid IBM Patent Suit, wIred, Mar. 7, 2014, 2:42 PM, available at https://www.wired.com/2014/03/twitter-ibm/ (last visited Oct. 17, 2019). 11 See Love, et al., supra note 2, at 391–2 (comparing the most active buyers and sellers in the brokered market between 2012 and 2016). 12 For a discussion of the various kinds of NPEs, see, e.g., Shawn P. Miller, et al., Who’s Suing Us? Decoding Patent Plaintiffs since 2000 with the Stanford NPE Litigation Dataset, 21 stan. teCh. l.

258 Research handbook on intellectual property and technology transfer among multiple buyers with similar interests.13 “Defensive” aggregators like Allied Security Trust (AST) and RPX use a membership fee-based model to accumulate funds that can be used to purchase patents of interest to their members.14 Generally, these patents fall into one of three categories: (i) patents currently being enforced against members, (ii) patents that members fear may be asserted against them in court down the road, or (iii) patents that members wish to hold for possible defensive use against either competitors or corporate patent asserters. Other aggregators buy with monetization in mind and are generally considered NPEs.15 Intellectual Ventures (“IV”), which claims to have acquired over 70,000 patent assets since it was formed in 2000,16 is the most well-known example.17 Despite the large sums of money at stake and the sophistication of many market participants, there remains no central clearinghouse for patent offerings. While third-party platforms play an important and increasingly common role in bringing potential buyers and sellers together,18 the lion’s share of patent offerings are never made public and the terms of consummated sales are revealed even less often.

III.

UNIVERSITY PATENT ASSIGNMENTS

Due to the market’s opaqueness, little is known at present about universities’ participation. On one hand, we know that universities regularly transfer patent assets. In a recent study, Fusco et al. found that universities collectively transferred more than 20,000 US patent assets between

rev. 234 (2018). Our use of the term in this Chapter excludes universities and IP-holding subsidiaries of operating companies, both of which are sometimes included in NPE taxonomies. 13 See generally Sell Your Patent, rpx, available at https://www.rpxcorp.com/rpx-sell-your-patent/ (last visited Mar. 6, 2018). 14 Company, RPX, available at https://www.rpxcorp.com/about-rpx/ (last visited Mar. 6, 2018) (“By acquiring problem patents, RPX helps to mitigate and manage the risk of potential patent assertions for its growing client network.”); About Us, allIed seCurIty trust, available at http://www.ast.com/ about-us/asts-mission/ (last visited Mar. 6, 2018) (“We are an independent member-based, not-for-profit cooperative that helps companies who use innovative technologies mitigate the risk of patent assertions and litigation by securing rights from patents available on the open market.”). 15 See, e.g., Erik Oliver, Kent Richardson, & Michael Costa, How Intellectual Ventures is Streamlining Its Portfolio, Intell. asset MgMt., May/Jun. 2016, at 9. 16 See Kent Richardson & Erik Oliver, What’s Inside IV’s Patent Portfolio?, Intell. asset MgMt., July/Aug. 2014, at 14. 17 Though largely funded by operating technology companies like Microsoft, Intel, and Sony, IV’s business model goes well beyond defensive acquisition. See Oliver, Richardson, & Costa, supra note 15, at 9 (“Since its founding, it has reportedly raised over $6 billion in capital. A large portion of this has come from corporate investors in the high-tech space, such as Microsoft, Intel, Sony, Nokia, Apple, Google, Yahoo, American Express, Adobe, SAP, NVIDIA and eBay.”). In fact, it is generally understood that IV earns the “vast majority” of its revenue from licensing its portfolio. See id. (“The vast majority of IV’s revenue does not come from making products or offering services. Rather, it comes from licensing its portfolio to other companies—IV is the quintessential NPE.”). 18 At present the most noteworthy example is the IAM Market, available at https://www.iam-market .com/ (last visited Mar. 7, 2018), where an estimated 25 percent of patents offered for sale by brokers are listed by IAM. See Richardson, Oliver, & Costa, supra note 2, at 34 (reporting that between October 2015 and May 2016 the IAM Market “listed 194 packages, with 3,724 assets, from 17 sellers”).

US patent sales by universities and research institutes 259 1990 and 2017.19 In addition it has been shown that assets initially owned by universities sometimes find their way into the hands of NPEs.20 On the other hand, extrapolating from raw assignment records is complicated by the fact that universities transfer patent assets in a number of disparate circumstances. For example, universities regularly transfer assets to “spinout” companies, firms established by university affiliates (typically one or more of the transferred assets’ faculty inventors) for the express purpose of attempting to commercialize the technology. According to surveys conducted by the Association of University Technology Managers (“AUTM”), research institutions formed over 1,000 spin outs in 2017 alone and have spun out more than 11,000 startups since 1995.21 However, spin out formation has more in common with venture capital financing or vertical disintegration than it does with deals negotiated in the secondary market for patent rights. For example, spinouts generally exchange equity (not cash) for patent assets and expect to benefit from the knowhow of university personnel or recent graduates,22 both of which are virtually unheard of in the brokered and private patent markets.23 In addition to forming new companies, universities also regularly collaborate with industry partners that fund or actively participate in research conducted by faculty members and graduate students, generally with an understanding that the industry partner will receive rights to any technology developed during the partnership. Again according to AUTM surveys, industry funds account for about eight percent of all university R&D expenditures, or about one quarter of R&D expenditures not covered by government funds.24 Consequently, universities regularly transfer assets to partner companies pursuant to ex ante joint and funded research agreements that, like spinout formation agreements, are not readily comparable to ex post deals negotiated in the patent marketplace. Perhaps the surest thing that the existing literature shows is that it is comparatively rare for universities to transfer assets following arm’s-length negotiations with true third-parties. For

19

Stefania Fusco, et al., Dissemination of Academic Knowledge and Monetization of University Patents (Aug. 10, 2018) (unpublished manuscript, on file with the authors). 20 Yarden Katz, Universities Have Turned over Hundreds of Patents to Patent Trolls, Medium, Oct. 13, 2016, available at https://medium.com/@yardenkatz/universities-have-turned-over-hundreds -of-patents-to-patent-trolls-99d5cdec1d8a#.315c8xj7c (last visited Oct. 17, 2019) (reporting that “nearly 500 of IV’s patents originally belonged to universities, including state schools”). 21 Driving the Innovation Economy: Academic Technology Transfer in Numbers, AUTM, available at https://autm.net/AUTM/media/SurveyReportsPDF/AUTM_2017_Infographic.pdf (last visited Oct. 26, 2018). 22 AUTM licensing survey respondents report holding equity in about two-thirds of start-ups “formed that were dependent upon the licensing of your institution’s technology for initiation.” See Statistics Access for Technology Transfer Database, AUTM, available at https://autm.net/surveys-and -tools/databases/statt/ (last visited Mar. 28, 2019) [hereinafter “AUTM-STATT”]. 23 See Love, et al., supra note 2, at 371–2 (explaining that “[l]ess than one percent of the packages [offered for sale on the brokered market] included another type of IP right or some form of know-how”). 24 This is the average percent reported among all AUTM licensing survey respondents that reported both overall R&D expenditures and amounts funded by industry. See AUTM-STATT, supra note 22; see also Robert D. Atkinson, Information Technology and Innovation Foundation, Industry Funding of University Research: Which States Lead? 1 (2018), available at http://www2.itif.org/2018-industry -funding-university-research.pdf (last visited Oct. 17, 2019) (reporting that “the share of university research funded by industry increased from 4.9 percent in 1980 to a high of 7.4 percent in 1999” but has “fallen since then, even as federal funds have dropped overall … [to] just 5.9 percent of U.S. academic research” in 2016).

260 Research handbook on intellectual property and technology transfer one, we know that universities rarely participate in the brokered market. In a recent study of patent packages offered for sale through patent brokers between 2012 and 2016, we found that only fourteen packages were sold by universities or other non-profit research institutions.25 In addition, the few times in recent memory when universities (publicly) attempted to attract private sales, their efforts were noted primarily for, initially, their uniqueness and, ultimately, their failure.26 As a result, important questions about the market for university patents remain unexplored: How often are university patents sold? What types of patents are sold? Which universities sell most often? Who buys university patents? And, what circumstances appear to drive university patent sales?

IV.

IDENTIFYING UNIVERSITY PATENT SALES

To shed light on these questions, we set out to identify arm’s-length patent transfers from universities and research institutes that took place between 2012 and 2017. We began by constructing the (known27) universe of university patent assignments that were executed during this period. To do so, we used the USPTO’s assignment database28 to identify all recorded patent assignments involving nonprofit entities focused on basic research.29 After compiling this list, we removed all assignments from entities that were not in fact nonprofits, normalized

25 Love, et al., supra note 2, at 395–6; see also Matt Levy, Why Universities Oppose Real Patent Reform: Money, patent progress, May 7, 2015, available at https://www.patentprogress.org/2015/05/ 07/why-universities-oppose-real-patent-reform-money/ (last visited Oct. 17, 2019) (presenting statistics from Allied Security Trust indicating that universities and labs attempted to sell 277 patent assets between 2007 and 2014). 26 For example, Penn State held an online patent auction in 2014 that netted just one bid. See Goldie Blumenstyk, Penn State’s Patent Auction Produces More Lessons Than Revenue, Chron. hIgher eduC. (May 1, 2014), available at https://www.chronicle.com/blogs/bottomline/penn-states-patent -auction-producesmore-lessons-than-revenue/ (“[T]he university got just one bid on [a] single pair of patents out of 53 batches” and that bidder “offered the minimum bid, $10,000.”). For more detail on Penn State’s efforts to auction patents, see Daniel R. Cahoy, et al., The Role of Auctions in University Intellectual Property Transactions, 54 duQ. l. rev. 53, 58 (2016). 27 There are at least two reasons why USPTO assignment records are incomplete. First, patent assignments are recorded on a voluntary basis and, thus, some sales may never be recorded. Second, it is not uncommon for assignments to be recorded months, and sometimes years, after they were executed. Thus, our data for recent years may not include some assignments that, while executed in 2016 or 2017, will not appear in USPTO records until 2019 or thereafter. 28 Patent Assignment Search, uspto, available at https://assignment.uspto.gov/patent/index.html#/ patent/search (last visited Oct. 26, 2018). There is a programmatic API for accessing the same data which was used for this research at https://assignment-api.uspto.gov/documentation-patent/#/ (last visited Nov. 26, 2018). 29 We started with a list of known research entities including: the names of all universities, hospitals, or other entities that have ever completed the annual survey conducted by the AUTM; US government entities; all branches of the US armed services, including the Army, Navy, Airforce and assignments from the secretaries of each; and any other entity with the word “research,” “institute,” “university,” or some foreign translations of these keywords in its name. We then used a combination of software and human review to recursively analyze the names of assignors to group variations of the same entity name and determine if that entity is a university or research entity.

US patent sales by universities and research institutes 261 and consolidated entity names, and removed all internal transfers30 as well as all transfers between two universities or research institutes. We then reviewed by hand the approximately 3,400 assignments that remained. For each of these assignments, we reviewed the assignment agreement filed with the USPTO, the assigned assets, and the names of the inventor(s), assignor(s), and assignee(s). Using this information we were able to determine (using, e.g., web searches) the circumstances under which the overwhelming majority of assignments occurred. The vast majority of these assignments—about 2,500—were the result of a joint or funded research arrangement and/or the formation of a spinout company. Another 635 assignments transferred patent assets back to one or more of their inventors.31 For the remaining 277 assignments, we found no public evidence of a pre-existing relationship between the assignor and assignee that would allow us to categorize the transfer as one resulting from spinout formation or joint research. Of these, we determined that 220 very likely represented arm’s-length patent transfers to third-party entities.32 For the majority of these, we were able to locate news articles, press releases, securities filings, or other public information confirming our determination. For a minority, however, we made an educated guess based on factors such as how long the assignment occurred after the transferred assets’ filing, publication, or issuance; when and where the assignee was formed;33 whether the assignee frequently sues or is sued for patent infringement; whether the assignee was involved in patent litigation close in time to the transfer; and whether the assignee is included in RPX’s database of NPEs.34 For the final 57 assignments, we were unable to make a determination. The majority of these assignments involved a foreign assignor or foreign assignee about which we were able to find negligible information. While many of these assignments occurred well after the publication or grant of transferred assets, we did not consider this evidence dispositive in isolation. Thus, in an abundance of caution we have excluded them from the analysis that follows. For each of the 220 assignments that we identified as associated with a sale of academic US patents, we collected data on the entities and assets involved. For each assignor, we determined the country in which it is located, whether it is public or private, and whether it is a university or another type of nonprofit research institute. Similarly, for each assignee, we

30 For example, transfers from one arm of a university to a university-affiliated patent-holding partner like the Wisconsin Alumni Research Foundation (WARF). 31 Though we did not examine the subsequent history of these assets, we observed anecdotally that several were later sold to NPEs. See Assignment Abstract of Title for Application 09165316, USPTO, available at https://assignment.uspto.gov/patent/index.html#/patent/search/resultAbstract?id=6185678 &type=patNum (last visited Mar. 21, 2019) 32 Twelve of these recorded transfers (involving 60 total assets) were exclusive licenses, rather than outright assignments. All twelve exclusive licenses were to NPEs. We chose to include them because, in each case, the license appeared to function identically to a sale. One noteworthy example is Stanford’s exclusive license of patents covering a dietary supplement to ThermoLife International, an (all but) non-practicing entity that asserted the patent in 117 lawsuits in 2013. See, e.g., Laura Sydell, Bodybuilders Beef over a Workout Supplement—And a Stanford Patent, nat’l pub. radIo, July 8, 2016, available at https://www.npr.org/sections/alltechconsidered/2016/07/08/483438151/bodybuilders-beef -over-a-workout-supplement-and-a-stanford-patent (last visited Oct. 17, 2019). 33 For example, assignees formed in another country after a transferred patent issued are both unlikely to have funded or participated in the patented research and unlikely to be a spinout. 34 We used RPX’s “Entity Search.” See Entity Search, RPX, available at https://insight.rpxcorp.com/ advanced_search#entityTab (last visited Mar. 21, 2019).

262 Research handbook on intellectual property and technology transfer determined its home country and classified it as a large operating company, a small operating company, a defensive aggregator, or a “non-practicing entity.” In addition, we classified each set of transferred assets as falling within one of ten technology categories,35 and we determined whether any transferred assets had been asserted in district court or challenged before the Patent Trial and Appeal Board.36 Finally, we categorized assignments according to the purpose for which the assets appear to have been acquired—i.e., for monetization by an NPE, for defensive or offensive use of the assets by an operating company,37 or for commercialization of the patented technology. Because patent sale prices are virtually always kept confidential,38 we were able to find pricing information for just a handful of unrepresentative transactions. Accordingly, we are unable to report data on prices or to estimate the fiscal size of this segment of the patent market.

V.

FINDINGS

Basic statistics for the sales that we identified are shown below in Table 12.1. Overall, these 220 assignments involved just 544 patent assets, resulting in an average package size of less than three assets per transfer and a median of just one. By comparison, packages shopped on the brokered market are significantly larger, with an average of 15 transferred assets (including an average of 9 issued US patents) per package.39 This relatively small package size may be explained in part by the fact that we observed an unusually large number of distinct assignments between the same two parties that were executed at roughly the same time. Altogether, the 220 assignments that we identified as sales involved just 81 unique assignors, 65 unique assignees, and 108 unique assignor-assignee pairs.40 If we assume that all assignments between the same two parties that were executed within a 30-day period constitute a single “sale” that was broken up into multiple assignments, our data consolidates to 154 sales. One possible explanation for the fragmented nature of these transfers is that proceeds from university patents often flow to a variety of inventor-specific budget centers, including the inventors’ schools or departments within the university and the inventors themselves.41

35 We began by automatically classifying assets into the technology “areas” that make up the International Patent Classification system, and then consolidated these areas into the ten categories used below. We then performed a manual check of the assets in each “sale” during which we re-classified about 19 percent of assets to more accurately reflect the transferred technology. 36 We collected litigation data by searching Docket Navigator for each sold US patent. See Docket Navigator, available at https://www.docketnavigator.com (last visited Oct. 26, 2018). 37 We categorized a transfer as defensive when it appeared that the acquisition was made to avoid the possibility that the assignor or a subsequent purchaser would enforce the patent against the assignee. We instead categorized a transfer as offensive when it appeared that the assignee acquired the asset for potential enforcement against one or more competitors. 38 See Love, et al., supra note 2, at 360, n. 4 (collecting citations). 39 Love, et al., supra note 2, at 371. 40 A list of assignors and assignees is provided in the Appendix. 41 University policies generally provide a one-third share (net of prosecution costs) to both. See Walter D. Valdivia, University Start-Ups: Critical for Improving Technology Transfer, brookIngs Inst., Nov. 2013, at 9 (“[U]niversities generally split licensing revenue in three parts: a third for

US patent sales by universities and research institutes 263 Table 12.1

Overall statistics on US patent sales by universities and research institutes (2012–2017) US Patent Assets Sold Recorded Assignments Avg. No. Assets/Assignment Median No. Assets/Assign. “Sales”* Avg. No. Assets/Sale Median No. Assets/Sale Unique Buyer-Seller Pairs Avg. No. Assets/Pair Median No. Assets/Pair Unique Sellers Avg. No. Assets/Seller Median No. Assets/Seller Unique Buyers Avg. No. Assets/Buyer Median No. Assets/Buyer

544 220 2.47 1 154 3.53 1 108 5.04 2 81 6.72 3 65 8.37 3

Note: * Here and throughout, we define a “sale” as incorporating all assignments between the same two parties that were executed within the same 30-day period.

A.

Sellers

Figures 12.1, 12.2, and 12.3 provide information on subsets of academic patent sellers. Perhaps the the most striking thing this data reveals is that US-based universities and labs are responsible for less than one-quarter of all transfers. By contrast, Asian universities and labs—especially those located in China, South Korea, and Japan—are more active participants in the secondary market for US patent rights. We consider why this might be so further below. Consistent with prior studies of university patent licensing,42 we also find that universities transferred far more assets than government labs and other nonprofit research institutes. While universities were more active sellers worldwide, we also observe that non-universities played a comparatively larger role outside the US. In fact, Taiwan’s Industrial Technology Research Institute43 was the most active assignor during the period of our study, with 21 recorded assignments transferring a total of 88 US patent assets to eight different buyers. In addition, among universities, we see a high level of participation by public universities. Overall, public universities account for more than 60 percent of all assignments and about 47 percent of all assets sold. Because public universities generally have public-spirited missions and are often governed by public officials, it seems reasonable to hypothesize that technology transfer programs at public universities feel both greater pressure to make the fruits of their

the faculty-inventors, a third for their department or lab, and a third as discretionary funds for the university.”). 42 See Joseph Allen, Increasing the ROI from the Federal Labs, Ip watChdog, May 23, 2018, available at http://www.ipwatchdog.com/2018/05/23/increasing-roi-federal-labs/id=97547/ (“[U]niversities received just over 50% [more] R&D funding [than federal labs], but license nearly 10 times the value of technology.” (quoting Wilbur Ross, US Sec’y of Commerce)). 43 For more information about ITRI, see ITRI, available at https://www.itri.org.tw/eng/index.aspx (last visited Mar. 21, 2018).

264 Research handbook on intellectual property and technology transfer

Figure 12.1

US patent sales by universities and research institutes (2012–2017), grouped by seller type

Figure 12.2

US university patent sales (2012–2017), by university ranking

US patent sales by universities and research institutes 265 Table 12.2

US patent sales by universities and research institutes (2012–2017), grouped by country of origin Country

“Sales”

Recorded Assignments

US Patent Assets

Australia

14 (9.1%)

23 (10.4%)

29 (5.3%)

Belgium

1 (0.65)

1 (0.45)

1 (0.18)

Canada

1 (0.65)

1 (0.45)

14 (2.57)

China (incl. Taiwan & HK)

31 (20.1)

50 (22.7)

138 (25.4)

Germany

1 (0.65)

1 (0.45)

1 (0.18)

Israel

1 (0.65)

1 (0.45)

1 (0.18)

Japan

22 (14.3)

27 (12.3)

36 (6.62)

Korea

37 (24.0)

47 (21.4)

154 (28.3)

New Zealand

1 (0.65)

1 (0.45)

1 (0.18)

Spain

1 (0.65)

1 (0.45)

3 (0.55)

UK

7 (4.54)

16 (7.27)

18 (3.30)

US

37 (24.0)

51 (23.2)

148 (27.2)

Total

154 (100)

220 (100)

544 (100)

research freely-available to the public and less pressure to monetize their portfolios simply to make ends meet.44 However, our results suggest instead that public universities contribute to the secondary market roughly in proportion to their share of university students and faculty members. While there are more private than public four-year universities in the US, American public universities are on average much larger and account for more than 60 percent of students and more than 70 percent of faculty members.45 We also observe that US universities participating in the secondary patent market are drawn from across the ranking spectrum and, predominantly, fall outside the highest tiers. As shown in Figure 12.2, we find that more than 80 percent of sales made, assignments recorded, and assets transferred by US universities came from universities ranked outside the top 50 by US News and World Report.46 In addition, more than three-quarters of sales, assignments, and transferred assets came from universities ranked outside the top 50 highest grossing university patent licensors according to AUTM survey results.47 These results are particularly noteworthy because tech-transfer statistics are typically dominated by a small cohort of highly-ranked universities, most of which do not appear in our data at all.48

44 See, e.g., Peter Lee, Patents and the University, 63 duke l.J. 1, 78 (2013) (“The norms, histories, and missions of public versus private, secular versus religious, and land-grant versus non-land-grant institutions may lead to different visions about a university’s role in society and how patenting and technology transfer advance (or do not advance) that role.”). 45 Digest of Education Statistics, us dep’t eduC., available at https://nces.ed.gov/programs/digest/ current_tables.asp (last visited Nov. 1, 2018). 46 National University Rankings, u.s. news, available at https://www.usnews.com/best-colleges/ rankings/national-universities (last visited Nov. 2, 2018). 47 See AUTM-STATT, supra note 22. 48 See, e.g., Valdivia, supra note 41, at 6 (“In 2012, a year very much in line with the ten-year trends in this sector, the top 5% of earners (8 universities) took 50% of the total licensing income of the university system; and the top 10% (16 universities) took nearly three-quarters of the system’s income.”).

266 Research handbook on intellectual property and technology transfer B.

Buyers

Table 12.3 and Figures 12.3 and 12.4 provide information on entities that acquired US patents from academic sellers. First, we observe a higher level of participation by US entities on the buy-side of the market. Relative to sellers, US academic patent buyers were twice as likely to be located in the US. That said, foreign entities still collectively made up more than half of the market, with entities based in China, Korea, and Japan actively purchasing US patents from academic sellers during the period of our study. In addition, the vast majority of sales transferred patents to NPEs. Overall, NPEs accounted for two-thirds of assignments and acquired about 60 percent of transferred assets. By contrast, and despite the emphasis that many tech transfer offices place on startup formation, only about ten percent of the sales that we observed transferred patents to small operating companies. Not surprisingly, given this breakdown of buyers, the vast majority of acquisitions appear to have been made with patent assertion in mind. In addition to the 60 percent of assets transferred directly to NPEs, another 29 percent appear to be defensive or offensive acquisitions by operating companies hoping to avoid or impose litigation risk. In all, then, roughly 90 percent of the market for academic patents appears to take place in the shadow of patent litigation. That said, we also observe a number of transfers that appear to have been made for the purposes of commercializing patented technology. Compared to the brokered and private markets as a whole, this relatively small percentage stands out as comparatively large.49 Then again, several transfers to prospective commercializers stand out for another reason as well: six of these twelve assignments appear to follow unrecorded licenses to spinouts that failed or were acquired.50 Thus, while our data does in each case show a direct transfer from a university to an unrelated commercializer, these transfers arguably should instead be considered transfers from the spinout, not the university. C.

Technology and Litigation

Finally, Table 12.4 provides information on the technologies covered by acquired assets, and Table 12.5 shows whether those assets were asserted post-transfer. First, we observe that almost 70 percent of transferred assets cover inventions related to computing, telecommunications, and other audio-visual technology. Interestingly, this is true despite the fact that “high tech” products tend to have the shortest product life cycles51 and thus tend to be the 49 See Love, et al., supra note 2, at 404 (concluding that “our data strongly suggests that the brokered market for patents is primarily, and perhaps almost exclusively, a market for the transfer of potential legal liability, not a market for the transfer of technology”). 50 For example, our data includes an assignment from Louisiana State University to Proteovec, a startup formed to commercialize the patented technology. Press reports, however, characterize this transfer not as a sale from LSU, but instead as the result of Proteovec buying LSU spinout “TransGenRx’s intellectual property and technical equipment for $400,000 as the sole bidder in a bankruptcy court auction.” Regenerating TransGenRx: The Promising Discovery of a Now Defunct Biotech Firm May Yet Realize its Commercial Potential and Save Lives, greater baton rouge bus. rep., Dec. 10, 2013, available at https://www.businessreport.com/article/regenerating-transgenrx-the-promising-discovery -of-a-now-defunct-biotech-firm-may-yet-realize-its-commercial-potential-and-save-lives (last visited Oct. 17, 2019). 51 In the computer industry, products become twice as powerful roughly every two years. This observation, which has held true for decades, is known as “Moore’s law.” See Gordon E. Moore, Progress

US patent sales by universities and research institutes 267 Table 12.3

US patent sales by universities and research institutes (2012–2017), grouped by country of destination

Country

“Sales”

Recorded Assignments

US Patent Assets

Australia

1 (0.65%)

1 (0.45%)

1 (0.18%)

B.V.I.

1 (0.65)

4 (1.82)

4 (0.74)

Canada

2 (1.30)

2 (0.91)

2 (0.37)

China (Taiwan)

14 (9.10)

27 (12.3)

99 (18.2)

Japan

19 (12.3)

23 (10.4)

26 (4.78)

Korea

36 (23.40

45 (20.4)

161 (29.6)

Netherlands

2 (1.30)

2 (0.91)

2 (0.37)

Seychelles

2 (1.30)

2 (0.91)

8 (1.47)

Singapore

1 (0.65)

1 (0.45)

9 (1.65)

U.K.

3 (1.95)

5 (2.27)

5 (0.91)

U.S.

73 (47.4)

108 (49.1)

227 (41.7)

Total:

154 (100)

220 (100)

544 (100)

Figure 12.3

US patent sales by universities and research institutes (2012–2017), grouped by buyer type

in Digital Integrated Electronics, Int’l eleCtron devICes MeetIng, IEEE (1975), reprinted in ssCs: Ieee solId-state CIrCuIts soC’y news, Sept. 2006, at 36, 37 (predicting that computing power will double approximately every two years).

268 Research handbook on intellectual property and technology transfer

Figure 12.4

US patent sales by universities and research institutes (2012–2017), grouped by reason for sale

most amenable to transfer using other forms of intellectual property.52 By contrast, biotechnology and pharmaceuticals—both of which are commonly cited as US technology transfer successes53—account for just 20 percent of transferred assets. Moreover, many of the biopharmaceutical asset sales that we observe do not appear to further the commercialization of new drugs. Almost one-third of all transferred pharmaceutical assets relate to an over-the-counter dietary supplement,54 and over 60 percent of all biotech, pharmaceutical, and medical device

52 See, e.g., Stuart J. H. Graham, et al., High Technology Entrepreneurs and the Patent System: Results of the 2008 Berkeley Patent Survey, 24 berkeley teCh. l.J. 1255, 1277, 1289–90 (2009) (finding in a 2008 survey of entrepreneurs that only 24 percent of software startups owned patent assets and that “first-mover advantage,” not patent protection, was ranked the most “important” means to “capture competitive advantage” in the software industry). 53 See, e.g., Joseph Allen, NIH Pressured to Misuse Bayh-Dole to Control Drug Prices, IpwatChdog, March 30, 2016, available at https://www.ipwatchdog.com/2016/03/30/nih-bayh-dole-control-drug -prices/id=67600/ (last visited Oct. 17, 2019) (“One of the greatest successes of the law is the development of life saving therapies from federally funded research. While no new drugs were developed under preceding policies, at least 200 new drugs and vaccines are now protecting human health through Bayh-Dole.”). Noteworthy examples include recombinant DNA, which earned Stanford University and the University of California system approximately $255 million in royalties, see Louis G. Tornatzky, et al., Innovation U: New University Roles in a Knowledge Economy 161–2 (2002), and blockbuster pharmaceuticals Lyrica and Emtriva, which earned Northwestern University $700 million and Emory University $525 million, respectively, see Irene Abrams, et al., How Are U.S. Technology Transfer Offices Tasked and Motivated—Is It All About the Money?, 17 res. MgMt. rev. 1, 3 (2009). 54 Eight of 26 total pharmaceutical assets relate to arginine, a supplement that Stanford University exclusively licensed to the (all but) NPE Thermolife International. See Sydell, supra note 32.

US patent sales by universities and research institutes 269 Table 12.4

US patent sales by universities and research institutes (2012–2017), grouped by technology category Technology

“Sales”

Recorded Assignments

US Patent Assets

98 (63.6%)

137 (62.3%)

377 (69.3%)

Telecomm.

40 (26.0)

45 (20.4)

143 (26.3)

Semiconductors

25 (16.2)

52 (23.6)

111 (20.4)

Electrical

Elec. Other Chemical

33 (21.4)

40 (18.2)

123 (22.6)

40 (26.0%)

58 (26.4%)

133 (24.4%)

Pharma.

5 (3.25)

6 (2.73)

26 (4.78)

Biotech.

26 (16.9)

38 (17.3)

80 (14.7)

Nanotech.

2 (1.30)

3 (1.36)

3 (0.55)

Chem. Other

7 (4.55)

11 (5)

24 (4.41)

Medical Device

9 (5.84%)

11 (5%)

12 (2.21%)

Mechanical Other

4 (2.60%)

9 (4.09%)

17 (3.13%)

Misc. Other Total

3 (1.95%)

5 (2.27%)

5 (0.92%)

154 (100%)

220 (100%)

544 (100%)

sales were made to NPEs.55 That said, despite the concentration of computer- and communication-related assets, the market for academic US patents nonetheless exhibits a greater diversity of technology than the larger brokered market, which is even more heavily dependent on computer hardware and software.56 Second, we observe that relatively few assets have to date been enforced in court or challenged in an administrative proceeding. As of November 2018, 21 of the 544 total assets have been litigated by the entities that purchased them (all but two of which are NPEs).57 Thus, despite the fact that a large share of US patents sold by universities and labs wound up in the hands of NPEs, these assets have rarely been selected for enforcement in court to date. That said, assets that have been enforced in court have generally been enforced numerous times, with litigated assets asserted in an average of 14 cases (and a median of 5). It also seems likely that additional patents, especially those purchased in 2016 and 2017, will be enforced in coming years.58

55 In addition to Thermolife, three other NPEs purchased biopharmaceutical patents, including Intellectual Ventures and the IP Strategy Network, a Japanese sovereign patent fund. Intellectual Ventures also purchased six of the nine total medical device packages that we observed. 56 See Love, et al., supra note 2, at 398 (reporting that approximately 88 percent of packages sold on the brokered market between 2012 and 2016 related to telecommunications, computer hardware, or software). 57 An additional two assets were litigated against the entity that purchased them as part of the case’s settlement. 58 NPEs generally do not immediately enforce the assets that they acquire. See Brian J. Love, An Empirical Study of Patent Litigation Timing: Could a Patent Term Reduction Decimate Trolls without Harming Innovators?, 161 u. pa. l. rev. 1309, 1333 (2013) (finding in a study of litigated patents that NPEs “wait 2.4 additional years on average before filing suit”).

270 Research handbook on intellectual property and technology transfer Table 12.5

Litigation of US patents sold by universities and research institutes (2012–2017)

No. Litigating Buyers

9

No. Litigating NPE Buyers

7

No. Assets Litigated by Buyers*

21

No. Assets Litigated by NPE Buyers

18

No. Asset-Case Pairs**

297

No. NPE Asset-Case Pairs

282

Avg. No. Cases/Asset

14.1

Avg. No. NPE Cases/Asset

15.7

Median No. Cases/Asset

5

Median No. NPE Cases/Asset

5

No. Unique Cases

131

No. Unique NPE Cases

122

Avg. No. Unique Cases/Buyer

14.5

Avg. No. Unique Cases/NPE

17.4

Median No. Unique Cases/Buyer

5

Median No. Unique Cases/NPE

5

Notes: * This total includes some suits filed by the buyer before the sale was recorded. ** Many sold assets were asserted together in the same cases.

VI.

ANALYSIS

Synthesizing the results presented above, we make two broad observations about the market for academic US patents and, finally, consider what our data suggests about patent policy. A.

Foreign Entities, Not US Research Universities, Dominate the Market

Perhaps the most noteworthy fact revealed by our data is that elite US research universities almost entirely refrain from selling their patents to third parties. The transactions that we observed overwhelmingly involved patents awarded to foreign universities and research institutes, which collectively account for over 80 percent of the market volume. Moreover, the fraction of the market that involves sales from US entities is, itself, dominated by universities that are otherwise marginal players in the overall market for technology transfer. Why might this be the case? While our data itself offers little guidance on this topic, we suspect that several factors are at play. First, a large percentage of US university research is funded by the US government, and the Bayh-Dole Act limits universities’ ability to outright transfer patents that cover publicly-funded research. Among other limitations, the US government retains the right to use (and in some instances unilaterally license) the technology, and patents covering US-funded research generally cannot be assigned without the government’s permission.59 According to AUTM survey responses, about two-thirds of all funding for US university research comes

59 For a more detailed list of limitations, see Cahoy, et al., supra note 26, at 58. While universities are generally prohibited from assigning patent rights to federally funded research without government approval, the Bayh-Dole Act provides an exception for assignments “made to an organization which has as one of its primary functions the management of inventions.” 35 U.S.C. § 202(c)(7)(A). While it is generally accepted that this exception was originally drafted to exempt assignments to university-affiliated research foundations (like WARF), some argue that the language should be read to additionally exempt assignments to NPEs on the theory that NPEs’ “primary function[…]” is to “manag[e] … inventions” by filing lawsuits to monetize them. See Jacob H. Rooksby, Innovation and Litigation: Tensions between Universities and Patents and How to Fix Them, 15 yale J.l. & teCh. 312, 329 n.52 (2013) (noting that section 202(c)(7)(A) “would not seem to prevent universities from assigning patent rights to … patent assertion entities … created for the sole purpose of bringing litigation” and reporting that “several TTO directors have acknowledged the existence of such a market”).

US patent sales by universities and research institutes 271 from government sources.60 Thus, it is likely the case that a large percentage of US university portfolios simply cannot be transferred “free and clear.” This may give foreign universities and labs, as well as less prestigious US universities, a comparative advantage in US patent sales. Patents obtained by foreign entities will rarely cover the fruits of US-funded research and, thus, will rarely be constrained by US government restrictions. In addition, less prestigious US universities that attract relatively few government research grants may, despite obtaining few patents overall, benefit in the secondary market from having portfolios that are less encumbered by the Bayh-Dole Act.61 Public relations considerations likely play a role as well. In recent years, patent sales and suits by US research universities have generally been viewed with suspicion and have come under fire in press reports.62 Foreign universities likely face less criticism, especially when they sell assets to US entities that will primarily enforce them against US companies. Similarly, lower ranked US universities may be less sensitive to reputational risks associated with selling patents because they rely less on their reputations to attract students, faculty, and donors. Litigation experience and strategy is yet another potential factor. Many elite US universities have considerable experience in patent litigation, including experience litigating suits that simply monetize patent rights against technology commercializers.63 Universities with prior litigation experience may be more willing and better able to manage subsequent suits without the help of NPEs that specialize in patent monetization. It may also be true that large, elite US universities are viewed more favorably by US jurors than foreign universities or smaller, less well-known US institutions. If so, it may be more advantageous for popular US universities to file suit in their own names and in their own (often insular) hometowns,64 rather than transfer assets to other faceless entities for assertion.

60

See AUTM-STATT, supra note 22. For a ranking of almost 1,000 universities based on federal funding received, see Rankings by Total Federal Obligations, nat’l sCI. found., available at https://ncsesdata.nsf.gov/profiles/site ?method=rankingBySource&ds=fss (last visited Nov. 20, 2018). According to these statistics, the top 25 recipients of federal research funding between 2007 and 2016 collectively received about 41 percent of all federal funding, and the top 50 institutions received about 61 percent. Id. 62 Many universities that filed suit in recent years were labeled “patent trolls” in press reports. See, e.g., Daniel Engber, In Pursuit of Knowledge and Profit: How Universities Aid and Abet Patent Trolls, SLATE, May, 8, 2014, available at http://www.slate.com/articles/technology/history_of_innovation/ 2014/05/patent_trolls_universities_sometimes_look_a_lot_like_trolls.html (last visited Oct. 17, 2019); John Koetsier, Congratulations, Boston University, You’re Now a Patent Troll, venturebeat, July 3, 2013, 12:17pm EDT, available at http://venturebeat.com/2013/07/03/congratulations-boston-university -youre-now-a-patenttroll/; Timothy B. Lee, Patent Trolls Have a Surprising Ally: Universities, wash. post: the swItCh, Nov. 30, 2013, 11:05am EDT, available at http://www.washingtonpost.com/blogs/ theswitch/wp/2013/11/30/patent-trolls-have-a-surprising-ally-universities/. 63 See Brian J. Love, Do University Patents Pay Off? Evidence from a Survey of University Inventors in Computer Science and Electrical Engineering, 16 yale J.l. & teCh. 285, 289–92 (2014) (collecting examples). 64 While this was possible under patent venue rules applicable during the period of our study, today universities would be able to sue in their hometown only alleged infringers that either are incorporated in the same state or have “a regular and established place of business” in the same district. TC Heartland, LLC v. Kraft Foods Grp. Brands, LLC, 137 S.Ct. 1514, 1517 (2017) (holding “that for purposes of [28 U.S.C.] § 1400(b) a domestic corporation ‘resides’ only in its State of incorporation”). 61

272 Research handbook on intellectual property and technology transfer Finally, it seems likely that differing government policies are a contributing factor. A good deal of foreign participation in the market for academic US patents is driven by so-called “sovereign patent funds” (SPFs), publicly-financed patent holding entities established to aggregate domestic patent rights and wield them against foreign companies.65 Between 2012 and 2017, both the most active seller and the most active buyer of academic US patents were SPFs.66 Overall, SPFs sold 16 percent and bought 32 percent of the patents that we studied.67 While the US is home to a number of large, private patent aggregators—a fact that has been cited by foreign governments in response to criticisms of SPFs—none are controlled or subsidized by the US government.68 Government dynamics also likely contribute to the higher rate of patent sales among mid-range US universities. US public funding for higher education decreased markedly— almost 20 percent in real dollars per student—between 2000 and 2017.69 At the same time, US university enrollment has fallen each year since 2011, making it even more challenging for universities to replace public funding with additional tuition.70 The resulting need to create or grow sources of revenue may place disproportionate pressure on mid-tier US universities— which generally attract fewer donations and face stiffer competition for students—to squeeze additional money from their patent portfolios, even if doing so might generate negative publicity. B.

The Academic Patent Market Mostly Transfers Liability, Not Technology

In addition, our data strongly suggests that the market for university patents is primarily a market for the transfer of potential legal liability, not a market for the transfer of technology. As discussed above, over two-thirds of the assignments that we identified transferred academic US patents to either NPEs or defensive patent aggregators, neither of which commercialize technology. In addition, the majority of assignments to operating companies appear to be motivated by companies’ desire to assert or avoid the assets transferred. Few assignments were made to small operating companies, even fewer to true startups, and just 10 percent of assignments appeared to transfer software or knowhow in addition to the (typically years old) patent assets listed therein. In this respect, the market for academic US patents (despite mostly not involving patent brokers) strongly resembles the larger brokered market for patents, which moves patent assets in similar

65

See Xuan-Thao Nguyen, Sovereign Patent Funds, 51 u.C. davIs l. rev. 1257 (2018). The most active seller was Taiwan’s ITRI, and the most active buyer was South Korea’s Intellectual Discovery. Though ITRI operates as a nonprofit “semi-public research center,” it has also been categorized as an SPF because it, for example, runs the IP Bank-Taiwan and has amassed a portfolio of 19,000 patents. See id. at 1268–70. 67 The SPFs that bought patents in our dataset were South Korea’s Intellectual Discovery and IP Cube Partners, and Japan’s IP Strategy Network (which includes the Life-Science Intellectual Property Platform Fund). 68 See Nguyen, supra note 65, at 1260 (“Alarmed by the rise of powerful patent aggregators in the United States, governments from other countries have decided to counter with their own initiatives of aggregating patents through the establishment of Sovereign Patent Funds.”). 69 See State Higher Educ. Exec. Officers Assoc., State Higher Education Finance Report 18, 2018, available at http://www.sheeo.org/sites/default/files/SHEF_FY2017.pdf. 70 Id. 66

US patent sales by universities and research institutes 273 proportions to similar cohorts of buyers.71 Though our data cannot say, it may be the case that the market for academic US patents operates much like the broader brokered market, with tech transfer office personnel playing a role similar to that of patent brokers, including perhaps circulating patents to multiple potential buyers identified as either likely infringers or likely enforcers.72 In addition, the central role that actual or threatened litigation plays in the market suggests that universities engaged in true ex ante technology transfer are both willing and able to accomplish their objectives without selling patent rights. Indeed, yet another reason for the lack of participation by top-tier US research universities may be that their tech transfer offices place a greater focus on maximizing commercialization, not revenue. Stanford University’s Office of Technology Licensing, for example, currently warns inventors “that financial interests are not the primary consideration when making licensing decisions” because the office is tasked with “making licensing decisions that it believes will effectively transfer the technology for society’s benefit.”73 This sentiment echos prior statements issued by the prestigious, 60-member Association of American Universities (“AAU”) in 2015,74 and a group of 11 elite universities in 2007.75 C.

The Academic Patent Market and Patent Policy

Traditionally, technology transfer policy in the US has focused on themes such as technology commercialization, regional economic development, and startup formation. Our data however

71 Indeed, if anything, the academic patent market is even more heavily dependent on NPE buyers than the brokered market. Between 2012 and 2016, NPEs purchased only 37 percent of packages shopped by brokers. Love, et al., supra note 2, at 388. 72 For example, it is not unheard of for TTO employees to transition into careers as patent brokers. See, e.g., news wIre, Thomas Major Joins IPOfferings as Vice President in Charge of Brokerage for Larger Patent Portfolios, Nov. 22, 2010, available at https://www.newswire.com/thomas-major-joins -ipofferings/74535 (noting that Mr. Major was previously “Director of Technology Transfer for the University of Utah”). Anecdotally, at least some universities have circulated patents with “Evidence of Use” documentation, which is common in the brokered market. See GreyB, Patent Licensing–A Way for Universities to Generate More Revenue, Mar. 30, 2016, available at https://www.greyb.com/how -universities-can-generate-revenue/ (urging universities to use “evidence of use charts” more often to “provide evidence that your patented technology has been used by other companies” which “helps prospective buyer [sic.] of your patent get a fair idea about the worth of your patent”). 73 OTL and the Inventor: Roles in Technology Transfer, stan. off. teCh. lICensIng, available at https://otl.stanford.edu/otl-and-inventor-roles-technology-transfer (last visited Nov. 17, 2018). While a transaction from Stanford to an NPE does appear on our list, Stanford released a statement about this transaction explaining that they believed the buyer “would be selling products in the space,” were “not interested in licensing ‘patent trolls’ and have avoided any arrangement with them,” and “were not aware that they intended to institute so many law suits.” Sydell, supra note 32. 74 Statement to the AAU Membership on University Technology Transfer and Managing Intellectual Property in the Public Interest 3, Mar. 2015, available at https://www.eff.org/files/2016/08/17/aau _patent_tech_transfer_working_group_statement.pdf (encouraging AAU members to “[d]evelop procedures and criteria for evaluating a university’s technology transfer units that do not rely solely upon measuring revenue generation, but focus on aligning the work of these units with the research university’s core missions of discovery, learning, and the promotion of social wellbeing”). 75 In the Public Interest: Nine Points to Consider in Licensing University Technology, Mar. 6, 2007, at 8, available at http://news.stanford.edu/news/2007/march7/gifs/whitepaper.pdf (explaining “that universities would better serve the public interest by ensuring appropriate use of their technology by requiring their licensees to operate under a business model that encourages commercialization and does not rely primarily on threats of infringement litigation to generate revenue”).

274 Research handbook on intellectual property and technology transfer suggests that, overwhelmingly, the market for buying and selling academic patents neither directly contributes to the commercialization of new technologies nor directly facilitates the formation of new companies. How, then, should policymakers or university administrators view the market for US patents issued to universities and research institutes? Should US universities be encouraged to aggressively enter the market, as some have suggested?76 Or should US policymakers restrict or even abolish the market entirely, as others propose?77 While our data presents just a static snapshot of academic patent sales, we are skeptical that expanding the market in its current form is compatible with the dominant justifications for university patenting. It is generally accepted that the primary policy goal underlying university tech transfer is to facilitate the commercialization of university research.78 Apart from the role that patents may play in facilitating commercialization, most other theories that support patent ownership generally tend to support university patenting relatively weakly, if at all. For example, because university research is heavily subsidized ex ante by the public,79 universities have less need than private entities to rely on patents to recoup their upfront costs. Also, academic researchers generally report that they are not motivated to research by the prospect of obtaining patents80 and that obtaining patents tends to delay, rather than hasten, the public disclosure of their research.81

76 See, e.g., Jacob H. Rooksby, When Tigers Bare Teeth: A Qualitative Study of University Patent Enforcement, 46 akron l. rev. 185 (2013) (recounting a presentation during AUTM’s 2011 Annual Meeting that “challenged universities to ask themselves, ‘Why do we get patents if we are not willing to enforce them?’” and “[n]ot[ed] that enforcement action is a ‘tremendous untapped revenue source’”). 77 See, e.g., Elliot Harmon, Tell Your University: Don’t Sell Patents to Trolls, (Aug. 17, 2016), available at https://www.eff.org/deeplinks/2016/08/tell-your-university-dont-sell-patents-trolls. In 2017, a bill was introduced in Maryland that if enacted would have prohibited state universities from “assign[ing] or exclusively licens[ing] a patent to a patent assertion entity.” H.B. 1357, 2017 Leg., 437th Sess. (Md. 2017), available at http://mgaleg.maryland.gov/2017RS/bills/hb/hb1357f.pdf. 78 The Bayh-Dole Act itself states that: “It is the policy and objective of the Congress to use the patent system to promote the utilization of inventions arising from federally supported research or development [and] . . . to promote the commercialization and public availability of inventions made in the United States by United States industry and labor…”. 35 U.S.C. § 200. See also Ian Ayres & Lisa Larrimore Ouellette, A Market Test for Bayh-Dole Patents, 102 Cornell l. rev. 271, 276 (2017) (noting “the conventional wisdom … that Bayh-Dole patents are justified only by their commercialization incentive”). 79 See AUTM-STATT, supra note 22. 80 Love, supra note 63, at 316 (reporting that about 70 percent of surveyed faculty members disagreed (approximately 15 percent) or strongly disagreed (approximately 55 percent) with each of the following statements: “The ability to patent my university research encourages me to do MORE research than I would otherwise” and “The ability to patent my university research encourages me to do HIGHER QUALITY research than I would otherwise”). See also Christopher D. Hazuka, Supporting the Work of Lesser Geniuses: An Argument for Removing Obstructions to Human Embryonic Stem Cell Research, 57 u. MIaMI l. rev. 157, 196 (2002) (“[A]cademic scientists, who have driven the revolutionary advances in biomedical science, are not generally motivated by the possibility of obtaining patents. Instead, they seek publication and the esteem of their peers. Indeed, much biotechnology upstream, basic research would take place in the absence of the patent system.”); Arti Kaur Rai, Regulating Scientific Research: Intellectual Property Rights and the Norms of Science, 94 nw. u. l. rev. 77, 89–90 (1999) (observing that norms in the scientific research community “promote a public domain of freely available scientific information” and eschew “claiming property rights in invention … as immoral”). 81 Love, supra note 63, at 323 (reporting that, among surveyed faculty members who expressed an opinion one way or the other, about three-quarters (or 30 percent of all respondents) indicated that university patenting “HINDERS the dissemination of new research because, e.g., faculty members keep their research secret and/or delay publication until after filing a patent application,” rather than “HELPS

US patent sales by universities and research institutes 275 Unfortunately, little in our data suggests that academic patent sales have a net positive impact on technology commercialization. As discussed above, the vast majority of the market is based upon transferring legal liability, not technology, and in most instances those transfers are made to NPEs. Though there is a lack of consensus on the topic, many have concluded that NPE patent monetization has a net negative effect on innovation.82 Cohen et al. find that NPE suits have a “negative impact on innovation activity at targeted firms,”83 and Bessen et al. estimate that patent litigation costs have generally exceeded patent rents since 1999 due to the rise of NPEs.84 Moreover, even when patents are not sold to NPEs, we see little evidence that academic patent sales facilitate commercialization. While some sales likely do assist commercializers, most non-NPE sales that we observe appear to be defensive and offensive acquisitions by large, established operating companies that very likely independently developed the patented technology long before the transfer took place. If this reading of the facts is correct, it is hard to see how such transactions encourage commercialization. To the contrary, transferring patent rights to an existing commercializer (under implicit threat of otherwise enforcing against them in court, or selling the assets to an NPE or competitor to enforce) would seem far more likely to discourage than encourage the commercialization of technology that might be patented by an academic institution.85 Indeed, Ian Ayres and Lisa Ouellette have argued that universities should be permitted to transfer exclusive patent rights only after first “offer[ing] the invention under a nonexclusive license for a nominal fee.”86 Accordingly, our data suggests that policymakers who agree with the narrow commercialization goals underlying the passage of the Bayh-Dole Act, as well as university administrators who agree with the sentiments expressed in recent statements by the AAU and other university groups, should continue to question whether academic patent sales serve the public interest and, in particular, may wish to take further action concerning sales to NPEs.

disseminate new research because, e.g., researchers read patents to learn about new research they wouldn’t otherwise be exposed to”). See also David Schwartz, 5 Things Tech Transfer Offices Wish Their Start-ups Knew, teCh transfer enews, June 5, 2013, available at http://techtransfercentral.com/2013/ 06/05/5things-tech-transfer-offices-wish-their-start-ups-knew/ (“[D]on’t publish your work in any way unless you know it’s safe from an IP protection standpoint.”); Vid Mohan-Ram, Patent First, Publish Later: How Not to Ruin Your Chances of Winning a Patent, sCI., Oct. 26, 2001, available at http://www .sciencemag.org/careers/2001/10/patent-first-publish-later-how-not-ruin-your-chances-winning-patent. 82 See Mark A. Lemley & Robin Feldman, Patent Licensing, Technology Transfer, and Innovation, 106 aM. eCon. rev. 188 (2016). 83 Lauren Cohen, Umit G. Gurun, & Scott Duke Kominers, The Growing Problem of Patent Trolling, 352 sCI. 521 (2016). See also Lauren Cohen, Umit G. Gurun, & Scott Duke Kominers, “Empirical Evidence on the Behavior and Impact of Patent Trolls: A Survey” in Patent Assertion Entities and Competition Policy 27 (D. Daniel Sokol, ed., 2017); Catherine Tucker, Patent Trolls and Technology Diffusion: The Case of Medical Imaging (Working Paper, 2014), available at https://pdfs .semanticscholar.org/e22b/c09be54ee671bf1a67547fd8f8e2ee48783e.pdf. 84 James E. Bessen, Peter Neuhausler, James L. Turner, & Jonathan W. Williams, Trends in Private Patent Costs and Rents for Publicly-Traded United States Firms (Working Paper, 2018), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2278255. 85 See Ayres & Ouellette, supra note 78, at 276–7 (explaining that the “commercialization rationale for exclusive patents on publicly funded inventions makes little sense when numerous firms are already eager to use the inventions without exclusivity”). See also Rebecca S. Eisenberg, Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research, 82 va. l. rev. 1663, 1710 (1996); Rai, supra note 80, at 120, 135. 86 Ayres & Ouellette, supra note 78.

276 Research handbook on intellectual property and technology transfer

VII.

APPENDIX

Table 12A.1

Assignors (Sellers) No of

No. Recorded

Purchases

“Sales”

Assignments

Private Univ.

Defensive

1

2

2

Private Nonprofit

Monetization

1

1

2

Country

Entity Type

Ajou Univ.

Korea US

Alfred E Mann Found. For Sci. Res.

No. of

Likely Context(s) of

Entity Name

Transferred Assets

Arizona State Univ.

US

Public Univ.

Defensive, Monetization

3

6

41

Battelle Memorial Inst.

US

Private Nonprofit

Monetization

1

3

15

Boston Univ.

US

Private Univ.

Defensive

2

2

3

Cardiff Univ.

UK

Public Univ.

Offensive

1

6

6

Carnegie Mellon Univ.

US

Private Univ.

Offensive

1

1

1

Taiwan

Private Univ.

Defensive

1

1

1

US

Private Univ.

Monetization

1

1

1

Korea

Public Nonprofit

Monetization

1

1

4

US

Public Univ.

Defensive

1

1

1

US

Private Univ.

Defensive

1

1

1

Korea

Public Univ.

4

7

9

Japan

Private Univ.

Monetization

1

1

1

Hanyang Univ.

Korea

Private Univ.

Monetization

6

8

9

Hiroshima Univ.

Japan

Public Univ.

Defensive

1

2

4

Hokkaido Univ.

Japan

Public Univ.

Monetization

1

1

1

Hong Kong

Public Univ.

Monetization

3

3

9

US

Private Univ.

3

3

10

Monetization

16

21

88

Public Nonprofit

Commercialization

1

1

1

Public Nonprofit

Monetization

1

3

3

Chung Yuan Christian Univ. Cornell Univ. Elec. & Telecomm. Res. Inst. Florida Atlantic Univ. George Washington Univ. Gwangju Inst. of Sci. & Tech. Hamamatsu Univ. Sch. of Med.

Hong Kong Univ. of Sci. & Tech. Illinois Inst. of Tech. Industrial Tech. Res. Inst. Inst. of Geological & Nuclear Sci. Japan Sci. & Tech. Agency

Taiwan New Zealand Japan

Private Nonprofit Defensive,

Commercialization, Monetization

Commercialization, Monetization

US patent sales by universities and research institutes 277

Entity Name

Country

Entity Type

Likely Context(s) of Purchases

No of “Sales”

No. Recorded Assignments

No. of Transferred Assets

Katholieke Universiteit

Belgium

Private Univ.

Commercialization

1

1

1

Germany

Public Univ.

Defensive

1

1

1

UK

Public Univ.

Monetization

1

1

3

Korea

Public Univ.

Monetization

5

5

14

Korea

Private Nonprofit

Monetization

3

3

24

Korea

Public Nonprofit

Monetization

2

2

19

Korea

Public Nonprofit

Monetization

1

1

11

Korea

Public Nonprofit

Monetization

1

1

1

Korea

Private Univ.

Monetization

2

5

8

Korea

Private Univ.

Monetization

2

2

4

Japan

Public Univ.

Monetization

1

1

1

Kyung Hee Univ.

Korea

Private Univ.

Monetization

2

3

5

Kyushu Univ.

Japan

Public Univ.

Monetization

2

3

9

US

Public Univ.

Commercialization

2

2

12

US

Public Univ.

Monetization

1

1

1 4

Leuven Kiel Univ. King's College London Korea Adv. Inst. of Sci. & Tech. Korea Elec. Tech. Inst. Korea Inst. of Mach. & Materials Korea Institute of Sci. & Tech. Korea Res. Inst. of Biosci. & Biotech. Korea Univ. Kwangwoon Univ. Kyoto Prefectural Univ.

Louisiana State Univ. Louisiana Tech Univ.

US

Private Nonprofit

Monetization

1

1

Nagoya Univ.

Mitre Corp.

Japan

Public Univ.

Defensive

1

1

1

Nanjing Univ.

China

Public Univ.

Monetization

1

1

1

Japan

Public Nonprofit

Defensive

1

1

1

Canada

Public Nonprofit

Commercialization

1

1

14

Taiwan

Public Univ.

4

13

27

US

Public Univ.

2

4

4

Japan

Public Univ.

2

2

3

UK

Public Univ.

1

3

3

US

Private Univ.

2

2

2

Korea

Private Univ.

1

1

1

National Inst. for Materials Sci. National Res. Council of Canada National Tsing Hua Univ. North Carolina State Univ. Osaka Univ. Queen Mary Univ. Rochester Inst. of Tech. Sejong Univ.

Defensive, Monetization Defensive, Monetization Defensive, Monetization Monetization Defensive, Monetization Monetization

278 Research handbook on intellectual property and technology transfer

Entity Name Seoul National Univ. Shanghai Jiao Tong Univ. Shinshu Univ. Southwest Res. Inst. Stanford Univ. Sungkyunkwan Univ. Taipei Medical Univ. Tel Aviv Univ. Tokyo Inst. of Tech. Tokyo Med. & Dental Univ. Univ. of Alabama Univ. of Arkansas Univ. of California Univ. of Cincinnati Univ. of Colorado Univ. of Delaware Univ. of Electro-Comm’ns Univ. of Leeds Univ. of Liverpool

Country

Entity Type

Korea

Public Univ.

China

Public Univ.

Japan

Public Univ.

US

Private Nonprofit

US

Private Univ.

Likely Context(s) of Purchases Defensive,

No of “Sales”

No. Recorded Assignments

No. of Transferred Assets

3

3

5

Monetization

5

10

11

Monetization

1

1

1

3

3

4

1

1

8

3

3

38

Monetization

Defensive, Monetization Monetization Defensive,

Korea

Private Univ.

Taiwan

Private Univ.

Defensive

1

1

1

Israel

Public Univ.

Commercialization

1

1

1

Japan

Public Univ.

Monetization

4

5

5

Japan

Public Univ.

Monetization

1

1

1

US

Public Univ.

Monetization

1

4

4

US

Public Univ.

Monetization

1

1

1

US

Public Univ.

Monetization

1

1

5

US

Public Univ.

Monetization

1

5

6

US

Public Univ.

2

2

15

US

Public Univ.

Defensive

1

1

1

Japan

Public Univ.

Defensive

1

1

1

UK

Public Univ.

Commercialization

1

1

1

UK

Public Univ.

Monetization

1

2

2

US

Public Univ.

Defensive

1

1

2

UK

Public Univ.

Monetization

1

2

2

Spain

Public Univ.

Monetization

1

1

3

Australia

Public Univ.

Monetization

12

21

22

Australia

Public Univ.

Monetization

1

1

1

UK

Public Univ.

Monetization

1

1

1

Monetization

Commercialization, Offensive

Univ. of Louisiana at Lafayette Univ. of Manchester Univ. of Murcia Univ. of New South Wales Univ. of Newcastle Univ. of Nottingham Univ. of Texas

US

Public Univ.

Monetization

1

1

1

Univ. of the Arts

US

Private Univ.

Defensive

1

1

1

Univ. of Tokyo

Japan

Public Univ.

Monetization

1

1

1

US patent sales by universities and research institutes 279

Entity Name

Likely Context(s) of

No of “Sales”

No. Recorded

No. of Transferred

Country

Entity Type

Univ. of Toyama

Japan

Public Univ.

Monetization

1

1

1

Univ. of Tsukuba

Japan

Public Univ.

Monetization

1

1

1

Australia

Public Univ.

Commercialization

1

1

6

Univ. of Western Australia Utah State Univ. Waseda Univ.

Purchases

Assignments

Assets

US

Public Univ.

Offensive

1

1

2

Japan

Private Univ.

Monetization

1

1

1

280 Research handbook on intellectual property and technology transfer Table 12A.2

Entity Name* 22nd Century Grp Inc Acer Inc Allergan Inc Allied Security Trust Amazon Inc

Assignees (Buyers)

Country

US

Entity Type**

Small Op Co

Likely Context of Purchase(s) Commercialization

No of “Sales”

1

No. Recorded Assignments 1

No. of Transferred Assets 14

Taiwan

Large Op Co

Defensive

7

8

17

US

Large Op Co

Commercialization

1

1

14

Defensive

2

2

3

US

Defensive Aggregator

US

Large Op Co

Defensive

1

1

1

Taiwan

Large Op Co

Defensive

2

2

11

BaroFold Inc

US

Small Op Co

Defensive

1

1

1

Cargill Inc

US

Large Op Co

Commercialization

1

1

1

US

Small Op Co

Offensive

1

1

2

UK

NPE

Monetization

1

2

2

US

Large Op Co

Offensive

1

1

1

Netherlands

Small Op Co

Defensive

1

1

1

Seychelles

NPE

Monetization

2

2

8

Taiwan

Large Op Co

Defensive

1

9

9

US

NPE

Monetization

1

3

15

US

Small Op Co

Defensive

1

1

1

US

NPE

Monetization

1

1

5

BVI

NPE

Monetization

1

4

4

US

NPE

Monetization

1

1

7

US

NPE

Monetization

1

1

1

US

Large Op Co

Defensive

1

2

2

Taiwan

Small Op Co

Defensive

1

3

3

Korea

Small Op Co

Commercialization

1

1

1

Korea

NPE

Monetization

29

38

154

US

NPE

Monetization

27

44

47

US

NPE

Monetization

1

1

3

Korea

NPE

Monetization

1

1

1

AU Optronics Corp

ClearOne Comm’ns Inc Clifton Cowley Cross Match Tech Inc DyeCoo Textile Sys BV Dynamic Invention LLC E Ink Corp Element 26 Tech LLC ENCO Systems Inc Eolas Tech Inc Forwarding Tech Ltd Garrett Consulting Ltd LLC Geospatial Tech Assoc LLC Google Inc Innovative Turnkey Sol’n Corp INTEKPLUS Co Intellectual Discovery Co. Intellectual Ventures Interdigital IP Cube Partners Co

US patent sales by universities and research institutes 281

Entity Name* IP Strategy

Country

Entity Type**

Likely Context of Purchase(s)

No of “Sales”

No. Recorded Assignments

No. of Transferred Assets

Japan

NPE

Monetization

14

17

17

Australia

NPE

Monetization

1

1

1

Taiwan

NPE

Monetization

2

3

9

Life Tech Corp

US

Large Op Co

Defensive

1

4

23

Lifecell Corp

US

Large Op Co

Offensive

1

1

1

Medico Int’l Inc

US

Small Op Co

Commercialization

1

1

1

US

Large Op Co

Commercialization

1

1

1

Network iQNovate Ltd Kaituozhe IP Mgmt Consultants

Mettler Toledo LLC Micoba LLC

US

NPE

Monetization

1

1

1

Microsoft Corp

US

Large Op Co

Defensive

1

1

1

US

Large Op Co

Offensive

1

6

6

Korea

NPE

Monetization

2

2

2

Japan

Large Op Co

Defensive

1

1

1

Japan

Large Op Co

Defensive

1

1

1

US

NPE

Monetization

1

3

3

US

NPE

Monetization

1

1

4

Defensive

1

1

1

Myriad Genetics Inc NEWRACOM Inc Nichia Corp Nissan Chem Indus Ltd Nytell Software LLC Oakham Tech LLC Open Invention Network OSSimTech Inc

US

Defensive Aggregator

Canada

Small Op Co

Defensive

2

2

2

UK

NPE

Monetization

1

2

2

ProteoVec LLC

US

Small Op Co

Commercialization

1

1

11

Prowire LLC

US

NPE

Monetization

1

2

2

Quarkstar LLC

US

NPE

Monetization

1

5

6

Rakuten Inc

Japan

Large Op Co

Defensive

3

4

7

Raytheon Co

US

Large Op Co

Commercialization

1

1

6

Defensive

3

3

5

Plant Bioscience Ltd

RPX Corp Samsung Elec Co.

US

Defensive Aggregator

Korea

Large Op Co

Defensive

3

3

3

US

NPE

Monetization

1

1

2

US

Small Op Co

Commercialization

2

2

9

US

Small Op Co

Defensive

1

1

2

Taiwan

Small Op Co

Defensive

1

2

50

UK

NPE

Monetization

1

1

1

Santa Monica Semiconductor LLC Software Motor Co Sotera Wireless Inc Star Trend Enter Corp Sybre Ltd

282 Research handbook on intellectual property and technology transfer

Entity Name* Telescope Time Inc Tessera Thermolife Int’l LLC Transpacific IP Ltd Uniloc Corp Ventana Med Sys Inc Virdia Inc

Country

Entity Type**

Likely Context of Purchase(s)

No of “Sales”

No. Recorded Assignments

No. of Transferred Assets

US

NPE

Monetization

1

1

1

US

NPE

Monetization

4

4

10

US

NPE

Monetization

1

1

8

Singapore

NPE

Monetization

1

1

9

US

NPE

Monetization

1

1

3

US

Large Op Co

Commercialization

1

1

1

US

Small Op Co

Commercialization

1

1

1

Notes: * Transactions involving subsidiaries have been consolidated under the name of their parent. A small number of transactions that quickly passed through an intermediary have also been consolidated under the name of the ultimate recipient. ** Here, we define a “small” operating company as one with less than 500 employees or less than $100 million in annual revenue.

13. Intellectual property exchanges and auctions: non-traditional mechanisms for technology transfer Daniel R Cahoy

I.

INTRODUCTION

Technology transfer through the sale and licensing of intellectual property (“IP”) exists amidst a constantly evolving rights environment. In the early 20th century, it became clear that blocking patents could create roadblocks impacting the development of entire industries.1 Weak enforcement of patents in the middle of the century depressed interest in transfer as industry found it easier to ignore rights.2 In the 1980s, the creation of the Court of Appeals for the Federal Circuit and the enactment of the Bayh-Dole Act conferred increased value to technology rights, and more sophisticated valuation techniques allowed firms to take greater advantage.3 At the dawn of the 21st century, the pendulum continued to swing in favor of IP power as “assertion entities” emerged to take advantage of favorable rules to threaten firms for nuisance payments.4 And most recently, uncertain subject matter rules5 and realignment of litigation advantages6 have tilted the power balance yet again, inserting new variables in the freedom to operate equation.

1

See, e.g., Herbert A. Johnson, The Wright Patent Wars and Early American Aviation, 69 J. aIr l. & CoM. 21, 38 (2004) (describing the aviation industry in view of the Wright Brothers’ patents and stating, “[o]verall progress of aviation suffered from the inability, or unwillingness, of the Wrights to exploit their patent rights and to commence broad scale production of the airplane.”). 2 Rochelle Cooper Dreyfuss, The Federal Circuit: A Case Study in Specialized Courts, 64 N.Y.U. L. rev. 1, 6 (1989) (detailing conclusions by the Hrushka Commission that lead to the creation of the Federal Circuit, including that the “research community considered the value of patents to be in decline.”). 3 David C. Mowery & Bhaven N. Sampat, The Bayh-Dole Act of 1980 and University-Industry Technology Transfer: A Model for Other OECD Governments?, 30 J. teCh. transfer 115, 119 (2005) (noting the impact of both the Bayh-Dole Act and the establishment of the Federal Circuit as aspects of a broad “shift in U.S. policy toward stronger intellectual property rights”). 4 Exec Office of the President, Patent assertion and U.S. Innovation 3–4 (2013) (describing the “aggressive litigation practices” used by patent assertion entities or “trolls”). 5 Christopher M. Holman, Patent Eligibility Post-Myriad: A Reinvigorated Judicial Wildcard of Uncertain Effect, 82 geo. wash. l. rev. 1796, 1816 (2014); PTO, Roundtable on Patent Subject Matter Eligibility, Dec. 5, 2016, 406–7. 6 A number of changes to US law have made so-called patent trolling more difficult. This includes the Supreme Court’s decision restricting venue shopping, TC Heartland LLC v. Kraft Foods Group, 137 S.Ct. 1514 (2017), and its ruling that relaxed the rules for fee shifting in response to abusive plaintiffs, Octane Fitness, LLC v. Icon Health & Fitness, Inc., 135 S.Ct. 1749 (2014), as well as legislation that limited the ability of plaintiffs to join multiple defendants in a single suit. Leahy-Smith America Invents Act, Pub. L. No. 112-29, 125 Stat. 284 (2011) (codified in scattered sections of 35 U.S.C.).

283

284 Research handbook on intellectual property and technology transfer Firms must make decisions on licensing, sales and acquisition in the midst of these shifting sands, looking for a path to market protection and possibly an advantage against competitors. Not surprisingly, technology transfer as a field is under pressure to evolve as well, as firms struggle to make intelligent decisions with available assets. Traditionally, firms might be able to identify the owners of the most important technologies and deal directly with them. Either preemptively or as a consequence of litigation, technology owners and interested users could have some confidence in meeting minds at some relatively early point in the innovation process. However, this simple system is inadequate for an increasing number of modern technology transfer contexts. The greatest problem with the bilateral model of licensing or sale is that too much information is missing from the participants. Identifying the most important rights, particularly in the most cutting-edge fields with thicket potential, is difficult.7 And even if such rights are discovered, it is difficult for firms to come together on value assessments. Essential elements of past licensing or sales transactions—particularly price—are often confidential,8 and it can be difficult to determine baseline numbers. Additionally, because technology assets covered by patents are by definition unique in critical ways, true “comparables” do not exist.9 A past sale by the same firm in the same technology area may be entirely irrelevant if the inventions are different in scope. Additionally, there can be complementary attributes to different technologies such that the value of one is dependent on one or more other technologies.10 A single patent may be of limited value without a portfolio of complementary patents.11 To fill in such information gaps and reduce asymmetries, more complex mechanisms have emerged to reshape the marketplace. The late 19th and early 20th century witnessed the formation of pooling arrangements as an early solution to identification and pricing problems.12 Later, standard setting organizations attempted to coordinate IP interests related to specific technological solutions.13 However, recent years have given rise to a virtual explosion of new

7

See Daniel R. Cahoy & Leland Glenna, Private Ordering and Public Energy Innovation Policy, 36 fla. st. u. l. rev. 415, 446–8 (2009) (describing as one of the conditions for private ordering or transfer/consolidation of patent rights the ability to identify the relevant field of patents); Mario Benassi & Alberto Di Minin, Playing in Between: Patent Brookers in Markets for Technology, 30 r&d MgMt. 68, 76–7 (2009) (describing conditions for a patent secondary market to develop, including patenting intensity and IP “blindness”). 8 See Mark A. Lemley & Nathan Myhrvold, How to Make a Patent Market, 36 hofstra l. rev. 257, 257 (2007) (describing the “blind market’ that exists for patent transactions in contrast to stocks); Andrei Hagiu & David B. Yoffie, The New Patent Intermediaries: Platforms, Defensive Aggregators, and Super-Aggregators, 27 J. eCon. perspeCtIves 45, 45–6 (2013). 9 Hagiu & Yoffie, supra note 8, at 46–7. 10 Id. at 47. 11 See Gideon Parchomovsky & R. Polk Wagner, Patent Portfolios, 154 u. pa. l. rev. 1, 42 (2005) (“firms patent heavily to maximize the benefits of patent portfolios, and such benefits are directly determined by the quantity of patents assembled.”). 12 Gavin Clarkson & Joshua Newberg, Blunt Machetes in the Patent Thicket: Modern Lessons from the History of Patent Pool Litigation in the United States Between 1900 and 1970, 22 J. teCh. l. & pol’y 1, 2 (2018) (describing the early history of patent pools in the United States). 13 See George S. Cary, Paul S. Hayes & Larry C. Work-Dembowski, Antitrust Implication of Abuse of Standard Setting, 15 geo. Mason l. rev. 1241, 1242–3 (2008) (describing the role of patents in standards and the need to facilitate access). See generally Jorge L. Contreras, A Brief History of FRAND: Analyzing Current Debates in Standard Setting and Antitrust Through a Historical Lens, 80 antItrust L.J. 39 (2015).

Intellectual property exchanges and auctions 285 techniques and systems for connecting willing buyers and sellers in ways that permit monetization and exchange outside of wasteful litigation. Considered at its most extensive parameters, transfer mechanisms beyond bilateral negation can include any marketplace, exchange or auction, whether run by a third party or by an owner or purchaser/licensor of rights. A myriad of new options exists for IP owners and it is fair to say that the technology transfer environment is transforming more quickly than in the past. To some extent, this has been covered in the literature in the context of “intermediaries” who facilitate new mechanisms14 or the perceived “commoditization” of IP driving them.15 Some would argue that recent developments were more of a fad and the failure of some models indicates nothing has changed. This is shortsighted. Non-traditional mechanisms, which are broader than the subset that is intermediary-administered, are undergoing alignment in a quest to prove value, for sure. But this new way of interacting is here to stay. This Chapter will describe non-traditional mechanisms for technology transfer, with a goal of evaluating the success of recent players and providing some rules for assessing the future of this active marketplace. Although the discussion could be broadened to all technology-related IP, including at least trade secrets and know-how, it will primarily address patents as the form most subject to standardized transfer mechanisms. The assessment concludes that it is likely that the growth of technology rights along with big data systems will only fuel the growth of alternative exchange mechanisms and foster more creativity and experimentation in the future.

II.

EARLY SYSTEMS FOR ORGANIZED TECHNOLOGY MARKETS

It is likely that soon after patents emerged in the historical record, sharing the rights through licenses existed.16 The very idea of a patent—an open “letters patent” describing the invention and who created it17—by its nature facilitates commercial exchange. Eventually this exchange came to include the wholesale conveyance of the right to another. However, a one-to-one system of transfer was not always sufficient as technologies became more complicated and inventions involved more than one element, perhaps created at different times by different people. This led to one of the first problems encountered in technology licensing that thwarted

14

See, e.g., Hagiu & Yoffie, supra note 8; Allen W. Wang, Rise of the Patent Intermediaries, 25 berkeley teCh. l.J. 159 (2010); Benassi & Di Minin, supra note 7, at 68. 15 See, e.g., Richard C. Bulman, Devining the Future of Intellectual Property, 42 les nouvelles 441, 441–2 (2007) (detailing IP commoditization in the context of auctions and insurance); James Thuo Gathii, Construing Intellectual Property Rights and Competition Policy Consistent with Facilitating Access to Affordable AIDS Drugs to Low-End Consumers, 53 fla. l. rev. 727, 749–51 (2001) (describing the “commodity logic” for IP that is embedded in TRIPs). 16 The Venetian Patent Act of 1474 is an early example of invention protection that appeared to permit a patent holder to license to others. Edward C. Walterscheid, The Early Evolution of the United States Patent Law: Antecedents (Part 1), 76 J. pat. & tradeMark off. soC’y 697, 709–10 (1994). Notably, the Venetian Act was a positive privilege from the state to practice rather than a negative right to exclude. Ted Sichelman & Sean O’Connor, Patents as Promoter of Competition: The Guild Origins of Patent Law in the Venetian Republic, 49 san dIego l. rev. 1267, 1277–8 (2012). Still, the ability to open that privilege to others was important, and it eventually expanded to a right to exclude. Id. at 1269. 17 Walterscheid, supra note 16, at 701.

286 Research handbook on intellectual property and technology transfer bilateral licensing: diverse ownership of patents necessary to produce the commercially viable product. In other words, thickets.18 An important historical example of this diverse ownership issue is the invention of the sewing machine in the 1850s. A machine that could speed up the process of garment sewing was perceived as an important need in the early days of the industrial revolution.19 Early inventions, such as those made by Howe, had elements of practical sewing machines, but also problems that prevented wide-spread adoption.20 Eventually, Singer was able to identify and combine all of the essential parts of a viable machine and sales took off.21 However, the owners of patents covering parts of this comprehensive machine sought to redeem the fruits of their labors, and multiple litigations ensued. Eventually, lawsuits depressed sales and threatened the entire industry.22 To resolve their disputes, equitably share royalties and ensure that any manufacturer could identify and access the most important rights, patent owners formed the Sewing Machine Combination.23 This is by most accounts the first patent “pool.” A patent pool is a structure that allows for the licensing of technology that reads on multiple rights.24 Although one commonly thinks about such pools as being a recent solution and need, their presence in historic technologies, such as airplanes and radio in addition to sewing machines, establish that they have been understood for well over a century. Modern examples of pools include the MPEG LA consortium and the Radio Frequency ID (RFID) consortium.25 In addition to enabling a particular technology form, pools may gather technologies for a social goal. A primary example of this model is the Medicines Patent Pool, promulgated by Unitaid, with the goal of enabling the production of low-cost HIV, hepatitis C and tuberculosis medicines in low-income countries.26 In essence, pools reduce innovation uncertainty. They solve the problem of multiple negotiations, patent hold-up and the potential for “royalty stacking,” which can impede the commercialization of a technology.27 By collecting the rights and providing defined royalties, they ensure that both innovators and low-cost competitors can make use of a core technology. They establish a price that can be figured into production costs. They facilitate technology adoption.28 18

Michael A. Heller, The Gridlock Economy: How Too Much Ownership Wrecks Markets, Stops Innovation, and Costs Lives 43 (2008) (explaining the origin and meaning of the phrase “patent thicket”). 19 Adam Mossoff, The Rise and Fall of the First American Patent Thicket: The Sewing Machine War of the 1850s, 53 arIz. l. rev. 165, 172 (2011). 20 Id. at 175. 21 Id. at 179–80. 22 Id. at 182. 23 Id. at 194–5. 24 Erik Hovenkamp & Herbert J. Hovenkamp, “Patent Pools and Related Technology Sharing” in Cambridge Handbook of Antitrust, Intellectual Property and High Tech 358 (Roger D. Blair & D. Daniel Sokol, eds., 2017). 25 Id. at 364; Richard J. Gilbert, Ties that Bind: Policies to Promote (Good) Patent Pools, 77 antItrust L.J. 1, 6 (2010) (referring RFID pool). 26 License Overview, MedICInes patent pool, https://medicinespatentpool.org/what-we-do/global -licence-overview/ (last visited Sept. 25, 2018). 27 See Mark A. Lemley & Carl Shaprio, Patent Holdup and Royalty Stacking, 85 tex. l. rev. 1991, 2010–11, 2014–15 (2007) (explaining why royalty stacking is a problem for firms and how pools play a role in addressing it). 28 See Gilbert, supra note 25, at 3 (“The underuse caused by a patent thicket can harm patent owners as well as consumers if collective royalty demands or the transaction costs from licensing many patents slow the adoption or impede the use of products covered by the patented technologies.”).

Intellectual property exchanges and auctions 287 On the other hand, pools also place limits on individual owners to collect the maximum return for their invention because they present all rights as a united product. This is arguably more problematic if a patent owner is not a manufacturer and cannot benefit from licensing from the pool. For example, in the Sewing Machine Combination, one of the non-manufacturing initial patentees was compelled to negotiate an additional royalty stake to account for the fact that other manufacturers could profit from machine sales.29 The potential for patent pools to create controversy related to collusion is a constant problem. Some of the earliest voluntary pools attracted scrutiny, including the famous Motion Picture Patents Company—also known as the Edison Trust30—the avoidance of which was responsible for bringing the movie industry to Hollywood.31 Department of Justice guidelines provide the outlines of permissible patent pools.32 In general, a pool consisting of complementary patents is more likely to be permitted than one of substitute (competing) patents.33 An established mechanism with attributes similar to pools are standard setting organizations (“SSOs”). These entities allow innovators to agree on a technology form that will become the basis for other products, avoiding wasteful competition when consumers would otherwise be ambivalent.34 Although the standard is not in and of itself an owned technology, it may incorporate owned technologies from one or more participants. Such initiatives have a checkered history of patent owners attempting to extort participants after the standard is agreed upon, leading to either hold-ups or inflated royalty pricing.35 To avoid this problem, modern standard setting organizations often require an agreement among members to license proprietary technology on a “fair and reasonable” (“FRAND”) basis.36 This seems like a fair exchange for adopting a standard that requires use of the owned technology. As with a pool, a FRAND-licensed patent avoids the information problems that might occur if an owner remained independent. Users can predict pricing and availability of technology and incorporate the expense into product development. And, similar to pooled rights, the owner gives up some measure of control to facilitate widespread adoption. Due to the heavy-handed approach of pooling and standards, regulators have concerns about antitrust effects with standards as well.37 By working together even in cross-licensing, tech-

29

Mossoff, supra note 19, at 195. Clarkson & Newberg, supra note 12, at 26–7; 31 Jon M. Garon, Content, Control, and the Socially Networked Film, 48 U. louIsvIlle l. rev. 771, 774–5 (2010). 32 U.S. Dep’t of Justice & Fed. Trade Comm’n, Antitrust Enforcement and Intellectual Property Rights: Promoting Innovation and Competition 66–74 (2007). 33 Id. at 74–6. 34 See Mark A. Lemley, Intellectual Property Rights and Standard Setting Organizations, 90 CalIf. l. rev. 1889, 1896–7 (1992) (describing the role of standards and standard setting organizations). 35 See Joel M. Wallace, Note, Rambus v. F.T.C. in the Context of Standard-Setting Organizations, Antitrust, and the Patent Hold-Up Problem, 24 berkeley teCh. l.J. 661, 677–8 (2009) (relating the story of Rambus, Inc’s efforts to assert patents against competitors using industry standards for computer memory). 36 Timo Ruikka, “FRAND” Undertakings in Standardization: A Business Perspective, 43 les nouvelles 188, 189 (2008) (describing the meaning of FRAND commitments or undertakings). 37 George S. Cary, Mark W. Nelson, Steven J. Kaiser, & Alex R. Sistla, The Case for Antitrust Law to Police the Patent Holdup Problem in Standard Setting, 77 antItrust l.J. 913, 921 (2011) (“[T]here is little debate that the activity of SSOs (and their members) can raise serious anticompetitive issues, which may—in certain cases—violate Sections 1 and/or 2 of the Sherman Act.”). 30

288 Research handbook on intellectual property and technology transfer nology owners are, to some extent, colluding and sharing in joint use of their inventions. For that reason, the US and European antitrust regulators scrutinize these arrangements. They may, in some cases, preclude or invalidate the collaboration.38 Such scrutiny makes these historic alternatives to bilateral agreements somewhat dangerous and therefore somewhat rare. An emerging concern related to standards is the converse bargaining problem, in which standard setting organizations employ reverse “hold-up” power against patent owners.39 In other words, SSOs use the power in competitors determining the final standard to force individual patent owners to accept a low royalty or risk exclusion. To date, regulators have been reluctant to investigate this concern40 as they tend to treat potentially relevant technologies as available, rather than created in anticipation of use in a standard. From the firm’s view, exclusion is an investment disincentive rather than inclusion being a windfall. Pools and standards are well characterized mechanisms for reducing the uncertainty associated with identifying necessary technology and respective owners. But they apply in a somewhat narrow range of circumstances. Particularly, they require the existence of an overall technology on which a set of patents will overlay. Patent owners with less of an organized need of collaboration, but the desire for a system stronger than bilateral licensing, will seek out other solutions. In this space, many new systems of exchange and consolidation have flourished.

III.

RISE OF THE PATENT EXCHANGE AND OTHER CONSOLIDATORS

To be clear, patent licensing models have always existed, and even early forms of patent aggregators played a role in technology diffusion.41 But, arguably, the environment is evolving more quickly. After many years of conventional and rather staid licensing practices, the IP world has witnessed a creative explosion of new systems to bring together prospective owners and users. It is likely that a number of factors underlie this change. One is certainly the massive growth of IP rights, particularly patents, over the last twenty years. The number of patents issued each year in the US has increased over five-fold since the 1980s, and there are likely nearly three million patents currently in force (about twelve million globally).42 Such numbers mean that it has become increasingly harder to monitor a given field, even with the use of increasingly sophisticated computer tools. The increasing number of players is also an important factor. IP ownership is a global phenomenon, and the rapid growth of new environments such as in China make traditional

38

Cary, et al., supra note 13, at 1242–5. See J. Gregory Sidak, Patent Holdup and Oligopsonistic Collusion in Standard Setting Organizations, 5 J. CoMpetItIon l. & eCon. 123, 142–6 (2009) (describing the collusive power of SSO members to extract low patent royalties). 40 Id. at 164–71. 41 Michael Risch, Licensing Acquired Patents, 21 geo. Mason l. rev. 979, 979 (2014). 42 World Intell. Prop. Org. (WIPO), World Intellectual Property Indicators 2017, at 38 (“The estimated number of patents in force worldwide rose … to 11.8 million in 2016 … The USPTO recorded the most, with 2.8 million patents in 2016 …”). 39

Intellectual property exchanges and auctions 289 bilateral licensing methods more difficult.43 Owners within a country may originate from anywhere, and choice of law barriers in addition to standard language hurdles make negotiations more complicated. At the same time, business is increasingly global. It is far more likely that a firm will seek international licenses rather than dealing with owners on a country-to country basis. Facilitating exchange amidst the complexity is the increasingly accessible electronic deal-making environment. It is possible to simply advertise on the web or even offer standard licenses in an electronic marketplace. If the rationale for investigating new mechanisms of transfer is primarily to provide information and reduce the costs of negotiation, Internet-based systems certainly provide the power. Moreover, electronic systems have lowered the barriers to entry for smaller entities, leading to a more robust market for exchange. The varied interests and subject matter available for transfer has led to a diversity of transfer mechanisms (see Figure 13.1). Some are specific to a firm, industry segment or technology. Other, more recent forms seek to be a broad forum for exchange of any kind of technology right. And finally, some mechanisms are specifically directed to promoting exchange to fulfill an important social purpose. A large firm may make use of one or all of these forms in commercializing innovation.

Figure 13.1

Basic forms of non-traditional transfer mechanisms

43 Id. at 12 (“Applications filed in China increased from 18,700 in 1995 to 1.3 million in 2016, amounting to average yearly growth of 23%.”).

290 Research handbook on intellectual property and technology transfer A.

Subject Specific Transfer Marketplaces

A subject matter-specific exchange or marketplace is one that is intentionally more limited than an open forum. It could be restricted to particular technology, such as pharmaceuticals44 or renewable energy.45 Alternatively, the mechanism could be limited by type of participants, such as small businesses or universities. Potential licensees or purchasers look to the mechanism as a semi-curated source, perhaps with a stronger potential that complementary rights will be available. 1. Single-entity marketplaces or portals The original limited forum is probably best described as the single entity marketplace. Under this model, a firm or similar owner provides a forum for unspecified parties to identify useful inventions and enter into a license or purchase agreement. Typically, large firms have underutilized rights, which may be identified during an audit.46 Firms may determine that they can effectively recoup investment by opening up use to others.47 In addition, encouraging widespread use of a core technology may help build an “open innovation” ecosystem that can lead to greater profits than one firm acting on its own.48 One may even be able to use a licensing program to shape the competitive environment.49 Thus, even competitors may be a potential target for technology distribution. When Tesla announced that it would not enforce its electric vehicle patents against anyone who wants to use the technology “in good faith,” it was essentially prioritizing the environment possible from shared use rather than exclusivity.50 Firms commonly make such a determination for non-essential rights; they conclude that some licensing revenue is better than leaving the assets to gather dust in a basement. Single-entity marketplaces have existed throughout history. Inventors such as Thomas Edison and Charles Goodyear famously sold and licensed critical technology in phonographs and vulcanized rubber, respectively.51 Bell Labs became an important licensing portal for tran-

44 For example, there is an exchange that highlights licensing opportunities that is utilized by firms such as Novartis, Sanofi and Bayer called Phamalicensing, pharMalICensIng, available at https:// pharmalicensing.com (last visited Sept. 10, 2018). 45 See Negotiable Technology Licensing, nat’l renewable energy lab, available at https://www .nrel.gov/workingwithus/licensing.html (last visited Sept. 10, 2018). 46 Robert J. Roby & Carolina Paschoal, The Intellectual Property Audit: An Opportunity and a Necessity, 10 No. 2. landslIde 46 (2017). 47 Kevin G. Rivette and David Kline, Rembrandts in the Attic 124–9 (2000) (describing patent “fix” strategies to generate revenue). 48 See Tomoya Yanagisawa & Dominique Guellec, The Emerging Patent Marketplace, OECD Science, Technology and Industry Working Papers, 2009/09 (2009) (describing the “open innovation” strategy); Stephanie Vu, Note, Pledging Patents Effectively: Copyright and Open Source as a Framework for Patent Pledges, 14 Colo. teCh. L.J. 437, 440–5 (2016) (listing patent pledges to create an open source commons). 49 Katherine E. Rocket, Choosing the Competition and Patent Licensing, 21 RAND J. eCon. 161, 168–70 (1990) (describing the cases of cellophane, polyester and nylon). 50 Elon Musk, All Our Patent Are Belong to You, tesla (June 12, 2014), available at https://www .tesla.com/blog/all-our-patent-are-belong-you (last visited Oct. 17, 2019). 51 Adam Mossoff, Patent Licensing and Secondary Markets in the Nineteenth Century, 22 geo. Mason l. rev. 959, 962–5 (2015).

Intellectual property exchanges and auctions 291 sistor technology, albeit as an involuntary move resulting from an antitrust consent decree.52 In a time before globalization and efficient supply chains, such licensing was effectively the only way to distribute some innovations across a broad geography. Despite long historical threads, the most important leaps in this single entity model are recent. The penultimate single firm marketplace innovator is IBM. For years, IBM has been on top of the annual tally of US patent grants to companies, achieving a record 9,043 in 2017 alone.53 It has long been IBM’s strategy to license some portion of its rights as an alternate revenue stream.54 In the 1990s, the firm made a specific decision to open up part of its vast portfolio in order to make better use of its rights, as well as to use them to gain access to competitor technology.55 IBM engaged in a large-scale licensing program that essentially made its role as a marketplace an essential business function.56 The firm now advertises its holdings online and interested parties are invited to engage to investigate licensing opportunities.57 A distinguishing characteristic of a single-entity marketplace is that it usually involves licensing “out.” In other words, rather than a true exchange, it is more of a storefront for the managing entity to disseminate its technology to other firms. There are notable exceptions, of course, involving “in” marketplaces. Proctor & Gamble is quite upfront about its interest in working with outside inventors and developing collaborative products.58 It accepts submissions and advertises needs in order to create new products that might fall outside of the traditional corporate pipeline.59 However, most firms are reluctant to provide such intake opportunities, perhaps out of concern that it will serve as an invitation for patent assertion entities or those who might attempt to imply development contracts. Some firms have taken the single-entity marketplace to another level and exist primarily to invent and license or sell, rather than sell to consumers. They may engage in novel research within a particular field and subsequently transfer it, or they may partner with another firm in a consulting arrangement. In the former iteration, the firm is acting as a type of marketplace.

52

Jonathan M. Barnett, Property as Process: How Innovation Markets Select Innovation Regimes, 119 yale L.J. 384, 434–5 (2009) (describing the IP “giveaways” of large firms, including AT&T/Bell Labs). 53 25 Years of IBM Patent Leadership, IBM, available at https://www-03.ibm.com/press/us/en/ presskit/42874.wss (last visited Sept. 25, 2018). 54 Julie L. Davis & Suzanne S. Harrison, Edison in the Boardroom 80–1 (2001) (IBM’s Jerry Rosenthal tells the story of his firm’s licensing practices). 55 Peter Grindley & David Teece, Licensing and Cross-Licensing in Semiconductors and Electronics, 39 CalIf. MgMt. rev. 8, 14–15 (1997). 56 Kevin Rivette & David Kline, Discovering New Value in Intellectual Property, harv. bus. rev., Jan.–Feb. 2000, at 55–6; Robert C. Bird, Pathways of Legal Strategy, 14 stan. J.l. bus. & fIn. 1, 31–3 (2008). 57 Intellectual Property Licensing, IBM, available at https://www.ibm.com/ibm/licensing/ (last visited Sept. 26, 2018). 58 The portal is called P&G Connect + Develop, and has a sophisticated web-based submission system. Partnering for Mutual Value, proCtor & gaMble, available at https://www.pgconnectdevelop .com (last visited Sept. 25, 2018); Lydia Dishman, How Outsiders Get Their Products to the Innovation Big League at Proctor & Gamble, fast CoMpany, July 13, 2012, available at https://www.fastcompany .com/1842577/how-outsiders-get-their-products-innovation-big-league-proctor-gamble (last visited Oct. 17, 2019). 59 Open Innovation in Action, proCtor & gaMble, available at https://www.pgconnectdevelop.com/ needs/ (last visited Sept. 25, 2018) (“Sometimes innovators bring us breakthrough solutions we didn’t know we needed.”).

292 Research handbook on intellectual property and technology transfer One well-known example is SRI International and its East Coast subsidiary, the Sarnoff Corporation. The firm, which is a non-profit, exists to identify market opportunities, research, invent and transfer resulting technology.60 An interested firm can peruse the SRI license opportunities as one would real estate listings.61 More limited but in a similar vein is Interdigital, a non-practicing firm that specializes in mobile communications.62 Interested parties can consider technologies available for licensing and engage the firm. Industry players that do not carefully consider Interdigital’s IP holdings may risk a lawsuit. And therein lies a particular complaint about the non-practicing invention firms: if their rights merely capture information already known, some believe that they are less a marketplace for technology transfer and more of a toll taker on the path to commercialization.63 A separate but related form of a single entity technology marketplace exists in many universities. In this mode, the university offers access to patented technology and potentially related know-how from researchers in numerous areas. The University of Wisconsin’s licensing program is among the most well-known and impressive examples. Through its Wisconsin Alumni Research Foundation (“WARF”), the university has worked to serve as a technology dissemination service for the work of its researchers.64 Of course, its efforts greatly accelerated after the passage of the Bayh-Dole Act in 1982.65 Now, the university offers access to its more than 1700 active patents, individually and grouped into portfolios.66 The breadth of the university’s patent program has led industry to view it as a general technology licensing source. However, universities are often more limited in technology transfer than private firms. The Bayh-Dole Act does not allow the sale of university patents created with federal funds unless permission is granted from the federal government.67 Additionally, universities, particularly public universities, often view technology transfer as a service that is part of their public mission,68 reducing the need to identify the most profitable partners and maximizing non-exclusive arrangements over exclusive. Government research labs and agencies that create and license inventions can also be viewed as single entity marketplaces. Such forums are particularly prevalent at the federal level and

60 Michael Svanevik & Shirley Burgett, Matters Historical: The Slow Establishment of SRI in Menlo Park, san Jose MerCury news, Jun. 6, 2017, available at https://www.mercurynews.com/2017/06/06/ matters-historical-the-slow-establishment-of-sri-in-menlo-park/ (last visited Oct. 17, 2019). 61 Technology for License, srI InternatIonal, available at https://www.sri.com/engage/technology -for-license (last visited Sept. 30, 2018). 62 Our Licensing Program, InterdIgItal (2018), available at https://www.interdigital.com/page/ licensing (last visited Oct. 17, 2019). 63 Kristen Osenga, Formerly Manufacturing Entities: Piercing the “Patent Troll” Rhetoric, 47 Conn. l. rev. 435, 465 (2014) (noting that Interdigital has been named one of the “most fearsome patent trolls,” even though it has also commercialized some of its technology). 64 David C. Mowery & Bhaven N. Sampat, University Patents and Patent Policy Debates in the USA, 1925–1980, 10 Indus. & Corp. Change 781, 787–8 (2001) (relating the early history of, and rationale for WARF). 65 WARF & the Bayh-Dole Act, WARF, available at https://www.warf.org/about-us/history/warf -bayh-dole-act/warf-bayh-dole.cmsx (last visited Sept. 30, 2018). 66 Explore WARF Inventions and Patents, WARF, available at https://www.warf.org/technologies/ inventions-patents-and-portfolios.cmsx (last visited Sept. 30, 2018). 67 35 U.S.C. § 202(c)(7). 68 Daniel R. Cahoy, Toward a Fair Social Use Framework for College and University Intellectual Property, 41 J. C. & U.L. 485, 500–1 (2015).

Intellectual property exchanges and auctions 293 can be found at agencies as varied as the US Army,69 NASA,70 and the National Institutes of Health.71 Similar to a university, the market need is present here because government agencies have a limited ability to commercialize and often need a partner to develop and disseminate basic research. A related variation on this model is IP bundling, in which multiple agencies or universities put together complementary rights that would have limited appeal on their own.72 This appears to be a rather difficult endeavor given the need for diverse institutions to identify complementary rights, and bundling success has been somewhat limited.73 Limited exchanges and brokerages 2. A true marketplace or exchange with multiple buyers and sellers is one of the fastest growing areas in IP transfer. Even when restricted to a particular technology or industry, there are a number of different forms with varied business models. Some utilize a neutral forum, while others have an active third party or “intermediary” who facilitates the transfer. No doubt, many exchanges have enabled connections that have been essential for commercialization. However, some have been categorized as extractors or trolls for their tendency to prey on existing firms desperate to avoid the uncertainty and expense of IP litigation. And some serve a primarily defensive mechanism, essentially taking IP out of circulation or at least diminishing its threat value. On the downside, an exchange forum for multiple entities has the potential for more problems than single-entity settings. This is because such forums may bring together competitors, creating marketplace advantage issues as well as antitrust concerns. Thus, regulatory authorities are likely to scrutinize such endeavors, and designers must be careful to ensure that exchanges are something more than collusive behavior. The most basic form of exchange—in which the forum is a mere connecting entity for the transacting parties—appears to be the rarest. In this case, a platform allows rights holders to advertise opportunities to license or acquire and interested parties can review and engage as they see fit. The platform typically takes some participation fee or percentage to fund the marketplace. The narrow offerings and known (or at least industry-active) participants theoretically lower some of the transaction costs. The reason for the rarity is the multiple complications that accompany such an open forum. Foremost is the fact that they may offer merely a first stage of discussion, leaving the actually licensing or transfer arrangements to the parties who have made a connection.74 This is significantly more involved than an eBay or Amazon style transaction, and may not contribute much in terms of efficiency if the information costs are low to begin with. Second is the need to communicate a significant amount of information in the forum such that parties can potentially compete for opportunities without fear that lack of knowledge will yield a suboptimal deal.

69 Technology Transfer, arMy, available at https://www.arl.army.mil/technologyoutreach/ (last visited Oct. 1, 2018). 70 Patent Portfolio, NASA, available at https://technology.nasa.gov/patents (last visited Oct. 1, 2018). 71 Licensing Opportunities, nat’l Inst. health, available at https://www.ott.nih.gov/opportunities (updated Sept. 27, 2018). 72 Andrew R.O. Watson & Angus Livingstone, “Intellectual Property Clustering 7–11” in AUTM Technology Transfer Practice Manual (3d ed., 2010). 73 Id. (noting that several planned bundling initiative were on hold or never came to fruition). 74 Hagiu & Yoffie, supra note 8, at 55.

294 Research handbook on intellectual property and technology transfer For those reasons, some firms will not promote their most valuable rights, or they will keep alternative, bilateral options open.75 A good example of a subject-specific multi-actor exchange in the pharmaceutical industry is Pharmalicensing.76 This service, a spinoff of IP Exchange, operates as a site that allows IP owners (including universities and governments) to advertise their assets for purchase or license related to specific areas (type of solution, inventing entity, geographic origin, etc.). Enough information is presented for a potential licensee to determine the detailed nature of the technology and stage of development. Significantly, a service like Pharmalicensing includes market analytics and negotiation assistance to avoid some of the pitfalls of a naked exchange.77 Given the diversity of sources that can produce useful, basic-research information in the pharmaceutical industry, along with the fact that much value is produced by the development yet to come, an open forum makes a great deal of sense. At the more aggressive end of the spectrum is an exchange organized by an intermediary.78 This could appear as something similar to a patent pool, and in some cases the differences may be simply semantic. Typically, a pool is narrowly directed to certain technology contexts and focuses on disseminating facilitating technology rather than the final commercial product. A multiparty exchange need not be so narrow. Regardless, the organizing entity helps gather information on the assets (or the assets themselves) and packages them in a manner that is attractive to potential licensees or purchasers with less information. One form of intermediary-assisted exchange is the licensing agent, sometimes referred to as IP advisory or management firms, depending on their business practices.79 In the weaker form, this system organizes rights and provides a portal for interested licensees. Such a venture may also offer negotiation assistance. A firm known as Aqua Licensing gained attention in 2017 for providing licensing opportunities to startups that might be useful in bridging their “patent gap.”80 Still acting as a neutral exchange, Aqua envisions its activities as a service to new firms that are willing to give up equity for defensive patents.81 The more active advising form of intermediary-assisted exchange seeks out licensees for IP assets or vice versa. Such an entity may take ownership of the rights or simply serve as an agent with some percentage of the transaction captured in compensation. In fact, some large firms may see the advantage of transferring rights to a third party to carry out their licensing and enforcement,82 suggesting

75

Id. (“the lack of valuable patents [means] that few large operating companies [will] participate actively … ”). 76 Search Page, pharMalICensIng, available at https://pharmalicensing.com/app/home (last visited Sept. 25, 2018). 77 Features, pharMalICensIng, available at https://pharmalicensing.com/features.html (last visited Sept. 25, 2018). 78 Wang, supra note 14, at 165–6 (describing the general role of intermediaries). 79 See Raymond Millien & Ron Laurie, Meet the Middlemen, IaM Mag., Feb./Mar. 2008, at 54 (describing licensing agents). 80 See Connie Loizos, This Outfit Wants Startups to Trade Equity for Patents from Big Tech Players, teCh CrunCh, Aug. 2, 2017, available at https://techcrunch.com/2017/08/02/this-outfit-wants-startups -to-trade-their-equity-for-patents-from-big-tech-players/ (last visited Oct. 17, 2019). 81 IP Investor Pool, aQua lICensIng, available at http://aqualicensing.com/ip-investor-pool/ (last visited Sept. 30, 2018). 82 Yanagisawa & Guellec, supra note 48, at 22. Entities to whom IP owners outsource enforcement may also take the form of brokers. See Raymond Millien, Landscape 2013: Who are the Players in the IP Marketplace?, Ip watChdog, Jan. 23, 2013, available at http://www.ipwatchdog.com/2013/01/23/

Intellectual property exchanges and auctions 295 it is not solely an option for the undercapitalized. Examples of this form are IPValue83 and Thinkfire.84 Theoretically, the expertise of the agent is relevant in crafting deals that otherwise might not exist. Licensing entities have received a good share of criticism in recent years, suggesting that they do not always act as positive force in technology transfer. Most concerning is the licensing agent that does not wait for potential licensees to approach, but actively seeks targets who are potentially infringing. Known as patent licensing and enforcement companies (“PLECs”), they may simply serve as one type of a patent assertion entity that has the capability of extracting value with low quality patents.85 The most abusive of such firms are referred to as “patent trolls” in the popular media.86 The Lemelson Foundation is one of the earliest targets of such scorn, as it enforced the “submarine” patents of Jerome Lemelson against unsuspecting firms as far back as the 1990s.87 The basic idea perfected in his model was to fit patents to existing producers and demand a licensing fee to avoid litigation. By conducting such enforcement through a non-practicing entity, the risk of retaliation for infringement of another’s rights was non-existent.88 The number of licensing agents that have been called patent trolls is large indeed, and their alleged negative impacts well-reported.89 On the other side of the debate is the argument that licensing entities provide IP owners too small to litigate with a means of extracting revenue from inventions.90 Recent changes in the law have made this model somewhat less valuable for those with only extraction goals.91 The much-discussed IP “brokers” fit into the intermediary scheme as well, though one would consider them part of a market or exchange if they engage in repeated transactions. A broker can be defined as an entity that plays an active role in connecting buyers with sellers, as opposed to arranging licensing transactions.92 The broker may have a standing role in an industry, acting as a “buy-side” and “sell-side” transacting entity for multiple firms’ IP assets. But it is also common for brokers to interact only case-by-case, finding a marriage for one

ip-landscape/id=33356/ (last visited Oct. 17, 2019) (IP management or brokerage firms “may function more like IT companies when the IP owner essentially ‘outsources’ the monetization of the IP and is not involved in the day-to-day sale efforts, but still collects a majority of the sale revenue …”). 83 Approach, Ipvalue, available at https://www.ipvalue.com/approach (last visited Oct. 2, 2018). 84 thInkfIre, available at http://www.thinkfire.com (last visited Oct. 2, 2018). 85 Millien & Laurie, supra note 79, at 53–54 (describing PLECs). 86 Cahoy, supra note 68, at 503–4 (defining patent trolls and relating media accounts of their exploits). 87 Mark A. Lemley & Kimberley A. Moore, Ending Abuse of Patent Continuations, 84 b.u. l. rev. 63, 76 (2004) (describing Lemelson’s submarine patenting). But see Adam Mossoff, Patent Thickets and Patent Trolls, volokh ConspIraCy, May 5, 2009, available at http://volokh.com/posts/1241494164 .shtml (arguing that Lemelson is not a patent troll as is commonly defined in the context of patent acquisition, because he did invent the technologies he asserted). 88 See Colleen V. Chien, Predicting Patent Litigation, 90 tex. l. rev. 283, 292–3 (2011) (describing attributes of trolling model). 89 See, e.g., Fiona M. Scott Morton & Carl Shapiro, Strategic Patent Acquisitions, 79 antItrust L.J. 463, 482–3 (2014) (“The empirical evidence on PAEs [patent assertion entities], taken as a whole, supports the conclusion that enhanced monetization by PAEs is discouraging innovation and harming consumers.”) 90 Christopher A. Cotropia, Jay P. Kesan & David L. Schwartz, Unpacking Patent Assertion Entities (PAEs), 99 MInn. l. rev. 649, 653 (2014) (describing purported benefits of patent assertion entities). 91 See TC Heartland LLC, 137 S.Ct. 1514; Octane Fitness, LLC, 135 S.Ct. 1749. 92 Millien, supra note 82; Wang, supra note 14, at 167.

296 Research handbook on intellectual property and technology transfer particular set of assets at the request of one party.93 They are particularly important when a portfolio of rights is involved. When a buyer or seller wishes to transfer an array of rights, the broker is critical in identifying a party interested in such a volume of complementary rights. Individual firms have been playing the same role for their own interests for some time, but the specialized intermediary role of a broker arguable facilitates greater transaction frequency. Recent research by Love, et al., takes an in-depth look at a large part of the brokered patent market.94 Their conclusion is that the market is robust and surprisingly resistant to change in legal conditions.95 Additionally, Love, et al. find that large operating companies are the primary participants in the brokered market, suggesting that it has relevance to innovators.96 But, given the fact that only patent rights are sold (as opposed to know-how or other IP), the authors conclude that it is not a significant pathway for technology exchange. Rather, most of the action represents mere trading of legal liability protection.97 Whether a party prefers a limited subject matter system over something broader may depend on the industry and the nature of the rights being exchanged. Within a specific field, actors are more likely to have background knowledge that allows them to ascertain value with less information from the mechanism. Moreover, relationships exist that facilitate one-on-one negotiations. On the other hand, the close connection between the parties may mean that such narrow systems will only be a side avenue to traditional bilateral technology transfer. B.

Broad Exchanges, Marketplaces and Aggregators

At the forefront of alternate technology transfer systems are the broad exchanges, marketplaces and general aggregators. These are systems that intend to be all encompassing and open to rights holders in any industry. They should be accessible by any purchasing parties (who have access to the platform). For that reason, they exist in a potentially different space that draws in investment interests in addition to operators. In theory, the open participation should provide more efficiency. At the very least, it should be more transparent, with successful transaction practices and even pricing being more broadly available. Such a comprehensive exchange is a dream for those who believe that a large portion of existing rights are undervalued and insufficiently liquid due to transaction costs and information asymmetries. With the increasing capabilities of online tools as well as easily understood analogies to existing online marketplaces, the number of such forums has risen dramatically in recent years. On the other hand, not all have been successful, and there are suggestions of a downturn after some initial excitement over the innovations in new business models.98 And similar to more limited exchanges, there is a sense that general forums serve as a supplement to traditional mechanisms as a means of core technology transfer.

93

See, e.g., Kent Richardson, Erik Oliver, & Michael Costa, The 2017 Brokered Patent Market—The Fightback Begins, IaM Mag., Jan./Feb. 2018, at 8 (describing the authors’ efforts to secure patent portfolios for client, LinkedIn). 94 Brian J. Love, Kent Richardson, Erik Oliver, & Michael Costa, An Empirical Look at the “Brokered” Market for Patents, 83 Mo. l. rev. 359 (2018). 95 Id. at 406. 96 Id. at 405. 97 Id. at 404–5. 98 Millien, supra note 82; Yanagisawa & Guellec, supra note 48, at 15.

Intellectual property exchanges and auctions 297 The exact organization and available offerings differ from site to site. The classic rights access site is yet2, which is also one of the oldest.99 In this basic forum, technologies are listed as postings with existing protections (e.g., patents) included to signal the kind of license required.100 Another long-term player in this field is Tynex, which lists patents available for purchase as well as postings for desired rights.101 A firm like this acts as a kind of passive broker (though Tynex is also a standard broker as well). A more recent entry into the field is IAM Market, which lists portfolios available for sale as well as available licenses and requests for technology.102 It is another proudly passive site, which advertises its vetting rather than advising skills.103 However, its connection to a publication well known in the IP field likely gives it an advantage over sites that exist only to transfer patent interests. Significantly, there is a great deal of cross listing of the same IP on a number of sites, as exclusivity is not necessarily required. Similar to the way that a home owner may list on Airbnb and VRBO, or a driver may list on Lyft and Uber, the attraction of catching as many eyes as possible seems to be a primary motivation of listing. Whether such tactics are attractive may depend on the intent of the purchasing party. As noted above, the interest of collecting and licensing for purposes of trolling argues for amassing many rights in various fields, even if the strength or scope of particular patents is unclear. Unlike the eBay or Amazon model, the parties in most exchange models do not complete their transaction online but must continue after making a connection. This fairly standard feature of a two-sided platform has also been the focal point of its detractors.104 The fact that parties must engage in additional one-on-one negotiations detracts from the efficiencies of an online marketplace and limits the ability to transform these services into something different. On the other hand, it is quite possible that a smart operating firm would be reluctant to release so much control over a truly valuable asset (or place such faith in purchasing one). Regardless, this fact may play a major role in keeping exchanges secondary to other means of transferring rights. Unfortunately, due to the private nature of exchange and marketplaces, it is not possible to know precisely how successful they actually are.105 Even if they involve rights transfers, the public record of assignments is notoriously incomplete.106 Some reports claim that most activity in recent years was conducted by non-practicing entities.107 However, most exchanges

99

Services: yet2 Marketplace, yet2, available at https://www.yet2.com/services/yet2-marketplace/ (last visited Sept. 30, 2018) (description of marketplace). 100 Yet2 Marketplace, yet2, available at http://marketplace.yet2.com/app/about/home (last visited Sept. 30, 2018). 101 Welcome, tynax, available at https://www.tynax.com (last visited Oct. 4, 2018) (listing 456,311 patents available). 102 IaM Market, available at https://www.iam-market.com (last visited Oct. 4, 2018). 103 Sell or License Intellectual Property and Technology, IaM Market, available at https://www.iam -market.com/node/32 (last visited Oct. 4, 2018) (one of the “reasons to become an IAM Market vendor” is “all potential buyers get vetted by IAM Market.”). 104 Hagiu & Yoffie, supra note 8, at 54. 105 Ashby H.B. Monk, The Emerging Market for Intellectual Property: Drivers, Restrainers, and Implications, 9 J. eCon. geography 468, 471–2 (2009). 106 See Love, et al., supra note 94, at 362–3 (“Currently available information about the patent marketplace is largely anecdotal and qualitative in nature.”). 107 See Anne Kelley, Practicing in the Patent Marketplace, 78 u. ChI. l. rev. 115, 118–19 (2011) (asserting that “[t]he bulk of the buying in the patent marketplace is by NPEs”).

298 Research handbook on intellectual property and technology transfer and marketplaces claim participation by large, mainstream firms.108 It is likely that interest in alternative mechanisms may shift from year to year. Direct evidence aside, given the industry commentary and observations of who has survived, one can intuit factors that seem to be relevant to the success of an alternative technology transfer mechanism (see Figure 13.2). Because faith in the reliability of the information on an exchange and as well as the intentions of its operator are relevant to transacting parties, a site’s reputation is extremely important. For this reason, building a brand is an incredibly important part of running an alternate transfer mechanism. As more and more firms enter and leave the space, cutting through the noise may be one of the most important distinctions between success and failure. This factor may be a primary reason that a startup like IAM Markets can instantly stake a claim to a significant part of the market. But reputation alone is not enough. An exchange must also be able to attract high quality rights such that potential buyers see it as a real source of technology. Too many marginal exchanges are viewed as dumping grounds for leftover IP, which reduces the incentive to spend resources on reviewing offerings. Additionally, for true technology transfer to take place, complementary IP must be available. Although some fields can capture a product with only a few rights, many require a portfolio of overlapping IP. Also, an often-lacking feature is information sufficient to allow parties to reliably value available rights. Too often, intermediaries desire to view rights as similarly valuable assets with average prices and applications. But that is anathema to valuable patents, and no serious technology transfer takes place in the absence of considered valuation. Finally, there is a timing aspect that cannot be overlooked. Transfer is inherently more attractive when an emerging technology grabs the collective attention of various industries

Figure 13.2

Building blocks of a successful exchange or market

108 See, e.g., Services: yet2 Marketplace, supra note 99. (stating the marketplace began with a “core group of leading companies and institutes”); Licenses, Ipvalue, available at https://www.ipvalue.com/ licensees (last visited Oct. 2, 2018) (offering “licenses with some of the largest and most sophisticated enterprises in the world.”); Intellectual Property Transactions, oCean toMo, available at http://www .oceantomo.com/intellectual-property-transactions/ (last visited Oct. 2, 2018) (stating relationships include “personal contacts with senior IP and C-level executives at the largest operating companies …”).

Intellectual property exchanges and auctions 299 (e.g., blockchain) or the enforcement value of rights is perceived to be strong. Conversely, when IP is viewed as an option adjunct to innovation and enforcement business models are limited, less transfer will occur. Exchanges and marketplaces that exist with more of these factors seem to be the ones that stay around. A related, but distinct approach to the general IP exchange is the idea exchange. In this case, the exchange is directed to highlighting problems, rather than IP for sale, and it allows solvers to engage with their potentially propriety solution. However, problem solvers do transfer access to the IP as a consequence of being picked as the winning solution. A successful version of this idea/IP exchange is Innocentive.109 This forum allows clients such as Ford, Thomson Reuters and the Environmental Protection Agency to post “challenges” for users to attempt to solve.110 A user submits a solution, typically in an attempt to win a cash award. The award winner grants either a non-exclusive license or exclusive transfer of IP in exchange. In recent years, a different type of general technology intermediary has taken center stage. The aggregator has emerged as a powerful source of incentive for individuals and companies to dispose of rights.111 As noted above, aggregators have existed in some form for practically the entirety of US IP history, and they gave rise to some of the first pools.112 In fact, the difference between an aggregator and a pool is somewhat subtle, but generally rests in a pool’s focus on facilitating the adoption of a particular technology covered by diversely-owned rights. In contrast, aggregators generally collect rights on a variety of technologies and make them available for license in a variety of industries. And true to this distinction, the more recent form of aggregator rose to prominence not as a technology consolidator or market shaper, but rather by keying in on patent rights as an investment. In highlighting the business model laid bare by the growing thickets of rights in certain technologies, as well as high litigation costs and inefficiencies, aggregators were able to generate significant interest in collecting rights. In turn, this interest increases activity in the secondary market and seems to have the positive effect of making firms more open to technology transfer. The basic model of a modern aggregator is to purchase rights outright.113 Those rights are then available for license. But rather than simply facilitating the exchange, the aggregator will have as its business model an interest in seeking infringing parties and offering a license as a settlement.114 By offering a license at a rate below the cost of litigating, potential licensees have little incentive to push back.115 Thus, an aggregator can take weak rights and collect royalties from license after license, providing a significant return on an investment in what

109

Maurizio Borghi, Kris Erickson & Marcella Favale, With Enough Eyeballs All Searches Are Diligent: Mobilizing the Crowd in Copyright Clearance for Mass Digitization, 16 ChI.-kent J. Intell. prop. 135, 156–7 (2016) (describing InnoCentive as a type of crowdsourced invention). 110 Challenges, InnoCentIve, available at https://www.innocentive.com/ar/challenge/browse (last visited Oct. 2, 2018). 111 Yanagisawa & Guellec, supra note 48, at 25; Millien, supra note 82; Wang, supra note 14, at 177–82. 112 See generally Mossoff, supra note 51. But see Risch, supra note 41, at 989 (claiming aggregation in the current sense in a more modern phenomenon). 113 Wang, supra note 14, at 177. 114 Mark A. Lemley & A. Douglas Melamed, Missing the Forest for the Trolls, 113 ColuM. l. rev. 2117, 2153–4 (2013) (describing the complaints about patent aggregation). 115 Morton & Shapiro, supra note 89, at 470.

300 Research handbook on intellectual property and technology transfer the inventing firm may have considered to be worthless technology.116 Another key aspect of an aggregator is the fact that they are insulated from counter-attack because they produce no products.117 Most firms are cautious about litigating weak rights for fear of facing a counter lawsuit on patents their products may infringe. A non-practicing aggregator faces no such concern. Although an aggregator is technically a single owner, one can consider it a component of the IP market because it serves as a broad repository for rights. Aggregators have less concern about the overall fit of the rights they acquire within a particular business segment because they are in the business of licensing. They become an outsized player that motivates other parties to consider how rights will be utilized in the system, as well as how certain technologies will be commercialized. The fact that they offer a broader outlet for orphaned IP may also spur upstream investment in research and development. The most well-known aggregator is certainly Intellectual Ventures. Founded in 2000 by Nathan Myhrvold and Edward Jung, the firm claims to own 95,000 patents in areas as diverse as agriculture, consumer electronics and physical sciences.118 Most of the patents were acquired from others, but some resulted from inventive work inside the firm.119 Funds to purchase rights came from investors interested in the potential return on investment that could come from steady royalty income.120 At first, Intellectual Ventures simply acquired and licensed rights, but eventually it began litigating alleged infringers, earning scorn as one of the world’s largest patent trolls.121 In fact, Intellectual Ventures is more complex, offering licensees the ability to use its portfolio to attack competitors engaged in litigation.122 Many viewed the firm as the bellwether of the acquisition and license business model. In recent years, the model has seemingly fallen prey to the new speedbumps set up to discourage trolling. Intellectual Ventures stopped acquiring patents in 2017123 and has lagged in investment returns.124 In response to the problem of aggressive litigation practices by aggregators and other non-practicing entities, firms have collaborated to share ownership or cross-license rights to protect themselves. Termed defensive patent aggregation, the practice generally involves the formation of a separate business entity to acquire and freely license rights.125 Alternatively, the

116

However, it has been argued that repeat player aggregators have an incentive to assert strong patents, lest licensing companies conclude all of their assets are weak. David L. Schwartz, On Mass Patent Aggregators, 114 ColuM. l. rev. sIdebar 51, 60 (2014). 117 Lemley & Melamed, supra note 114, at 2129–30. 118 Our Patent Portfolio, Intell. ventures, available at http://www.intellectualventures.com/ inventions-patents/patent-portfolio (last visited Oct. 2, 2018). 119 Hagiu & Yoffie, supra note 8, at 58–9. 120 Id. 121 Wang, supra note 14, at 181; James M. Rice, The Defensive Patent Playbook, 30 berkeley Tech. L.J. 725, 740 (2015) (describing Intellectual Ventures’ evolution after the dot-com bubble); Richard Waters, Intellectual Ventures Launches First Three Lawsuits, fIn. tIMes, Dec. 8, 2010. 122 Hagiu & Yoffie, supra note 8, at 59–60. 123 Dennis Crouch, Intellectual Ventures Stops Buying Patents, patentlyo, April 13, 2017, available at https://patentlyo.com/patent/2017/04/intellectual-ventures-patents.html (last visited Oct. 17, 2019). 124 Nathan Vardi, After 10 Years, Nathan Myrhvold’s $3 Billion of Private Equity Funds Show Big Losses, forbes, Jun. 1, 2018. 125 See James M. Rice, The Defensive Patent Playbook, 30 berkeley teCh. L.J. 725, 738 (2015) (detailing the early history of defensive patent aggregation); Hagiu & Yoffie, supra note 8, at 56–8 (describing the patent aggregation business model).

Intellectual property exchanges and auctions 301 entity may acquire and sell rights to members, ensuring they stay out of trolling hands. One of the first such defensive aggregators was Allied Security Trust (AST), a non-profit organized in 2008.126 It practices a model in which it acquires patents, licenses them to an interested member, and then sells them to ensure that AST will never practice aggressive licensing.127 Another firm that has received much attention with a slightly different model is RPX.128 That entity, which is for-profit, acquires and provides licenses to all of its subscribing members.129 An inverse twist is embodied by the LOT Network, a collaboration of nearly 300 companies that acts only when one of its members sells a patent to a non-practicing entity.130 If that happens, LOT members automatically obtain a license, ensuring that the purchasing entity cannot hold-up the members with the threat of litigation.131 The variety of defensive models suggest that firms are compelled to find creative models for preventing aggressive licensing practices without substantially devaluing the value of their respective portfolios. In many ways, defensive aggregators serve as more of an exchange or market than offensive aggregators because they provide a safe place for owners to dispose of unwanted rights without fearing the criticism that accompanies transacting with non-practicing entities. However, their interests are constrained by their membership. Thus, defensive aggregators are often much narrower in scope and are active primarily in technology areas with a great deal of overlap. C.

Social Responsibility-Oriented Platforms

An important variation on the subject matter-specific exchange is the forum created to achieve a social goal. Assuming exchanges and markets facilitate technology transfer, it stands to reason that one might support the commercialization of socially useful inventions with a specific market. A specific subset of platforms has emerged to fill this need. Again, this would be particularly important if one suspected that such inventions are underutilized, in part because the value proposition is less clear. The sponsoring entity or intermediary could be a private party, such as a non-government organization, an international treaty organization or a government. Critically, social-responsibility transfer organizations may have more leeway in their expected profits or other success measures. It is easy to view all alternative mechanisms through the lens of financial success, but it is important to keep in mind that some services are intended primarily to serve a social goal. Thus, the success may be found in fewer transactions, and even reside in proof of concept. Typically, a social responsibility platform will focus on under-appreciated technologies under the assumption that other, objectively-valuable IP will be transferred no matter what the system.

126 Proactive Patent Defense, allIed seCurIty trust, https://ast.com (last visited Oct. 2, 2018); Quiong Qi, Patent Alliance: Defensive Aggregation—The Market Solution to Non-Practicing Entities?, 57 Idea 73, 88–93 (2016) (describing AST’s defensive model). 127 Qi, supra note 126, at 90 (describing AST’s “catch, license and release” model). 128 The RPX Network, RPX, available at https://www.rpxcorp.com/solutions/rpx-network/ (last visited Oct. 2, 2018). 129 Qi, supra note 126, at 94–8 (describing RPX’s “catch and hold” model). 130 How LOT Works, lot network, available at https://lotnet.com/how-lot-works/ (last visited Oct. 2, 2018) (describing how system protects against Patent Assertion Entitities). 131 Rice, supra note 125, at 768–9.

302 Research handbook on intellectual property and technology transfer An early entry into this field was an exchange designed to facilitate technology transfer in alternative energy and other sustainable technologies. Funded by the World IP Organization, WIPO Green is a forum for matching owners of sustainable technology with potential users and providers.132 The exchange has a specific global focus, and it appears that technology owners are largely from the developed world, whereas users are typically from the developing countries. As with more general exchanges, WIPO Green requires additional negotiation to close deals. Thus, transaction costs are somewhat high, and the number of actual transactions has remained relatively low.133 Somewhat related to WIPO’s technology exchange are “commons” systems designed to make available rights in sustainable technology to incentivize follow-on research and development. In this form, companies agree to license for little or no charge basic technologies in the field. Rather than a full exchange, the model is more like the Creative Commons in copyright134—essentially a non-assertion pledge. GreenXChange and Eco-Patent Commons are important proof-of-concept iterations, but neither remained active for long after their founding.135 Harkening back to old models, socially responsible patent pools exist as well. In this space, they tend to operate more like the commons form described above; however, they tend to be focused on particular products. The UNITAID Medicines Patent Pool is a good example. To permit access to technology in the medicines space, firms provide low-cost or free licenses to specific drugs used to treat HIV, tuberculosis and hepatitis C.136 The idea is to allow generic versions of the drugs to be made and distributed in certain regions of the world (but not in the high-income countries that account for most profits).137 A similar example exists in agriculture related to golden rice technology.138 Because such licensing is less about technology transfer than providing infringement protection, one might reasonably conclude that it contributes less to socially-important innovation than a sustainability exchange.

132 WIPO Green—The Marketplace for Sustainable Technology, WIPO, available at https://www3 .wipo.int/wipogreen/en/ (last visited Oct. 2, 2018). 133 wIpo green, year In revIew 2016 (2017) (declaring only about 120 “connections” facilitated despite 5000 network members and noting that 65% of represented technologies come from universities (down from 90% initially)). 134 See generally Lydia Pallas Loren, Building a Reliable Semicommons of Creative Works: Enforcement of Creative Commons Licensees and Limited Abandonment of Copyright, 14 geo. Mason l. rev. 271 (2007) (describing the creative commons system as predicated on the retention and limited enforcement of copyright). 135 Watson & Livingstone, supra note 72, at 6–7 (describing the Eco-Patent Commons system); Kevin Greenleaf & Michael P. Byrne, Triumph of the Eco-Patent Commons, 4 no. 1 landslIde 43 (2011) (describing both the Eco-Patent Commons and GreenXchange). 136 Our Model, MedICInes patent pool, available at https://medicinespatentpool.org/who-we-are/ our-model/ (Oct. 2, 2018). 137 Krista L. Cox, The Medicines Patent Pool: Promoting Access and Innovation for Life-Saving Medicines Through Voluntary Licenses, 4 hastIngs sCI. & teCh. L.J. 291, 296–7 (2012) (overview of Medicines Patent Pool). 138 Watson & Livingstone, supra note 72, at 5–6.

Intellectual property exchanges and auctions 303

IV.

EMERGING AUCTIONS AND LICENSE EXCHANGES WITH UNREALIZED POTENTIAL

The quest to promote technology exchange and monetize underutilized assets has led to the exploration of a creative new set of transfer mechanisms. Most prominently discussed are auctions and per-use license exchanges. Of course, calling these mechanisms new is a bit of a misnomer, as both exist in other parts of the business world and have been used with success for some time. However, they are relatively new to IP, and some would say their time has come. There is a sense that the IP rights community is too staid and reliant on tradition. Many believe that the integration of new models is just what the field needs to maximize the return on investment and insure that those who most value technology rights can attain them. To date, though, results have been less than stellar. A.

Auctions

A patent auction is a mechanism where interested parties bid competitively for either the purchase of a right or a license at a particular price. They represent another level of abstraction from one-on-one negotiations or exchanges.139 There are a variety of forms, but the familiar English option is generally the default.140 In fact, auction theory suggests that the form chosen is actually irrelevant for the seller.141 After the property is offered (with or without a reserve price) bids are placed usually for a fixed, upfront payment. Bids may be accepted only one time (a sealed bid) or bidders may have the opportunity to increase their bids dynamically (an ascending auction). An auction can be for a single transaction across a wide field or consist of multiple bidding opportunities. At the end of the auction, the right is typically assigned to the single bidder with the highest bid. Although alternate pricing arrangements are available and might entail strategic advantages,142 the winner typically pays his or her bid for the auctioned right. The supposed benefit of an auction model in IP is essentially the same as with other property: price maximization for the seller. An auction that awards the object to the highest bidder and sets an appropriate non-trivial reserve price will maximize a seller’s expected revenue when (1) there is a single, indivisible item for sale by a monopolistic seller, and (2) a fixed set of bidders have private information about their own personal valuation of the item.143 In other words, auctions can theoretically bring in the greatest revenue under these conditions. In addition, we can efficiently identify the party who most values the right as well as the

139 John Jarosz, et al., Patent Auctions: How Far Have We Come?, 45 les nouvelles 11, 12–13 (2010). 140 Paul Milgrom, Auctions and Bidding: A Primer, 3 J. eCon. persp. 3, 13 (1989). 141 Roger B. Myerson, Optimal Auction Design, 6 MatheMatICs operatIons res. 58, 65–6 (1981). The Revenue Equivalence Theorem states that any auction, which assigns the object to bidders with the same probability (possibly depending upon their value), will always engender strategic behavior such that the expected revenue for the seller is the same. See John H. Kagel, “Auctions: A Survey of Experimental Research” in The Handbook of Experimental Economics 501, 503–4 (John H. Kagel & Alvin E. Roth eds., 1997). 142 William Vickrey, Counterspeculation, Auctions, and Competitive Sealed Tenders, 16 J. fIn. 8, 20–3 (1961). 143 Myerson, supra note 141, at 69.

304 Research handbook on intellectual property and technology transfer precise value they assign. Multiple negotiations are unnecessary, and there is some degree of transparency (either a party knows its wining price or that another party bid more). Moreover, under most iterations, it is egalitarian: once parties are qualified to bid, the personal attributes of the bidders are irrelevant. The overall price maximization is one reason that the auction has come to be viewed as a means of dealing with IP assets that have not found interested buyers in normal transactions. Although auctions are not entirely new to IP—reports of auctioned patents in bankruptcy cases extend back many years144—it was the offerings from Chicago’s Ocean Tomo Bank that created widespread attention to the mechanism. In 2006, the firm made a splash by auctioning 78 lots (groups) of patents in a variety of industries, including energy, medical and financial services.145 The initial offering made about $8 million in sales and about 50% of the participating sellers transacted their patents.146 In subsequent years, the number of successfully auctioned rights decreased along with the profits.147 Eventually, Ocean Tomo sold its auction business, only to restart a revised version in 2014.148 Auctions have been used by non-commercial entities as well. Purportedly the first university-run auction of its own patent assets was conducted by Penn State in 2014.149 The university offered 73 issued patents for license, primarily related to engineering technologies.150 A second auction took place later that same year, with licenses to 110 patents available.151 In general, these rights had been available for license for some time, but the university’s technology transfer office was not successful in connecting with a licensee.152 Both auctions resulted in a small measure of success, with three licenses being concluded in total.153 At the very least, they were a proof of concept of the utility of the mechanism. Given the potential advantages of auctions, not to mention the fact that they are not uncommon in other parts of the business world, why has the adoption remained small? One reason is that they maximize price over any other part of the transaction, such as accompanying know how or payment terms.154 Other exchange and market forms at least allow for the possibility

144 See Perry J. Viscounty, Michael Woodrow De Vries, & Eric Kennedy, Patent Auctions: Emerging Trend?, nat’l L.J., May 8, 2006 (reflecting on the fact that patent auctions are not new and citing a 2004 bankruptcy auction). 145 Jarosz, supra note 139, at 11. 146 Adam Andrzejewski, Patent Auctions: The New Intellectual Property Marketplace, 48 U. louIsvIlle l. rev. 831, 836 (2010). 147 Jarosz, supra note 139, at 30 (noting that activity eventually fell and many patents were sold at a substantial discount to their estimated value); Hagiu & Yoffie, supra note 8, at 54–5. 148 Kristi L. Stathis, Ocean Tomo to Re-Enter Live Intellectual Property (IP) Auction Market, Ocean Tomo Insights Blog (Oct. 6, 2014), available at http://www.oceantomo.com/blog/2014/10-06-ocean -tomo-re-enter-live-intellectual-property-ip-auction-market/ (last visited Oct. 17, 2019). 149 See Daniel R. Cahoy, Anthony M. Kwasnica, & Luis Lopez, The Role of Auctions in University Intellectual Property Transactions, 54 duQ. l. rev. 53, 72 (2016). 150 Id. (citing meeting with Dr. Ron Huss, Assoc. Vice President for Research & Tech. Transfer, Pennsylvania State University, in State College, Pa (Mar. 31, 2015)). 151 Id. 152 Id. 153 Id. 154 See Guhan Subramanian, Negotiation? Action? A Deal Maker’s Guide, harv. bus. rev., Dec. 2009, available at https://hbr.org/2009/12/negotiation-auction-a-deal-makers-guide (last visited Oct. 17, 2019) (“although auctions do a great job of boiling everything down to price, boiling everything down to price is often at odds with identifying valuable opportunities.”).

Intellectual property exchanges and auctions 305 of a one-to-one negotiation before a transaction is fully concluded. In addition, the notion that auctions are not attracting the most valuable property has a recapitulating effect. Buyers do not take them seriously, and sellers are less inclined to risk listing rights that will be perceived as less valuable because they are auctioned. Although it is far too soon to declare the model dead for IP transactions, it certainly has not reached the potential originally envisioned. B.

Patent Commodity Exchanges

Another creative and experimental design for IP transactions is embodied in the business model of IP Exchange International (“IPXI”). This exchange was not designed around transfers of whole patent rights or general patent licenses, but rather attempted to boil patent licenses down to a unit.155 Patent owners would transfer control of their patents to IPXI to serve as a master licensing agent. IPXI would then divide the patent into “unit license rights,” or ULRs, that could be sold individually or more likely in large sets. A firm purchasing a ULR would attain a license for one product (e.g., a single car or printer). If the buyer knew that its production needs were high, it would buy more rights.156 Critically, those ULRs could be resold—traded on the secondary market—over and over again. Through this arrangement, a market in ULR could be maintained that would reflect the true value of the license per unit over time. The point of IPXI was to address inefficiencies in traditional patent license markets. According to one of the firm’s founders, the fact that standard licensing requires resource intensive inputs to determine the actual value of the patent and litigation propensities of the two parties is a huge obstacle.157 IPXI would relieve the parties of this determination by performing extensive due diligence including prior art searches. Moreover, the fact that these calculations could change over time makes the initial negotiation all the more difficult. Finally, follow-on transfers of the right face all of the same transaction costs of the original exchange. And there is almost no transparency in traditional licensing, so the players cannot truly assess whether their arrangement is appropriately valued. IPXI was intended to solve these transaction problems similar to a commodities market. From the outset, IPXI received acclaim for its innovative system.158 Many believed that it heralded a new future for IP transactions with significantly greater efficiencies. However, the business community—specifically, potential purchasers of the ULRs—were less inclined to jump in. There are potentially a number of reasons why firms did not take to the new technology transfer model. One is general inexperience with option markets in the IP community. In addition, the lack of personal interaction and ability to negotiate other aspects of the deal was likely a barrier as well. And, ultimately, the lack of a perceived need for the licenses was suggested by one of the founders as one of the most important problems. According to IPXI

155 Yanagisawa & Guellec, supra note 48, at 25; Merritt L. Steele, Note, The Great Failure of the IPXI Experiment: Why Commoditization of Intellectual Property Failed, 102 Cornell l. rev. 1115, 1121–4 (2017) (describing the IPXI model). 156 Steele, supra note 155, at 1121–2. 157 IPXI: The Life and Death of an Experimental IP Exchange, ColuM. teCh. ventures, Feb. 2, 2017, available at https://techventures.columbia.edu/news-and-events/videos/ipxi-life-and-death -experimental-ip-exchange (last visited Oct. 17, 2019) (interviewing Ian McClure). 158 See, e.g., Tom Groenfeldt, New IP Exchange Promises Transparency in Patent Pricing, forbes, Dec. 6, 2013, available at https://www.forbes.com/sites/tomgroenfeldt/2013/12/06/new-ip-exchange -promises-transparency-in-patent-pricing/#288a0a4743f0 (last visited Oct. 17, 2019).

306 Research handbook on intellectual property and technology transfer founder, Ian McClure, many potential licensees were increasingly willing to risk infringement at the time of the exchange’s launch.159

V.

THE FUTURE OF NON-TRADITIONAL TRANSFER MECHANISMS

To many, the performance of alternative mechanisms for transferring technology has proved disappointing. The initial excitement that surrounded the myriad of new systems and models launched since 2000 has dissipated to a great extent. Perhaps we are not truly on the cusp of an entirely new paradigm for transferring technology and the old ways will remain the standards. This view is not unreasonable, but it is worth considering whether the dampened environment is at least partially related to certain missing elements. If the right pieces could be put together (see Figure 13.2), perhaps more efficient exchanges and even auctions could take hold. A.

Reliable Valuation

The one aspect that seems to be most tenuous for non-traditional transfer mechanisms is the value proposition for participating. Largely, this relates to the inability to assign significant value to the IP being transacted.160 Mechanisms that maximize volume and speed in transactions—particularly in an attempt to mimic forums on which assets with more accepted value are traded—risk leaving unsolved the greatest barrier to technology transfer: information asymmetry. On exchanges and auctions in particular, it is difficult to disclose sufficient information to make a potential purchaser confident that the right in question is worth the asking price. The availability of average asset prices does little to assuage this concern, as IP rights are by their very nature unique and not subject to simple comparative analysis as one might conduct with homes in a subdivision. Note that the upswing in transaction (secondary market) activity in the last ten years does not disprove the fact that valuation is a central problem. As the trolling model of patents evolved, firms were more willing to acquire questionable rights with the knowledge that inefficiencies in the judicial system, particularly in the U.S, could permit some return based on mere threat value.161 The push to accumulate massive numbers of rights for offensive and defensive capabilities had the effect of pushing up prices. However, once various corrective measures to curb trolling took hold, that value dissipated because it was not based on commercialization and consumer interest.162 There remains room to inject better valuation into alternative mechanisms for transfer. Some services provide detailed value assessment to back offerings, though potential purchasers may not give them much weight. This may change as IP valuation continues to evolve 159 IPXI: The Life and Death of an Experimental IP Exchange, supra note 157; see also Joff Wild, IPXI Demise Caused by a U.S. Patent System that Offers no Incentive for Good-Faith Licensing, Says Exchange’s CEO, IaM Mag., Apr. 7, 2015 (reporting comments of IPXI CEO Gerard Pannekoek). 160 Kelley, supra note 107, at 124–6 (2011) (describing the problem of valuation in patent marketplaces). 161 See Matthew Vella & Chris Donegan, The Secondary Market in Patents: What Went Right, What Went Wrong, and How to Fix It, IaM Mag., Oct. 12, 2017. 162 Id.

Intellectual property exchanges and auctions 307 as a field and standards become more accepted. There is a growing academic literature on reliable valuation that can be leveraged in this effort. Only when firms believe that intelligent and defensible financial decisions can be made in alternative technology transfer forums will a significant shift take place. B.

The Human Element

More difficult to address is the traditional importance of human interaction in technology transfer. In a practice area dominated (or at least heavily influenced) by lawyers, there may be a learned inclination to fall back on face-to-face negotiation. Moreover, a relational aspect to technology transfer exists in complex contracting. It is reasonable for parties to desire more information about the application and utility of transferred assets than can be gleaned from a patent application or summary market analysis. Peter Lee describes the “personal connection” problem in great detail in the context of university technology transfer.163 He notes that the market-oriented model of technology transfer has been reasonably criticized for ignoring the problems of poor disclosure and transaction costs.164 Professor Lee builds on this by describing the role of relationships and tacit knowledge in technology transfer that are some of the most important factors in successful transfer.165 Such considerations are not often accounted for under the standard “commoditized” view of IP. The human element that flows from bilateral negotiation between parties with repeated interactions cannot be fully replicated by an interactive website. However, it may be possible to inject additional personal connections into technology transfer mechanisms that at least facilitate some additional transactions. One method is through increased vetting and better use of indicators of experience, reliability and trust. Mechanisms limited to industries and contexts in which players know each other have a stronger basis than fully open markets. Lee suggests that such relational transfer should be explicitly enhanced by the public sector when it is the original source of the technology or funding.166 Additionally, industry buy-in will have a recapitulating effect. Such improvements are likely to run hand-in-hand with better valuation techniques. Equally important is to bolster private markets with public support. For example, government and public-sector organizations can facilitate better valuation systems as the holders of the primary data as well as continue to improve the objective quality of rights.167 Additionally, they could engage in matchmaking in certain contexts and provide more robust sharing platforms.168 WIPO’s Green Patent exchange and the Unitaid Medicines Patent Pool are modest example in the social responsibility context.169 If broadened to contexts with greater commercial applicability, the impact of public support may be significantly greater. 163

Peter Lee, Transcending the Tacit Dimension: Patents, Relationships, and Organizational Integration in Technology Transfer, 100 CalIf. l. rev. 1503 (2012). 164 Id. at 1518–20. 165 Id. at 1521–30. 166 Id. at 1560–1. 167 Yanagisawa & Guellec, supra note 48, at 36, 39. 168 Id. at 36–7, 39–40 169 See WIPO, supra note 132; wIpo green, supra note 133; Medicines Patent Pool, supra note 136; Cox, supra note 137; Watson & Livingstone, supra note 72, at 5–6.

308 Research handbook on intellectual property and technology transfer

VI.

CONCLUSION

The potential for non-traditional mechanisms to facilitate technology transfer has grown dramatically over time and will continue to progress with the assistance of new software and systems. To date, their full influence is yet to be felt, as the conversation has been largely influenced by a gold rush of novel methods that did not fully pan out. By analyzing the attributes of successful methods as well as paying close attention to the current needs of industry, we will likely see a more important role for alternate transfer systems as a supplement to the standard bilateral transaction.

14. Currents and crosscurrents in litigation of university and nonprofit related patents: is there a coming wave of patent litigation involving those patents? Teo Firpo and Michael S Mireles

I.

INTRODUCTION

Universities play a critical role in the creation and dissemination of knowledge. In particular, technology transfer offices (“TTOs”) and patents have received considerable attention since the passage of the Bayh-Dole Act. The Bayh-Dole Act allows universities and other nonprofits to take title to government funded invention.1 Grant recipients are allowed to take title because of a concern with the tragedy of “underuse” of government funded inventions—arguably many government funded inventions were not being commercialized prior to the passage of the Bayh-Dole Act. After passage of the Act, university involvement in patenting and licensing has increased substantially and continues to rise according to statistics collected by the Association of University Technology Managers (“AUTM”).2 A substantial amount of literature has been written concerning whether passage of the Bayh-Dole Act has resulted in unintended negative consequences, such as the development of an anticommons or shifting researchers from moving from basic to applied research projects.3 An issue that has not been explored in as much detail concerns the enforcement of patents generated by universities and other nonprofits. In examining numerous sources, this Chapter sets forth data concerning patent enforcement of US patents in the United States by universities, nonprofits and their licensees and assignees. This work builds on existing studies by authors such as Professor Jacob Rooksby and others.4

1 Bayh-Dole University and Small Business Patent Procedures Act of 1980, Pub. L. No. 96-517, 94 Stat. 3015 (codified as amended at 35 U.S.C. §§ 200-211 (2000)). 2 See National Academy of Science, Managing University Intellectual Property in the Public Interest 19 (2011) (analyzing AUTM Surveys). 3 See generally Michael S. Mireles, States as Innovation System Laboratories: California, Patents and Stem Cell Technology, 28 Cardozo l. rev. 1133 (2006). 4 See Jacob H. Rooksby, When Tigers Bear Teeth: A Qualitative Study of University Patent Enforcement, 46 akron l. rev. 169, 171–2 (2013); Jacob H. Rooksby, Innovation and Litigation: Tensions Between Universities and Patents and How to Fix Them, 15 yale J. l. & teCh. 312, 335–6 (2012–2013); Jacob H. Rooksby, University Initiation of Patent Infringement Litigation, 10 J. Marshall rev. Intell. prop. 623, 625 (2011); Arti K. Rai, John R. Allison, & Bhaven N. Sampat, University Software Ownership and Litigation: A First Examination, 87 n. C. l. rev. 1519, 1520–6 (2008–2009); Scott Shane & Deepak Somaya, The Effects of Patent Litigation on University Licensing Efforts, 63 J. eCon. behav. & org. 739, 740 (2007).

309

310 Research handbook on intellectual property and technology transfer Our research examines whether universities, nonprofits and their licensees exhibit behavior similar to the so-called “patent troll.” While the data appears to demonstrate that universities and other nonprofits may be behaving similar to so-called patent trolls, we caution drawing relatively harsh conclusions from the data. However, we also caution that three developing “currents” could increase the amount of patent enforcement litigation by universities, nonprofits and their licensees, and data should continue to be gathered. At the same time, two “crosscurrents” may stem the tide of patent enforcement. However, for certain university generated patents, it may be too late and those patents may be enforced. There are three currents which may lead to an increase in patent enforcement. First, there may be a shift in the attitudes of professors at universities who are willing to embrace technology transfer and believe that enforcement of patents is part of technology transfer. Indeed, some universities are considering technology transfer activities in the tenure process. Over a generational shift in the ranks of professors, more professors may be inclined to participate in the technology transfer process leading to an increase in patenting and licensing. Second, most university TTOs have not been able to adequately fund their activities. This may push TTOs to enforce patents to generate revenue. Third, there has been a decline in federal funding in terms of real dollars for research, particularly for the National Institutes of Health, which may pressure universities to use patent litigation to generate funding as they seek new revenue streams. A study by the Congressional Research Service explains the decrease in funding in detail. Moreover, President Donald J. Trump’s Office of Management and Budget has again proposed a decrease in funding for research generally and for the National Institutes of Health for FY2019. Two “crosscurrents” may result in less patent enforcement; however, it is unclear how those cross-currents may develop. First, the law of patent eligible subject matter has been in a state of disarray; thus, leading to a situation where some TTOs may be less inclined to invest in patenting technology. Moreover, patent eligible subject matter has contracted. However, some recent US Court of Appeals for the Federal Circuit (“CAFC”) decisions and proposed legislation may result in a substantial broadening of patent eligible subject matter. Additionally, patents are subject to challenge in Inter Partes Review (“IPR”) proceedings at the United States Patent and Trademark Office (“USPTO”). Moreover, IPRs may be limited to not include challenge of pharmaceutical related patents. Second, universities and other nonprofits are concerned about their reputations, particularly the viewpoints of their students—current and potential, alumni and potential donors. For example, the University of California recently changed practices concerning the admission of foreign students based on public outcry concerning availability of seats for California residents.5 Additionally, Yale University changed licensing practices based on concerns expressed by members of their community, including students.6 Universities and nonprofits who do not exercise care in choosing to enforce patents (or control enforcement by licensees

5

See Teresa Watanabe, UC Proposes Its First Enrollment Cap—20%—on Out-of-State Students, l.a. tIMes, Mar. 6, 2017, (“A scathing state audit … found that UC was hurting California students by admitting too many out-of-state applicants.”). 6 See Timothy B. Lee, University Patents Limit Access to Medicine. These Students Want to Change That., wash. post, Nov. 22, 2013; see generally Amy Kapcynski, Samantha Chaifetz, Zachary Katz, & Yochai Benkler, Addressing Global Health Inequities: An Open Licensing Approach for University Innovation, 20 berkeley teCh. l.J. 1031 (2005).

Currents and crosscurrents in litigation of university and nonprofit related patents 311 and assignees) may find themselves in the midst of the public debate concerning the merits of patents and patent trolls. Notably, there may be a risk of widespread enforcement for some university and nonprofit generated patents. First, many universities have apparently licensed their patents to the patent aggregator Intellectual Ventures. We found a small number of university generated patents related to Intellectual Ventures. In the future, the patents licensed to Intellectual Ventures may be packaged and infringement may be alleged based on a cease and desist letter. Second, we found a notable number of foreign entities, such as universities, asserting US patents in the United States. These entities may be less concerned with reputation. It is unclear whether the currents and cross-currents discussed in this Chapter will lead to an increase in patent enforcement of university and nonprofit generated patents that may be similar to patent troll behavior. However, some university generated patents that are subject to preexisting agreements may take away the ability of universities and nonprofits to control their reputation depending on the terms of those agreements. Section I of this Chapter provides an Introduction. Section II provides a review of data concerning university and nonprofit patent litigation, which includes litigation by their licensees concerning university generated patents. Section III reviews the currents and crosscurrents that may result in an increase or decrease in litigation of patents generated by university and nonprofits. Section IV discusses university and nonprofit-related patents that may be litigated despite the crosscurrent of reputational concern, and foreign university and nonprofit patent enforcement. Section V provides a brief conclusion.

II.

DATA CONCERNING UNIVERSITY AND NONPROFIT PATENT LITIGATION

The Dataset concerning university and nonprofit patent litigation is primarily based on the Stanford Nonpracticing Entity Database (SNE Database), which contains a 20% random sample of all patent-asserting cases filed in US district courts between 2000 and 2014, including all cases filed in 2011, 2012 and 2013, and is supplemented by data from the USPTO, Lex Machina’s database on patent litigation, and Google Patents.7 We further used publicly available information to determine which universities are public versus private and are land grant or not. We used publicly available information to determine university endowment size. We created two datasets: one for patents involved in infringement; and another for litigation matters. After cross-checking the various data sources, we found a total sample of 381 patents.8 The SNE Database has 585 cases which involve one or more of the previously noted 381 patents. This section discusses the data generally and further provides specific case examples of university-related patent enforcement that has been criticized in the press as resembling “patent troll-like” behavior.

7 For additional discussion concerning the merging of the databases, see Teo Firpo & Michael S. Mireles, Monitoring Behavior: Universities, Nonprofits, Patents, and Litigation, 71 s. MethodIst u. l. rev. 505, 557–9 (Spring 2018). 8 For a additional discussion concerning the selection of patents, see id.

312 Research handbook on intellectual property and technology transfer Table 14.1

Patents by type of organization Number of patents (and % of total) 252

Average number of cases

US based university

(66%)

4.96

Foreign university

33 (9%)

3.39

US based foundation

35 (9%)

4.34

Foreign foundation

58 (15%)

3.29

US government agency

1 (0%)

1

Foreign government agency

2 (!%)

5

A.

Description of the Dataset and Analysis

The dataset includes patents that are related to US universities primarily, but also includes patents owned by foundations, other non-profits, and non-university related research centers. In examining the data, we attempted to discern whether universities, other non-profits, and their licensees were engaging in strategic behavior that resembles the behavior or so-called “patent trolls.” Importantly, some behavior of so-called “patent trolls” may be very similar to strategic behavior engaged in by practicing entities—such as forum shopping or the enforcement of patents against multiple defendants. Moreover, the data collected seems to indicate that there are significant outliers—perhaps highlighting that some universities, other non-profits, and their licensees are engaging in behavior that may be particularly troubling. We address some of those cases below. Well over half the patents in our sample (66%) were related to a US-based university, with the second largest group being foreign foundations (15% of all patents), followed by foreign universities and US-based foundations (both at 9%). Table 14.1 above presents the full breakdown, as well as the average number of cases associated with a patent of each kind. Furthermore, we examined whether the university was private or public, a land grant university or not, and whether the federal government funded the development of the asserted patent.9 These variables provide a sense of whether an institution is essentially using public funding to engage in troll-like behavior. However, there is a relatively high level of nuance to this analysis. For example, many land grant universities have received public land to serve as the basis of their endowment. However, those universities are often tasked with local economic development and thus, commercializing inventions, which could include enforcing patents. Moreover, some commentators have noted that many universities may have underreported whether a patented invention was covered by government funding or not with a government interest statement.10 Thus, our numbers may be much lower than the actual number of government funded patents that have been asserted by entities in our dataset. Fifty-six percent of the patents in our dataset are associated with private institutions. About 11% relate to foreign universities. Over one-third of US institutions are land grant universities. Thirty-three percent of university-related patents in our dataset are associated with public universities. Almost one-half of the patents in the dataset include a government interest statement, including statements related to the National Institutes of Health (33%), NASA (23%), the National Science Foundation (6%), and DARPA (4%).

9

See Teo & Mireles, supra note 7, at 529–30. General Accounting Office, Technology Transfer: Reporting Requirements for Federally Sponsored Inventions Need Revision 2–14 (1999). 10

Currents and crosscurrents in litigation of university and nonprofit related patents 313 Another variable examined is the venue where the case is filed. Filing patent infringement suits in the US Eastern District of Texas is often associated with so-called “patent troll” behavior. The percentage of cases filed in the Eastern District of Texas in our sample is 11.5%, which is also less than the percentage of cases filed in the same district in the entire SNE Dataset (16.5%). Notably, 46% of the cases filed in the Eastern District of Texas were filed by US universities. However, one-third of those cases were filed by California Institute of Technology. We also examined patent age at litigation.11 If a patent is asserted relatively late in its term, it may indicate that the patent assertor is engaged in troll-like behavior by asserting a patent in an ex post situation—where the technology has already been commercialized.12 We find a relatively high degree of heterogeneity in the age of patents at litigation, as can be seen in the following chart: There is also some variation in patent age at litigation according to the organizational types described above, with the average age of a US-based university patent being 11.3, that of

Figure 14.1

Patent age at litigation

11 For purposes of this article, we defined age at litigation only as the number of years between a patent’s filing and the case filing. Patent age in our prior paper is defined as age of patent to the date of May 2017, which also provides information on the years left for the patent to be asserted. See Teo & Mireles, supra note 7, at 559–60. 12 See generally, Brian J. Love, An Empirical Study of Patent Litigation Timing: Could a Patent Term Reduction Decimate Trolls Without Harming Innovators?, 161 u. pa. l. rev. 1309 (2013).

314 Research handbook on intellectual property and technology transfer

Figure 14.2

Cases by year

a foreign university 10.6, 9.8 for US-based foundations, 9.7 for foreign foundations, 8.1 for US government agencies, and 13.3 for foreign governmental agencies. An additional important variable is the length of the case. So-called patent trolls are sometimes said to engage in a strategy that uses the threat of the cost of litigation to push for an early settlement. A short length of the case could indicate troll-like behavior. Interestingly, the average length of the case for entities in our dataset is similar to that of the cases filed in the Eastern District of Texas (often referred to as a preferred venue to file for so-called “patent trolls”). In fact, the average length for cases filed in the Eastern District of Texas is 1.8 years, while the overall average in the sample is 2.3 years. However, the average length of litigation does vary depending on the entity in the dataset: US-based universities (2.2 years); foreign foundations or other non-profits (1.6 years); foreign universities (2.5 years); and US-based foundations and non-profits (3.4 years). Another set of variables worth exploring are those that can act as a proxy for patent “value.” These include the number of claims, as well as ‘forward’ and ‘backward’ citations (the number of patents that cite the patent as prior art, and the number of previous patents cited, respectively). When compared to measures of value for patents that might be considered highly litigated, such as the sample analyzed by Allison, Lemley and Walker (2009),13 we find some evidence that the patents in our sample have similar measures of value. A similar picture arises when comparing the university, foundation and government related patents in our sample with all patents in the SNE sample that were litigated in the Eastern District of Texas. Thus, we find some evidence that the patents in our sample have, at least according to this measure of value, 13 John R. Allison, Mark A. Lemley, & Joshua Walker, Extreme Value or Trolls on Top? The Characteristics of the Most-Litigated Patents, 158 U. pa. l. rev. 1, 29 (2009).

Currents and crosscurrents in litigation of university and nonprofit related patents 315 a relatively high value, which places them on a somewhat similar level as that of patents found to be aggressively litigated. We reach a similar conclusion when looking at variables that might indicate strategic behavior. In particular, looking at the number of alleged infringers in a case (a higher number might indicate the plaintiff is seeking a way to profit from the litigation), we find that the patents in our sample have a relatively high value when compared to, for instance, non-university related patents filed in the Eastern District of Texas in the SNE sample. Figure 14.2 shows how many cases there were per year in our database. There is a rise in cases filed especially in the years 2012 and 2013. This corresponds with the enactment of the Leahy-Smith America Invents Act, effective as of September 16, 2011. That law effectively requires patent infringement suits to be separately filed against defendants.14 Notably, these are not all the cases involving universities because the SNE sample took a random 20% of all cases in the years 2000–2014, including 2011, 2012 and 2013. B.

Examples from the Data

The following discussion concerns the assertion of patents related to three academic institutions. The first concerns patents licensed by Stanford University to Thermolife International. The second discusses patents directly enforced by Boston University. The third involves patents asserted by the California Institute of Technology. Finally, this section ends with a comparison to filings in the Eastern District of Texas. 1. ThermoLife/Stanford University litigation A set of university generated patents were litigated by an alleged “patent troll,” which received some media attention. The supplement company ThermoLife apparently exclusively licensed numerous patents from Stanford University. Those patents were apparently asserted in 117 different patent infringement actions in one year.15 Stanford’s TTO notes that Thermolife is not “a classic troll” because it sells a product.16 However, a data analytics company points out that Thermolife makes substantially more money from enforcing its patents than from actually selling a product and thus should be classified as a “patent monetization entity.”17 Moreover, some of Thermolife’s patents were found invalid for obviousness.18 A defendant in one of the litigation matters, Hi-Tech Pharmaceuticals, “recovered $913,370.06 in attorneys fees” against Thermolife and Stanford University19 apparently because “ThermoLife and Stanford

14

35 U.S.C. § 299. See Laura Sydell, Bodybuilders Beef Over a Workout Supplement—and a Stanford Patent, all thIngs ConsIdered (July 8, 2016). 16 See As Negatives Rise, PAEs Become Less Attractive to TTOs, teCh. transfer taCtICs (July 2015). 17 See Sydell, supra note 15. 18 See Daniel Siegal, GNC, Supplement Cos, Get ThermoLife Amino Patents Nixed, law 360, Sept. 21, 2016. 19 See Hi-Tech Pharmaceuticals Wins Almost $1 Million in Legal Fees from Thermolife and Stanford After Winning Major Patent Case, IndIa pharMa news, Apr. 10, 2017. 15

316 Research handbook on intellectual property and technology transfer hadn’t done enough research before filing patent infringement litigation … and pursued a ‘file-and-settle’ strategy.”20 In the SNE sample, there are 41 Thermolife/Stanford cases. They were all filed in Northern District of California or the Southern District of California. The average number of alleged infringers is high (13.4), but that is driven mainly by a handful of outlier cases (in particular, one case (“Thermolife International, LLC v. GNC Corporation et al”) which brought a number of patents owned (or previously owned) by Stanford (and had 83 alleged infringers). The length of the cases filed in the SNE sample is relatively long, with an average well over 1,000 days (some of these cases actually went to trial). 2. Boston University litigation Boston University initiated patent infringement suits against almost 40 defendants, including Apple, Samsung, and Microsoft.21 The patents involved were directed to “material used in the production of LEDs (light emitting diodes).”22 The Provost of Boston University stated: “[t]he number of defendants—close to 40—reflects the demand for handheld devices whose performance is improved by Moustakas’ invention and the prevalence of companies using the technology without a license from the University.”23 Notably, Boston University undertook a careful examination of the alleged infringers’ technology and attempted to engage in licensing negotiations with several companies. To verify suspicions of patent infringement, the University retained Dallas-based law firm Shore Chan DePumpo, specialists in intellectual property law who have worked for other major research universities here and abroad (including Penn State, the Universities of Virginia and Texas, and the California Institute of Technology). Shore Chan DePumpo hired consultants and laboratories to “reverse engineer” each defendant’s products. Nijhawan says the experts dismantled devices and “cross-sectioned the LEDs into incredibly tiny samples. These samples were analyzed using several state-of-the art techniques, including transmission electron microscopy, X-ray diffractometry, and secondary ion mass spectroscopy. The analytical data were reviewed by several independent scientific experts before any legal allegations were made.” Even then, the University filed suit “only after meeting with several alleged infringing manufacturers and affording them a fair opportunity to take a license at a reasonable royalty,” he says. “Several of those manufacturers refused to negotiate a reasonable royalty, thereby forcing Boston University to proceed with litigation.”24 In 2015, a jury found in favor of Boston University and awarded $13 million.25 Notably, an article states that the lawyers for Boston University would receive half of the award because

20

See Nicola Narea, Stanford, ThermoLife Seek to Slash Fee Award in IP Suit, law 360, June 2,

2017. 21 See BU Brings Lawsuit Against Leading Tech Firms for Patent Infringement, b.u. today, Sept. 9, 2004. 22 Id. 23 Id. 24 Id. 25 See Joel Brown, BU Wins $13 Million in Patent Infringement Suit, b.u. today, Dec. 7, 2015.

Currents and crosscurrents in litigation of university and nonprofit related patents 317

Figure 14.3

Length of litigation in ED Texas

they took the case on a contingency fee basis.26 In July 2018, the CAFC determined that claim 19, one of Boston University’s patents, was invalid for lack of enablement.27 In the SNE sample there are only six cases directly related to Boston University, all related to US Patent number 5686738. All cases were filed within the span of three months in the same venue (D. Mass). The average age of the patent was a little over 18 years and relatively close to expiry. The average length of the case is 603 days, although two of those cases (against Motorola and Apple) ended within 225 days. California Institute of Technology litigation 3. There are 42 cases directly related to the California Institute of Technology (a further 19 are for a patent previously associated with them). The average age at litigation of the patents involved is not extraordinary (10.3 years). The average number of alleged infringers is relatively high at 9.5 (for comparison, it was 4.2 in the whole sample, and 4.8 among US-based universities). The number of cases filed in Eastern District of Texas is also high: 10 of 42 (24%, higher than the sample overall, which is 11.5%). The length is also quite short: only 432 days on average (almost half that of the overall sample, at 827). There does not seem to be a strong difference in the “value” metric of California Institute of Technology patents—for example, they have on average 22.7 claims, compared to 25.6 overall. Finally, for comparison, the cases filed in the Eastern District of Texas had the following characteristics. Average number of alleged infringers: 7.1. The mean age was 10.1. The

26

Id. See Trustees of Boston University v. Everlight Electronics Co., LTD., et al, 896 Fed. Cir. 1357 (July 25, 2018); see also Steve Brachmann, CAFC Invalidates Boston University Patent Claim for Lack of Enablement, Ip watChdog blog, July 31, 2018. 27

318 Research handbook on intellectual property and technology transfer average number of cases was 9.3. The average length was 515—although this is due to a number of outliers, as Figure 14.3 shows.

III.

CURRENTS AND CROSSCURRENTS IMPACTING PATENT ENFORCEMENT OF UNIVERSITY AND NONPROFIT GENERATED PATENTS

There are at least three reasons why patent enforcement by universities and nonprofits may increase in the future. There are also at least two reasons why patent enforcement may decrease. This section will discuss the reasons why university and nonprofit related litigation may increase and decrease. There are three reasons why there may be an increase in patent litigation involving university and nonprofit related patents in the future. The first reason why patent enforcement may increase is that some universities have begun to change their tenure policies to include consideration of commercialization activities performed by professors. Second, most TTOs have not been able to generate enough revenue to cover their own costs. Third, the federal government has been reducing funding for research. First, some universities are changing their tenure policies to include the consideration of patenting and other commercialization activities. Moreover, some scholars have proposed that university technology transfer activities should be made part of the tenure process.28 Indeed, at least one state has mandated its universities to include commercialization activities as factors relevant to the granting of promotion and tenure.29 Additionally, a vendor for commercialization education recently announced a webinar about how to place commercialization requirements in the tenure process.30 As examples, the University of South Florida and Purdue University (West Lafayette) have both included commercialization language in their tenure policies. The University of South Florida includes the following language in its tenure policy: “Where appropriate … consideration will be given to … the demonstrable impact of research through inventions, development and commercialization of, and technology transfer.”31

28 See Paul R. Sanberg, Morteza Gharib, Patrick T. Harker, Eric W. Kaler, Richard B. Marchase, Timothy D. Sands, Nasser Arshadi, & Sudeep Sarkar, Changing the Academic Culture: Valuing Patents and Commercialization Toward Tenure and Career Advancement, 111 proCeedIngs nat’l aCad. sCIs. 6542, 6544 (May 6, 2014) (“[Sixteen] United States and Canadian universities … consider patents and commercialization in tenure and career advancement decisions …”); Judy Genshaft, Jonathan Wickert, Bernadette Gray-Little, Karen Hanson, Richard Marchase, Peter E. Schiffer, & R. Michael Tanner, Consideration of Technology Transfer in Tenure and Promotion, 17 teCh. InnovatIon 197, 197 (2016) (working as APLU Task Force on Tenure, Promotion, and Technology Transfer and recommending “technology transfer activities should be explicitly included among the criteria relevant for promotion and tenure at the university … as appropriate to the respective disciplines”). However, those activities are considered “optional” and not “required.” Id. at 201. The authors also provide examples of several institutions with language concerning technology transfer in their tenure policies: Iowa State University, Texas A&M, and University of Arizona. Id. at 203–204. 29 Rachel Abbey McCafferty, State is Pushing Universities to Bring Research to Market, CraIn’s Cleveland bus., Mar. 5, 2017. 30 Michael S. Mireles, Basic Research—Soon to be a Thing of the Past, Ip fInanCe blog (Feb. 5, 2013). 31 See Guidelines for Tenure and Promotion, u. s. fla.

Currents and crosscurrents in litigation of university and nonprofit related patents 319 Purdue University’s tenure policy states: “a record would likely include accomplishments such as … patents, licenses, prototypes, and entrepreneurship activities that moves products from the bench to the marketplace; these activities are particularly encouraged in disciplines where there is a focus on addressing social needs.”32 One concern about the Bayh-Dole Act’s unintended consequences is that there may be a shift from basic to applied research, although in some fields the difference between the two may be unclear.33 A monetary incentive through the possibility of future royalties provides an incentive to move to applied research. Adding an incentive through the tenure process will likely increase the likelihood that more professors may seek to engage in commercialization type activities. Patenting and licensing has generally been on the rise by universities.34 These activities likely will continue with increased incentives. Moreover, as Professor Mark Lemley noted, this issue is one that concerns the generational change in faculty ranks.35 As new faculty are socialized into their positions with expectations concerning commercialization, those faculty may be more inclined to participate in those activities than their predecessors who did not enter the academy with those expectations. While more patenting and licensing may not lead to more litigation, there are other reasons to be concerned. Second, most TTOs do not generate enough revenue to cover their own costs.36 This means that TTOs may be looking for additional revenue to ensure their own survival. One way to generate that revenue is through an increase in patenting and licensing, which may lead to patent enforcement. The desire to obtain licensing revenue may lead to aggressive patent enforcement and the failure to control the litigation behavior of licensees. Finally, the third current is the decrease in federal funding for research and development. Notably, state funding for many state universities has be reduced as well, but the decrease in federal funding in real dollars may result in universities looking to supplement their research budgets through patent licensing and enforcement. One important conduit for federal funding is the National Institutes for Health (NIH).37 The Congressional Research Service recently released a report entitled “NIH Funding: FY1994–FY2019” that details the decrease in real dollars for research funding for the NIH from the high level of over $43 billion (in real 2019

32

See Criteria for Tenure and Promotion for the West Lafayette Campus 2, u. purdue, Jan. 1, 2016. See William M. Landes & Richard A. Posner, The Economic Structure of Intellectual Property Law 316 (2003) (“Being able to earn substantial income from patent licensing has, it appears, induced universities to substitute away from basic research, and the result may have been a net social loss.”); see Rebecca S. Eisenberg, Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research, 82 va. l. rev. 1663, 1714 (1996); see also Pierre Azoulay, Waverly Ding, & Toby Stuart, The Impact of Academic Patenting on the Rate, Quality, and Direction of (Public) Research 1 (2006) (“[S]urveys of academic scientists have found that patenting skews scientists’ research agendas towards commercial priorities, causes delay in the public dissemination of research findings, and crowds out effort devoted to producing public research results.”). 34 See Michael S. Mireles, The Bayh-Dole Act and Incentives for the Commercialization of Government-Funded Invention in Developing Countries, 76 uMkC l. rev. 525, 530–1 (2007). 35 See Mark A. Lemley, Are Universities Patent Trolls?, 18 fordhaM Intell. prop., MedIa, & ent. l.J. 611, 620–1 (2008). 36 See Walter D. Valdivia, University Start-Ups: Critical for Improving Technology Transfer, brookIngs Inst. report (Nov. 20, 2013) (“[W]ith 84% [of] universities operating technology transfer in the red, 2012 was a good year because over the last 20 years, 87% did not break even.”). 37 Judith A. Johnson, NIH Funding: FY 1994–2019, 5, Cong. res. serv. (May 2, 2018). 33

320 Research handbook on intellectual property and technology transfer dollars) in 2003.38 There has essentially been a double-digit decrease in funding in real dollars every year since 2008.39 In 2008, NIH funding dropped by 11% from the 2003 high. In 2013, 2014, and 2015, NIH funding dropped each year over 20% from the 2003 high.40 In 2018, the Trump Administration proposed to essentially cut the NIH budget in real dollars for 2019 about 19% from the 2003 high.41 The Trump Administration also proposed in 2018 to cut the Department of Energy’s applied R&D budget, “including a 67 percent cut to the $2 billion Office of Energy Efficiency and Renewable Energy.”42 The decreases in federal funding could result in less federal research funding available for university and other nonprofit researchers which may place pressure on universities to develop new sources of funding, such as through licensing and patent enforcement. While those currents may result in additional enforcement of patents, there are at least two reasons why patent enforcement may decrease. The first reason is the change in patent law doctrine concerning patent eligible subject matter and Inter Partes Review proceedings at the USPTO to challenge patents. The second reason is the prospect of shaming universities who behave like patent trolls by organizations such as the Electronic Frontier Foundation. First, the US Supreme Court has issued numerous decisions which impact so-called patent troll behavior.43 However, the most important decisions likely address patent eligible subject matter, including Mayo Collaborative Services v. Prometheus Laboratories and Alice Corp. v. CLS Bank International.44 The Supreme Court’s recent patent eligible subject matter decisions have resulted in a sea change in approach to determining if a particular invention can be patented in the first instance under Section 101 of the Patent Act. The two referenced cases impacted diagnostics and software, respectively.45 Importantly, both cases focus the analysis of patent eligible subject matter on the presence of an exception to patentability in the asserted claim and whether the claim includes an “inventive concept.”46 One alleged upshot of both of these cases is that what is considered patentable has contracted. However, the CAFC has issued numerous decisions which arguably have provided some breathing room for the patentability of certain inventions.47 David Kappos, former Director of the USPTO, has referred to the current state of the law of patent eligible subject matter as “a real mess,”48 despite the best efforts of the USPTO to provide guidance on navigating the doctrine for

38

Id. Id. 40 Id. 41 Id. Notably, Congress did not accept the President’s proposal for FY2018; however, the resulting budget was still significantly below the 2003 high in terms of real dollars. 42 See FY19 Budget Request: DOE Applied R&D Slashed Again, aM. Inst. physICs, Mar. 19, 2018. Congress, however, will apparently direct the DOE to use funding for applied R&D. See Colin Cunliff, Department of Energy RD&D Appropriations, Fiscal Year 2019, Info. teCh. & InnovatIon found., June 26, 2018. 43 See Teo & Mireles, supra note 7, at 524–7. 44 Mayo Collaborative Services v. Prometheus Labs, Inc., 566 U.S. 66, 73 (2012); Alice Corp. v. CLS Bank International, 134 S.Ct. 2347, 2352 (2014). 45 See generally Mayo Collaborative Services, 566 U.S. 66, 73; Alice Corp, 134 S.Ct. 2347, 2352. 46 See generally Mayo Collaborative Services, 566 U.S. 66, 73; Alice Corp, 134 S.Ct. 2347, 2352. 47 See, e.g., Enfish, LLC. v. Microsoft Corp., 822 F.3d 1327, 1335 (2016). 48 See David Kappos Calls for Abolition of Section 101, nat’l l. rev. (Apr. 14, 2016) . 39

Currents and crosscurrents in litigation of university and nonprofit related patents 321 patentees.49 Importantly, two CAFC judges have called for Congress to step in and clarify the doctrine.50 Several professional organizations have proposed legislation to Congress which would essentially remove the Mayo/Alice analysis from consideration of patent eligible subject matter.51 Depending on the evolution of patent eligible subject matter, this doctrine may be used to lessen the scope of what is patentable and result in fewer patents and maybe even less litigation. Congress also passed the Leahy-Smith America Invents Act, which allows for Inter Partes Review proceedings to be instituted at the USPTO.52 This legislation allows for the challenge of issued patents on specific grounds.53 Notably, IPRs has been used frequently. Some universities and nonprofits may not assert their patents for fear of an IPR proceeding filed against their patents. This may reduce the likelihood of an increase in litigation. However, public universities may have immunity from an IPR challenge.54 That immunity may be waived. Finally, another reason that litigation may be curbed in the future is the prospect that universities and other nonprofits may be “shamed” if they or their licensees file enforcement actions. For example, the Electronic Frontier Foundation (EFF) has been publishing the names of universities whose patents have been the basis of “troll-like” litigation. The EFF’s “stupid patent of the month” for March of 2017 was “US Patent No. 8,473,532 (‘532 patent), ‘Method and apparatus for automatic organization for computer files.’”55 That patent “began its life with publicly-funded Louisiana Tech University [and was] sold to Micoba LLC, a company that has all the indicia of a classic patent troll” that brought suit against numerous defendants.56 Moreover, the EFF has started a “Reclaim Invention” program attempting to encourage universities to not sell patents to so-called patent trolls, as well as proposed model legislation designed to penalize universities who sell patents to so-called patent trolls.57 The threat of public shame concerning patent related enforcement activity may chill universities and other nonprofits from asserting their patents or ensure that they retain the power to influence the choice of whether to enforce a patent after the patent is licensed or sold. The three currents concerning patent enforcement are relatively strong and raise the prospect of increased patent litigation. Moreover, the two crosscurrents may not effectively and substantially reduce litigation of university and nonprofit related patents, including litigation by their licensees. For example, new CAFC case law and potential new legislation from

49

See Subject Matter Eligibility, USPTO, available at https://www.uspto.gov/patent/laws-and -regulations/examination-policy/subject-matter-eligibility (last visited Sept.1, 2018). 50 See Michael S. Mireles, Judges Lourie and Newman of the Federal Circuit Critique Mayo/Alice and Myriad, Ip fIn. blog (June 20, 2018). 51 See Intellectual Property Owners Association (“IPO”), Proposed Amendments to Patent Eligible Subject Matter under 35 U.S.C. § 101, USPTO 1; American Intellectual Property Law Association (“AIPLA”), AIPLA Legislative Proposal and Report and Patent Eligible Subject Matter, USPTO 4; American Bar Association, (“ABA”), Supplemental Comments Related to Patent Subject Matter E ligibility, USPTO 3–4. 52 See 35 U.S.C. §§ 311–319. 53 See id. 54 See Covidien LP v. University of Florida Research Foundation, IPR2016-01274; IPR2016-01275; IPR2016-01276 (Jan. 25, 2017). 55 See Daniel Nazer, Stupid Patent of the Month: Storing Files in Folders, eleCtronIC frontIer found. deeplInks blog (Mar. 31, 2017). 56 See id. 57 See Reclaim Invention, eleCtronIC frontIer foundatIon (last visited Aug. 30, 2018).

322 Research handbook on intellectual property and technology transfer Congress may significantly lessen the importance of patent eligible subject matter as a way to invalidate patents and allow the early challenge of patents. Additionally, there is proposed legislation in Congress to disallow the use of IPRs against pharmaceutical related patents.58 Finally, it is unclear whether shaming will successfully push universities and nonprofits to choose not to enforce their patents. As discussed, even after the negative press concerning so-called patent trolls, Thermolife and Stanford University, as well as Boston University, proceeded with enforcing their patents.

IV.

IS IT TOO LATE FOR SOME UNIVERSITY AND NONPROFIT RELATED PATENTS?

There are, at least, two groups of university and nonprofit generated patents that may be enforced in ways similar to those exhibited by the so-called patent troll. First, according to Professor Robin Feldman and Tom Ewing, many universities have entered into agreements to license their patents to Intellectual Ventures, the well-known patent aggregator. Second, we found a relatively significant amount of patent enforcement in the United States by foreign universities and nonprofits. First, in The Giants Among Us, Professor Robin Feldman and Tom Ewing provide a significant amount of information concerning the secretive mass-patent aggregator Intellectual Ventures.59 Intellectual Ventures “has announced that it has relationships with some 400 universities,” wherein for some of those universities, Intellectual Ventures may have essentially “optioned” future inventions.60 Moreover, the authors state that Intellectual Ventures has entered in agreements to file Patent Cooperation Treaty patent applications for inventions patented by universities in developing countries.61 Presumably, the agreements would allow Intellectual Ventures to enforce those patents in other countries such as the United States. It is unclear whether the agreements concerning university related patents reserve sufficient rights for the university to control or influence patent enforcement by Intellectual Ventures. For patents subject to those agreements, Intellectual Ventures may choose to vigorously enforce those patents. Moreover, Intellectual Ventures may assert patent portfolios which contain some university generated patents and some that do not. This may serve to make it harder to determine whether a university generated patent is actually litigated. Intellectual Ventures also uses the privateering method of enforcement—essentially using a third party to enforce patents that may not be as careful as Intellectual Ventures.62 It is unclear what rights are retained by Intellectual Ventures to control patent enforcement. For the current and future patents licensed to Intellectual Ventures, universities and nonprofits may not have much influence over the decision to enforce those patents. Second, we found a relatively significant number of US patents asserted by foreign universities and other nonprofits. For these foreign universities and other nonprofits, a concern with

58 See Michael S. Mireles, Disincentivizing the Use of IPRs and PGRs Against Pharmaceutical and Biologic Patents, Ip fIn. blog (June 21, 2018). 59 See Robin Feldman & Tom Ewing, The Giants Among Us, 2012 stan. teCh. l. rev. 1 (2012). 60 Id. at 38. 61 Id. at 39–40. 62 Id. at 62–3.

Currents and crosscurrents in litigation of university and nonprofit related patents 323 their reputation in the United States may not be as great as for US universities and nonprofits. Indeed, the prospect of garnering significant revenue for their institution through patent enforcement and licensing may be a very strong motivator. A recent report noted that there are around 30 foreign universities in the list of the top 100 universities receiving US patents.63 Moreover, we found that 25% of the litigated patents in our dataset involved a foreign university (9%), foreign foundation (15%), or foreign government agency (1%). As discussed, the analysis concerning these patents may be different from US patents generated from US universities and other nonprofits. For example, it is likely that those patented inventions were not funded by the US federal government (although a patent funded by the US federal government may have a co-inventor who may not be a US citizen). Also, the foreign university and nonprofit likely does not benefit the same way as a US university does, whether private or public, land-grant or not. Given the level of patenting by foreign universities and a probable lack of shaming potential concerning the enforcement of those patents, foreign universities and other nonprofits may not reluctantly enforce patents.

V.

CONCLUSION

Given the data concerning the enforcement of university and nonprofit generated patents, we believe that continued enforcement of those patents should be carefully monitored. Many of the inventions covered by those patents that were litigated were funded by the government and some were asserted by land-grant and public universities. Importantly, the university and nonprofit generated patents appear to share some of the characteristics of valuable patents. Moreover, the Bayh-Dole Act does ensure that inventors receive royalties from the licensing and exploitation of their patents, although those inventions may be government funded. While several currents indicate a possible increase in litigation in the future, there are other crosscurrents which may slow patent enforcement. Without question, the future for US patent law is anything but clear, and litigation involving university and nonprofit related patents should continue to be monitored.

63 Top 100 Worldwide Universities Granted U.S. Utility Patents 2016, nat’l aCad. (2017) .

of Inventors

15. Is patent enforcement efficient?1 Mark A Lemley and Robin Feldman

I.

INTRODUCTION

Of the many governmental activities we undertake in this country, few are as purely and explicitly utilitarian as the patent system.2 The patent system exists to bring about a particular result, rather than out of a sense of an inventor’s moral rights or as a matter of equity.3 From the many commercial activities that might otherwise be open to anyone, we remove some, for a limited period of time, in the hope that dedicating them to the province of a few will redound to the benefit of us all.4 The benefit—in other words, what the patent system is designed to promote—is commonly referred to as “innovation.” The traditional utilitarian story supporting the patent system is that the lure of patent rights encourages invention that would not otherwise occur, or at least would occur later but for the patent.5 The invention the system is designed to promote is not what is known in science as “basic research,” such as an understanding of how nature works or what forces propel the universe. After all, for more than a century, the courts have reminded us that the proper subject matter of a patent does not include laws of nature, natural phenomena, or abstract ideas—no matter how valuable and essential to the progress of science these may be.6 Rather, the patent system is aimed at protecting “applied” inventions, or innovations, that are deployed in the world. Only when broad and basic principles are reduced to a particular practice and applied in a specific endeavor will they be eligible for protection.7 The patent system’s focus is consistent with economic literature, which distinguishes invention—an idea—from innovation—turning an idea into a viable product. The patent

1

Ideas presented in this Chapter first appeared in Boston University Law Review, 98 B.U. L. Rev. 649 (2018). 2 For a discussion of the version of utilitarianism applied in the patent system, see Robin Feldman, Rethinking Patent Law 76–8 (2012). For debate on this point, compare Robert P. Merges, Justifying Intellectual Property 102–36 (2011) (arguing for patent rights on moral basis), with Mark A. Lemley, Faith-Based Intellectual Property, 62 uCla l. rev. 1328, 1336–8 (2015) (critiquing those moral claims). 3 See generally Lemley, supra note 2 (arguing that justifications of IP rights based on moral claims are unpersuasive). 4 Robin Feldman, Intellectual Property Wrongs, 18 stan. J.l. bus. & fIn. 250, 252 (2013). 5 William M. Landes & Richard A. Posner, The Economic Structure of Intellectual Property Law 294–310 (2003); see also John F. Duffy, The Marginal Cost Controversy in Intellectual Property, 71 U. ChI. l. rev. 37, 52–53 (2004); Steven Shavell & Tanguy van Ypersele, Rewards Versus Intellectual Property Rights, 44 J.l. & eCon. 525, 530 (2001). 6 Alice Corp. v. CLS Bank Int’l, 134 S. Ct. 2347, 2354 (2014). 7 Bilski v. Kappos, 561 U.S. 593, 611 (2010) (explaining “that while an abstract idea, law of nature, or mathematical formula [can]not be patented, ‘an application of a law of nature or mathematical formula to a known structure or process may well be deserving of patent protection’” (quoting Diamond v. Diehr, 450 U.S. 175, 187 (1981))).

324

Is patent enforcement efficient? 325 system encourages not just invention in the abstract, but the creation of new products. This is the “[p]rogress” of the “useful [a]rts” mentioned in the Patent Clause of the Constitution.8 The focus on innovation, not simply invention, is particularly important with the emergence of the modern non-practicing entity (“NPE”) business model. Colloquially known as “patent trolls,” NPEs are those entities whose core activity involves licensing or litigating patents, as opposed to making products.9 By all accounts, the modern NPE business model has expanded rapidly since its emergence over the last two decades, an expansion that is particularly evident in the context of litigation.10 Different scholars slice the numbers differently, with some excluding NPEs organized as trusts as well as individual inventors and others excluding “failed startups,” for example.11 When the broader definition is applied, however, the data are remarkably consistent across studies, with all showing that NPEs now account for the majority of patent lawsuits filed in the United States.12 Our goal in this Chapter is to assess whether those lawsuits are efficient. By “efficient,” we do not mean “are the lawyers working as quickly and cheaply as they could?” Rather, our goal is to determine under what circumstances the enforcement of patent rights might benefit society. As we demonstrate, while there are various ways patent enforcement might serve utilitarian ends, those approaches all involve some sort of technology transfer from the inventor to implementers or to the public at large. Without that technology transfer, patent enforcement represents a pure cost to society and a tax on innovation. Unfortunately, a large fraction of the patent lawsuits filed today fall in the category of pure costs to society.

8

u.s. Const. art. I, § 8, cl. 8. Robin Feldman, Federalism, First Amendment and Patents: The Fraud Fallacy, 17 ColuM. sCI. & teCh. l. rev. 30, 32 (2015) (defining patent trolls and NPEs as those whose core business involves licensing and litigating stripped patent rights as opposed to making products with those patents); John R. Allison, Mark A. Lemley & David L. Schwartz, How Often Do Non-Practicing Entities Win Patent Suits?, 32 berkeley teCh. l.J. 235, 240 (2017); see also Robin Feldman, Patent Demands and Startup Companies: The View from the Venture Capital Community, 16 yale J.l. & teCh. 236, 244–53 (2014) (describing nuances in use of various terms—including non-practicing entity, patent assertion entity, and patent monetization entity—concluding that a broad definition of those whose core activity involves licensing and litigating patents rather than making products is best, and suggesting that monetizer might be a better term than NPE). 10 Allison, Lemley & Schwartz, supra note 9, at 237 (noting that NPEs “account for a majority of all defendants sued for patent infringement” and may “win both larger judgments or larger settlements than … operating companies”). 11 Id. at 240. 12 Compare Robin Feldman, Tom Ewing & Sara Jeruss, The AIA 500 Expanded: The Effects of Patent Monetization Entities, 17 uCla J.l. & teCh. (Issue 2) 1, 7 (2013) (“[I]n 2012, litigation by patent monetization entities represented a majority of the patent litigation filed in the United States.”), and Colleen Chien, Assistant Professor, Santa Clara Univ., Patent Assertion Entities, Presentation to the FTC/DOJ Workshop on PAEs (Dec. 10, 2012), available at http://papers.ssrn.com/sol3/papers.cfm ?abstract_id=2187314 (last visited Oct. 17, 2019) [https://perma.cc/863E-RU2B] (using data from RPX Corporation and concluding that percentage of litigation by non-practicing entities in 2012 had reached 61%), with Christopher A. Cotropia, Jay P. Kesan & David L. Schwartz, Unpacking Patent Assertion Entities (PAEs), 99 MInn. l. rev. 649, 655 (2014) (using narrower definition of NPEs and finding no real increase in NPE litigation when comparing years 2010 and 2012, but also noting that “when we repackage all [NPEs] into a single category, they are responsible for a majority of [patent lawsuits] in 2012”); see also Matthew Sag, IP Litigation in the U.S. District Courts: 1994–2014, 101 Iowa l. rev. 1065, 1081 (2016) (finding that patent litigation volume doubled from 2010 to 2012). 9

326 Research handbook on intellectual property and technology transfer

II.

INNOVATION-RELATED JUSTIFICATIONS FOR NPES

Consistent with the utilitarian goals of the patent system, all of the arguments suggesting NPEs benefit society rest on their contribution, either directly or indirectly, to the creation of products somewhere in the system. NPEs, unlike practicing entities, do not deploy the technology in the world themselves, but that does not answer the question of whether they contribute to innovation and the creation of new products. To promote innovation, however, they must not only invent, but that invention must lead to the creation of products by someone, somewhere in the system, at some point. NPEs may be acting as middlemen, transferring technology to those who would implement it; or they could be collecting revenue from those who copied their invention and implemented it. Neither possibility, however, appears broadly supported by the available evidence. There is substantial literature that calls into question whether the patent system in general encourages innovation that would not otherwise happen. The facts that most significant innovations are simultaneously created by two or more people working independently13 and that in most industries virtually all patent enforcement is done against independent inventors14 cast significant doubt on the claim that the innovations would not have happened but for the lure of a patent. The issue is, however, complicated by the very different characteristics of different industries. There may be industries in which invention is so complex and uncertain that it would not be undertaken without patent protection.15 But there also seem to be industries— perhaps most of them—in which the patent system does not seem to be driving new invention, and may even be retarding it.16 That might lead one to question the patent system as a whole,17 or at least the traditional innovation-based justification for it. In addition to doubts about how well patents in general actually drive innovations that would not otherwise have occurred, the evidence casts significant doubt on the efficacy of the patent disclosure as a way of disseminating ideas and leading to the creation of products. While writing down and publishing a description of the invention has long been a quid pro quo for a patent, in the modern world there is good reason to think that engineers in many fields rarely read patents in order to learn about a technology.18 There are many reasons for

13

Mark A. Lemley, The Myth of the Sole Inventor, 110 MICh. l. rev. 709, 711 (2012). Christopher A. Cotropia & Mark A. Lemley, Copying in Patent Law, 87 n.C. l. rev. 1421, 1424 (2009). 15 Dan L. Burk & Mark A. Lemley, The Patent Crisis and How the Courts Can Solve It 80–1 (2009) (arguing that in pharmaceutical industry, investment in research would be likely to drop substantially without effective patent protection due to high costs of innovation and relative ease of copying inventions). 16 James Bessen & Michael J. Meurer, Patent Failure: How Judges, Bureaucrats, and Lawyers Put Innovators at Risk 21–4 (2008) (arguing that patent system particularly fails with respect to software patents). 17 See generally Michele Boldrin & David K. Levine, Against Intellectual Monopoly (2008) (arguing that “intellectual property is an unnecessary evil”). 18 Mark A. Lemley, Ignoring Patents, MICh. st. l. rev. 19, 21–2 (2008) (“Companies and lawyers tell engineers not to read patents in starting their research, lest their knowledge of the patent disadvantage the company by making it a willful infringer.”); Lemley, supra note 13, at 711 (“[D]isclosure theory, which justifies the grant of patents on the assumption that scientists read and learn from them, fails to grapple with the way learning occurs in the real world.”); John M. Olin, Note, The Disclosure Function of the Patent System (or Lack Thereof), 118 harv. l. rev. 2007, 2025–26 (2005) (“[E]ngineers often 14

Is patent enforcement efficient? 327 this. Lawyers at many companies discourage their engineers from reading patents for fear of increasing legal liability.19 The quality of the disclosure in the patent may be poor, particularly in the information technology (“IT”) industries.20 There are simply too many patents in many fields to possibly keep up with,21 and 600,000 more applications are filed every year.22 And in a fast-moving industry like IT, a delay of several years between invention and disclosure may make the technology described obsolete by the time anyone could read the patent.23 Further, economic literature suggests that effective technology transfer—in other words, transfer that can lead to commercialization—requires more than just reading a patent.24 Such transfer generally must include not only the information publicly available in the patent, but also the transfer of know-how, complementary assets, and other peripheral disclosures.25 Thus, if patents actually drive innovation by third parties we would expect to see not simply patenting but business transactions that involve the transfer of other types of information assets.26 Alternatively, NPEs could drive innovation if they served as efficient middlemen, connecting those who invent but whose inventions have not been deployed with those who can produce something from that invention. Several people have argued that NPEs serve this role.27 Here,

find it difficult to extract useful information from the written description [in a patent application], which … weakens the disclosure value of patents.”). By contrast, in some fields patents may provide more useful guidance to engineers. Lisa Larrimore Ouellette, Do Patents Disclose Useful Information?, 25 harv. J.l. & teCh. 545, 547 (2012) (finding that in field of nanotechnology, patents contain “useful, nonduplicative technical information”); Lisa Larrimore Ouellette, Who Reads Patents?, 35 nature bIoteCh. 421, 422 (2017). 19 Mark A. Lemley & Ragesh K. Tangri, Ending Patent Law’s Willfulness Game, 18 berkeley teCh. l.J. 1085, 1100–1 (2003) (“Once a company becomes aware of a patent, it has an obligation to obtain a written opinion of counsel or risk later being held a willful infringer … [L]awyers regularly advise their clients not to read patents if there is any way to avoid it.”); Lemley, supra note 18, at 21–2. 20 Burk & Lemley, supra note 15, at 158 (citing software patents as likely to “be supported by very little in the way of detailed disclosure” due to the Federal Circuit’s relaxed enablement requirement); Robin Feldman, Rethinking Patent Law, 90–123 (2012) 90–123 (describing how the last fifty years of cases related to computer technology have led to approval of patents containing limited useful information). 21 Lemley, supra note 18, at 19 n.1 (noting that more than one-third of all patents issued as of 2008 had been issued in the preceding twenty years). 22 David Rogers, United States Patent Application Filings Exceed 600,000 for the Second Straight Year, law.CoM (June 4, 2015), available at https://perma.cc/A4N5-5X32 (last visited Oct. 17, 2019). 23 Robin Feldman & Mark A. Lemley, Do Patent Licensing Demands Mean Innovation?, 101 Iowa l. rev. 137, 159–60 (2015). 24 Id. at 155 n.40. 25 James Bessen, Learning by Doing: The Real Connection Between Innovation, Wages, and Wealth 3 (2015); Jason Rantanen, Peripheral Disclosure, 74 U. pItt. l. rev. 1, 7 (2012) (defining “peripheral disclosure” as disclosure of information that would not occur but for incentives provided by the patent system); David J. Teece, Profiting from Technological Innovation: Implications for Integration, Collaboration, Licensing and Public Policy, 15 res. pol’y 285, 293 (1986) (noting that knowledge and competencies necessary to produce even “modestly complex technologies” is quite large and difficult for one company to maintain on its own). 26 Acquisition is one means of technology transfer. See generally John F. Coyle & Greg D. Polsky, Acqui-Hiring, 63 duke l.J. 281 (2013). 27 Daniel A. Crane, Intellectual Liability, 88 tex. l. rev. 253, 286–7 (2009); B. Zorina Khan, Trolls and Other Patent Inventions: Economic History and the Patent Controversy in the Twenty-First Century, 21 geo. Mason l. rev. 825, 832–3 (2014); David F. Spulber, “Intellectual Property and the Theory of

328 Research handbook on intellectual property and technology transfer too, an innovation benefit requires technology transfer. The social benefit of the middleman story depends on the middleman providing something of value to the implementer.28

III.

DOES PATENT ENFORCEMENT INVOLVE TECHNOLOGY TRANSFER?

In short, then, the traditional justifications for NPEs contributing to social welfare all involve some form of technology transfer or learning dissemination by which the NPE or the patent it holds teaches the implementer a technology it did not otherwise possess. Practicing entities can benefit social welfare without technology transfer by making and selling the invention directly; NPEs cannot. Early evidence testing the positive impact of NPEs on commercialization goals is not encouraging. The evidence is largely observational in nature, flowing from small sample studies, with all of the attendant limitations. Nevertheless, the data provide a useful window into the NPE business model and suggest approaches for generalizable analyses. We provide survey evidence of the direct measure of new product creation as a result of patent assertions by NPEs. We also tested commercialization effects by measuring other markers of potential innovation, such as technology transfer beyond the patent. Including such markers creates a more dynamic picture of the potential for future commercialization, even if that commercialization has yet to occur. We know that actual technology transfer happens within the patent system in the ex ante context.29 Both practicing entities and some NPEs engage in ex ante technology transfer. In particular, universities and inventors create alliances with companies that can more easily develop and commercialize their inventions through joint ventures and other types of technology and research sharing agreements.30 These agreements frequently occur before a patent issues or even before any of the parties file for a patent.31 Notably, these agreements involve technology transfer.32 Universities and other inventors in these deals provide new technology

the Firm” in Perspectives on Commercializing Innovation 9, 31 (F. Scott Kieff & Troy A. Paredes eds., 2012). 28 Thus, we do not share what some have characterized as skepticism about IP licensing generally. See generally Jonathan M. Barnett, Why Is Everyone Afraid of IP Licensing?, 30 harv. J.l. & teCh. 123 (2017). To be honest, we’re not sure that most of the people Barnett points to do either. Barnett correctly articulates ways that IP licensing can improve firm performance. See generally id. But all those ways involve ex ante technology transfer. See also Oskar Liivak, Private Law and the Future of Patents, 30 harv. J.l. & teCh. 33, 48–52 (2016) (encouraging a focus on ex ante technology transfer as the basis for patent law). The skepticism about IP licensing is more properly understood in our view as skepticism about licensing IP rights without any technology transfer. 29 Ashish Arora, Andrea Fosfuri & Alfonso Gambardella, Markets for Technology: The Economics of Innovation and Corporate Strategy 116–17 (2001); See generally Colleen V. Chien, The Market for Software Innovation Through the Lens of Patent Licenses and Sales, 33 berkeley teCh. l.J. (forthcoming 2018) (manuscript on file with authors) (documenting technology transfer in many software licenses, but also a large number of software patent licenses without technology transfer). 30 Feldman & Lemley, supra note 23, at 155–6. 31 Id. 32 Id.

Is patent enforcement efficient? 329 to those in a position to implement it.33 And that technology often includes trade secrets and know-how beyond the to-be-patented technology itself.34 Further, technology transfer can occur informally, by the communication of information at scientific conferences, through journal articles, and even through commitments to open sharing of patented technologies.35 Patent litigation and licensing demands for existing patents, by contrast, tend to occur after the defendant has already developed and implemented the technology. This is particularly true of NPE patent assertions and licensing demands, which some evidence suggests tend to happen in the last few years of a patent’s life, although the picture is complicated.36 NPE licensing demands and litigation against companies that are producing products do not seem to involve technology transfer or other indicia of new innovation. Indeed, evidence suggests NPEs may buy patents with vaguely-worded claims that are optimized for litigation but lacking in technical merit37 and that they may delay licensing of patents in order to increase revenue by targeting successful implementers after the fact.38 While some have argued that NPEs serve as efficient middlemen through this activity— transferring inventions from creators to commercializers—we found no such evidence in our 2015 study.39 We surveyed 191 in-house licensing attorneys at companies that produce products on the theory that these parties have direct knowledge of whether the company implemented new technology and because in-house counsel tend to negotiate licenses both as patent holders and as potential licensees.40 The survey examined the effects of licenses that a company took after receiving a patent demand, which was defined to include calls or letters suggesting areas of mutual interest or joint ventures, offering to license patents, threatening litigation, giving notice of intent to file an infringement lawsuit, or actually filing an infringement lawsuit.41 Respondents were asked whether those licenses led to any markers of innovation.42 Direct markers of innovation included the addition of new products or features.43 Indirect markers of innovation included whether the patent holder transferred know-how, other technical knowledge, or personnel (including through a consulting agreement) along

33

Id. Id. at 155 n.40. 35 See Colleen V. Chien, Opening the Patent System: Diffusionary Levers in Patent Law, 89 s. Cal. l. rev. 793 (2016). 36 Feldman, Ewing & Jeruss, supra note 12, at 8–9 (analyzing patent litigation data and finding that newer patents were asserted more frequently and that NPEs were more willing to assert patents of any age); Brian J. Love, An Empirical Study of Patent Litigation Timing: Could a Patent Term Reduction Decimate Trolls Without Harming Innovators?, 161 u. pa. l. rev. 1309, 1312 (2013) (“NPEs … assert […] their patents relatively late in the patent term and frequently continue to litigate their patents to expiration.”). 37 Josh Feng & Xavier Jaravel, Who Feeds the Trolls? Patent Trolls and the Patent Examination Process (2016) (unpublished manuscript), available at https://papers.ssrn.com/sol3/papers.cfm?abstract _id=2838017 [https://perma.cc/6656-SS7D] (last visited Oct. 17, 2019). 38 Erik Hovenkamp, How Reasonable Royalties Suppress Patent Licensing (2017) (unpublished manuscript), https://perma.cc/ABR2-3RVG (last visited Oct. 17, 2019). For thoughts on how to break the “vicious cycle of excessive, socially harmful remedies,” see William F. Lee & A. Douglas Melamed, Breaking the Vicious Cycle of Patent Damages, 101 Cornell l. rev. 385, 385 (2016). 39 See generally Feldman & Lemley, supra note 23. 40 Id. at 144–9 (describing methodology of 2015 study). 41 Id. at 149–55. 42 Id. at 155–66. 43 Id. at 160 fig. 9, 161 figs 10 & 11. 34

330 Research handbook on intellectual property and technology transfer with the patent, and whether any joint ventures were created.44 Again, the survey considered only licenses taken in response to unsolicited licensing requests.45 It did not look at the practice, particularly among university inventors, of entering into technology transfer agreements before embarking on development of a new technology.46 The responses suggest that licensing requests from NPEs rarely lead to direct or indirect markers of innovation. Ninety-two percent of respondents reported that when they licensed technology from NPEs, they added new products or features as a result of that licensing zero to ten percent of the time.47 The results were even stronger when respondents were asked about indirect markers of innovation, with respondents reporting with complete unanimity that they rarely received technical knowledge, transfer of personnel, or joint ventures along with a patent license.48 Thus, the results suggest that NPEs do not appear to be playing the role of efficient middlemen. While it is certainly possible that a middleman role could be reflected in markers other than the ones we examined, we did not find such evidence in our preliminary work. Interestingly, the evidence was also dismal when ex post licensing requests came from those other than traditional NPEs.49 When product producing companies and universities made unsolicited approaches and those approaches resulted in a licensing agreement, the agreements were unlikely to lead to direct or indirect markers of innovation.50 Three-quarters of respondents reported new products or features from zero to ten percent of the time, ninety-four percent reported transfers of personnel (including through consulting agreements) zero to ten percent of the time, and ninety-one percent reported joint ventures from zero to ten percent of the time.51 These observational results suggest that ex post patent licensing demands do not appear to lead to technology transfer or other markers of innovation, no matter what type of party initiates the unsolicited approach.52 A middleman who is not making a product and not actually providing the licensee with new technology is operating at most as a tax collector, taking money from innovative companies, perhaps for the benefit of inventors who could not otherwise do battle against large companies who have implemented their ideas.53 And perhaps specialists are better at collecting money than some types of inventors, particularly independent inventors.54 But transactions are not desirable for their own sake.55 It is socially desirable to impose such a tax on innovators only

44

Id. at 162 figs 12 & 13, 163 figs 14 & 15, 164 figs 16 & 17, 165 figs 18 & 19, 166 fig. 20. Id. at 156. 46 Id. 47 Id. Zero to ten percent was the lowest category offered. We suspect, though we cannot prove, that for almost all of respondents the number was in fact zero. 48 Id. at 157. 49 Id. at 160. 50 Id. 51 Id. at 160, 163–4. 52 This was a pilot survey. We are currently at work on a much larger survey of responses to patent licensing demands. 53 Id. at 142. 54 Stephen H. Haber & Seth H. Werfel, Patent Trolls as Financial Intermediaries? Experimental Evidence 1 (Sept. 21, 2016) (unpublished manuscript) (on file with authors). 55 Michael J. Burstein, Patent Markets: A Framework for Evaluation, 47 arIz. st. l.J. 507, 512–14 (2015) (noting that efficient markets that match buyers and sellers of a particular asset are not usually thought to be ends in themselves; instead they are instrumentally useful when they serve other social goals, such as allocating useful goods and services or mitigating risk). 45

Is patent enforcement efficient? 331 if the world gains something from it. That might be true if, for instance, the implementer had actually copied the idea from the patentee. If we think copying could reduce incentives to invent, we might reasonably prefer to force copiers into licensing arrangements instead, compensating the inventor whose work is copied. The available evidence suggests it is unlikely that most patent enforcement targets such copying. For example, Cotropia and Lemley demonstrate that most patent lawsuits are filed against those who have developed a product independently, rather than those who have taken the idea from a patent holder.56 And while some have speculated that defendants may copy indirectly, learning about the invention from the patentee’s product or from scientific discussions of the idea in conferences or academic journal articles without ever reading the patent itself,57 that is far more likely when the patentee actually makes a product than when it produces nothing other than the patent. Nor are individual inventors and for-profit firms likely to disseminate their ideas in other ways, such as by publishing academic papers later read by others who copy them. Universities, by contrast, are more likely to produce this sort of technology transfer. Further, there is evidence that NPEs tend to assert patents at the end of their lives, while practicing entities assert patents early, casting further doubt on the copying story.58 And there is very little evidence that patentees have used a remedy created in 1999 to protect against copying of published patent applications.59 Nor does the tax collection via patent licensing demands seem a particularly good way to fund future research by inventors. Although it is true that patent litigation can generate revenue that inventors might put back into research on new inventions, it is a singularly inefficient way of generating that revenue.60 Bessen, Ford and Meurer find that only a small fraction of damages awarded to NPEs actually gets returned to inventors.61 Most of it is lost to legal fees and to the intermediaries who make money asserting the patents.62 Those inventors may or may not invest what return they do receive in further research and development, and that

56

Cotropia & Lemley, supra note 14, at 1424. Robert P. Merges, A Few Kind Words for Absolute Infringement Liability in Patent Law, 31 berkeley teCh. l.J. 1, 29 (2016); Rantanen, supra note 25, at 7. 58 Love, supra note 36, at 1312 (“Product-producing companies predominantly enforce their patents soon after they issue … NPEs … assert[] their patents relatively late in the patent term and frequently continue to litigate their patents to expiration.”). We (and Love) acknowledge that the interpretation of his data is complicated by the change in the number of NPE suits during the time of his study. Feldman, Ewing & Jeruss, supra note 12, at 74 (“Thus, some of the increase in litigation activity by monetizers that Love observed during the final nine years of the patent term may relate to the general increase in patent litigation attributable to NPEs that has occurred during those years, a possibility that Love identifies in the article.”). 59 35 U.S.C. § 154(d) (2012) (providing provisional rights to reasonable royalty from copier of published patent application). 60 James Bessen, Jennifer Ford & Michael J. Meurer, The Private and Social Costs of Patent Trolls, 34 reg. 26, 32–3 (2011); David Olson, On NPEs, Holdups, and Underlying Faults in the Patent System, 99 Cornell l. rev. onlIne 140, 143 (2014) (describing payments to independent inventors in the absence of commercialization as “wasteful for society.”). 61 Id. (arguing that inventors receive possibly less than two percent of awards NPEs receive after their pursuit of infringement suits). 62 Fiona M. Scott Morton & Carl Shapiro, Strategic Patent Acquisitions, 79 antItrust L.J. 463, 481–3 (2014) (“[A] relatively small share of the costs imposed by [NPEs] on targets is returned to the original patentees. In other words, the transfer of funds from allegedly infringing downstream firms to patentees is done using a very ‘leaky bucket.’”) . 57

332 Research handbook on intellectual property and technology transfer further R&D may or may not generate new inventions.63 But because the overwhelming majority of defendants in NPE suits are themselves independent inventors, not copiers, the system is taxing one inventor to pay another, and losing most of the money in the process. It would seem far more efficient to fund inventor research directly through general tax revenue.64 And indeed, we do fund the class of NPEs most likely to engage in tech transfer—universities—in ways that mostly have nothing to do with patent litigation. Some have suggested a different theory of tax collection—that NPE patent suits provide an alternative way for venture capitalists to recover some of their investment in a failed invention. We are skeptical that venture capitalists are motivated by the prospect of this sort of consolation prize. They tell us that they are not—in survey responses they indicate that it is the prospect of a big win, not the possibility of recovering some money from a failed investment, that motivates them.65 But even if there were some marginal incentive related to additional investment from NPE taxation, that has to be weighed against the cost imposed on successful, product-implementing businesses, as well as on startups themselves.66 In particular, venture capitalists unanimously agree that if a startup company has a patent assertion against it, that would be a significant deterrent for any funding request.67 Finally, it is worth noting that a majority of defendants targeted in NPE suits are small companies such as venture-funded startups, not large companies.68 Thus, at best, many of these NPE suits would be taxing some venture-backed startups for the benefit of subsidizing others. That seems a dubious policy idea.

IV.

IS OWNERSHIP A GOOD IN AND OF ITSELF?

Responding to some of the concerns with traditional justifications for patent protection and licensing, a number of scholars have articulated what we call “commercialization-plus” justifications for patent protection. These justifications differ from the traditional innovation-based justifications because they focus on the alleged need for early or additional protections to

63

Alberto Galasso & Mark Schankerman, Patents Rights and Innovation by Small and Large Firms 2–5 (Nov. 20, 2015) (unpublished manuscript) (on file with authors). 64 See generally Daniel J. Hemel & Lisa Larrimore Ouellette, Beyond the Patents—Prizes Debate, 92 tex. l. rev. 303 (2013) (arguing that innovation can be encouraged by incentives other than patent system, including R&D-related tax incentives); Camilla A. Hrdy, Commercialization Awards, 2015 wIs. l. rev. 13, 52–64 (describing federal and state funded commercialization awards and their efficacy). 65 Robin Feldman, Patent Demands & Startup Companies: The View from the Venture Capital Community, 16 yale J.l. & teCh. 236, 243, 276–81 (2014) (reporting results of survey in which sixty-four percent of venture capitalists responded that when deciding whether to invest in a startup, they did not consider the potential for selling patents to NPEs if the company they invested in failed). Indeed, VCs report that they view the technology itself as less important than the management team. Paul Gompers et al, How Do Venture Capitalists Make Decisions?, NBER Working Paper 22587 (Sept. 2016). 66 Bessen, Ford & Meurer, supra note 60, at 33 (arguing that even if NPE litigation has small positive effect on inventors, it may still decrease overall innovation in society by imposing losses on defendant firms, discouraging innovation in other firms for fear of inadvertent infringement, and affecting research agendas of small companies toward areas where patents could be asserted against larger companies). 67 Feldman, supra note 65, at 280. 68 Colleen Chien, Startups and Patent Trolls, 17 stan. teCh. l. rev. 461, 464 (2014).

Is patent enforcement efficient? 333 encourage post-invention investment in commercialization.69 The most famous of these, known as “prospect theory,” recommends granting strong patents early in the life cycle of an idea so that a single party can control development of the idea much the way that a mineral prospector manages a mineral claim site.70 Other scholars have suggested that we should grant patents to old technologies in areas like pharmaceuticals in order to encourage the patent owner to engage in clinical trials and bring the product to market.71 Still others have suggested granting normal patents at the outset, and then, if no one commercializes the idea, granting extra rights to the person who does.72 The Bayh-Dole Act, passed in 1980, was premised on the worry that university inventions would languish unless one party was given the right to turn those inventions into commercial products.73 Commercialization-plus theories are controversial. One of us has criticized Kitch’s prospect theory as “fundamentally anti-market” because it presumes that central control is superior to market allocation of existing resources, and the other has argued that patents are entirely unlike the more clearly defined rights in Kitch’s mineral analogy.74 Others have suggested that commercialization theory is poorly fitted to industries in which invention proceeds by stages and improvements rather than by discrete advances.75 Kitch’s theory may justify patent protection in certain industries. Burk and Lemley suggest that the theory maps best to the pharmaceutical industry, where government regulatory barriers 69 Michael Abramowicz, The Danger of Underdeveloped Patent Prospects, 92 Cornell l. rev. 1065, 1073 (2007) (noting that patent underdevelopment occurs when patentee decides not to commercialize product because patent term is not long enough); Michael Abramowicz & John F. Duffy, Intellectual Property for Market Experimentation, 83 N.Y.U. L. rev. 337, 340 (2008) (arguing that just as patents encourage risky but ultimately beneficial technological experimentation, some additional form of intellectual property protection could result in a socially beneficial increase in market experimentation and entrepreneurial activity); F. Scott Kieff, Property Rights and Property Rules for Commercializing Inventions, 85 MInn. l. rev. 697, 703 (2001) (arguing that patent rights must be treated as property rights to facilitate investment in commercialization of inventions); Edmund W. Kitch, The Nature and Function of the Patent System, 20 J.l. & eCon. 265, 275–80 (1977) (advancing the so-called “prospect theory” which suggests granting strong patent rights early so that a single party can control development of an idea); Ted Sichelman, Commercializing Patents, 62 stan. l. rev. 341, 345 (2010) (arguing for implementing “commercialization patent” which would only be granted if patentee committed to commercializing invention). 70 Kitch, supra note 69, at 275–80. 71 Benjamin N. Roin, Unpatentable Drugs and the Standards of Patentability, 87 tex. l. rev. 503, 507 (2009). 72 Sichelman, supra note 69, at 345–6. 73 Council on Governmental Relations, The Bayh-Dole Act: A Guide to the Law and Implementing Regulations 2 (1999), available at http://otd.harvard.edu/upload/files/The_Bayh-Dole_Act-__A_Guide _to_the_Law_and_Implementing_Regulations.pdf [https://perma.cc/3P7D-P8UB] (last visited Oct. 17, 2019) (describing the shift from non-exclusive licensing before passage of the Act to the exclusive licensing potential created by the Act). 74 feldMan, supra note 2, 29–34 (introducing the bargain theory of patents—which suggests that a patent provides no more than an opportunity to bargain over the definition of rights—by criticizing Kitch’s analogy to mineral rights and arguing that a better analogy is to hunting license); Mark A. Lemley, The Economics of Improvement in Intellectual Property Law, 75 tex. l. rev. 989, 1044–72 (1997) (describing prospect theory and critiquing theory in detail); Mark A. Lemley, Ex Ante Versus Ex Post Justifications for Intellectual Property, 71 u. ChI. l. rev. 129, 139–40 (2004) (same); Lemley, supra note 13, at 739–40 (same). 75 Robert P. Merges & Richard R. Nelson, On the Complex Economics of Patent Scope, 90 ColuM. l. rev. 839, 908 (1990).

334 Research handbook on intellectual property and technology transfer significantly raise the cost of entry and may require exclusivity, though not necessarily exclusivity provided by the patent system.76 And it may justify patent protection for some kinds of NPEs, like universities, that are ill-suited to commercialize inventions on their own but wish to transfer the patent to someone who is in a better position to do so.77 Notably, though, any form of commercialization theory is self-limiting in certain important respects that have not previously been discussed in the literature. First, if exclusivity is necessary to induce a firm to commercialize a technology, we should rarely, if ever, see multiple companies independently develop the same technology. The very concept of commercialization theory is that no one would invest in developing and commercializing the technology unless they were first confident they would have exclusive rights over that technology.78 There may be exceptions in which companies engage in patent racing, each hoping to be the first to reach an important invention and therefore obtain those rights.79 But the historical examples of patent races have tended to be races to invent, not races to commercialize.80 If commercialization theory is correct, even independent inventors will not commercialize the technology unless and until they are confident they will have exclusive rights over that technology. And racing to commercialize (rather than to invent) is presumably something mostly engaged in by practicing entities, not NPEs. For the same reason, under commercialization theory we should not see companies rely on open source or public domain technologies.81 If we do, that

76

Burk & Lemley, supra note 15, at 143 (noting that development of products in pharmaceutical sector is characterized by long development times and high costs, due to constraints placed on development by significant regulatory oversight). Even there, though, the claim may be overstated. The pharmaceutical industry has several forms of regulatory exclusivity in addition to patents. See generally Robin Feldman, Regulatory Property: The New IP, 40 ColuM. J.l & arts 53 (2016) (describing history of thirteen non-patent exclusivities that can be obtained through Food & Drug Administration). And we see commercialization of biotechnology inventions like CRISPR even in the absence of patent protection. Robin Feldman, The CRISPR Revolution: What Editing Human DNA Reveals About the Patent System’s DNA (working paper 2017). 77 See supra note 74. But see Feldman & Lemley, supra note 23, at 160–6 (presenting data showing ex post as opposed to ex ante university licensing of patents does not lead to much future innovation). 78 See Kieff, supra note 69, at 703 (arguing in favor of commercialization theory by noting that “the power to restrict use that is conferred by a patentee’s property right and the strict enforcement of this right with a property rule … are paradoxically essential to avoiding underuse” of an invention). 79 Jean Tirole, The Theory of Industrial Organization 394–9 (1988) (providing introduction to phenomenon of patent racing); Yoram Barzel, Optimal Timing of Innovations, 50 rev. eCon. & stat. 348, 352 n.11 (1968) (describing option of granting or auctioning monopoly or patent rights on inventions before resources are committed to avoid duplicative use of resources); Partha Dasgupta & Joseph Stiglitz, Uncertainty, Industrial Structure, and the Speed of R&D, 11 bell J. eCon. 1, 27 (1980) (analyzing effect of competition on R&D and arguing that patents are needed to get firms to engage in R&D and that competition may also stimulate R&D in less efficient manner); Mark F. Grady & Jay I. Alexander, Patent Law and Rent Dissipation, 78 va. l. rev. 305, 306 (1992) (“[Patents] encourage hopeful inventors to squander valuable social resources in the race to win the patent.”); Jennifer F. Reinganum, “The Timing of Innovation: Research, Development, and Diffusion” in 1 Handbook of Industrial Organization 849, 853–68 (Richard Schmalensee & Robert D. Willig eds., 1989) (analyzing symmetric models of innovation in which several firms are seeking invention simultaneously). 80 Lemley, supra note 13, at 711–12. Winners of races to invent are more likely to do follow-on innovation. Neil C. Thompson & Jeffrey M. Kuhn, Does Winning a Patent Race Lead to More Follow-on Innovation? (working paper 2017). 81 See Clark D. Asay, Enabling Patentless Innovation, 74 Md. l. rev. 431, 434 (2015) (noting that open innovation has been viewed as “non-commercial” means of development).

Is patent enforcement efficient? 335 is reasonable evidence that exclusivity is not necessary to induce commercialization in that industry.82 Further, if any form of commercialization theory is correct, and certainly if enhanced commercialization theory is, infringement should be rare. Independent later inventors should not commercialize because they will not have exclusive rights to the invention, which, by hypothesis, are necessary to develop it. Nor should we see much copying of the patentee’s invention because if commercialization requires market exclusivity the copier will not generally be any better off than an independent inventor who does not have exclusivity. That does not mean we would never see patent litigation. But it does mean that if commercialization theory is correct, it should involve particular circumstances such as a lower regulatory burden on second entrants, which is true of generic pharmaceutical companies, or some reason to think that simply knowing that a market exists dramatically reduces the costs of commercialization, encouraging others to enter despite the lack of exclusivity that was theoretically necessary to spur such entry.83 There may be such cases; Samsung Electronics Co. v. Apple Inc. is arguably one.84 But those cases would generally involve copying existing market participants, not independent development. A third implication of commercialization-plus theories is that non-practicing patent owners should generally not grant nonexclusive licenses. If market exclusivity is required for commercialization, universities should be granting exclusive licenses to practicing entities in any given market, because the licensee needs that exclusivity to commercialize the invention. Nonexclusive licenses should be rare and tightly controlled because the nonexclusive licensees would have to coordinate their production and pricing decisions under commercialization theory. And nonexclusive licenses to multiple parties should be nonexistent because allowing open entry into a market is inimical to the theory of commercialization.85 In fact, however, the evidence suggests not only are most university licenses now nonexclusive,86 but that the lack of exclusivity is an important driver of subsequent improvement for core enabling technologies.87 Even if commercialization theory justifies patent protection in some industries, it cannot justify most modern patent litigation. Nor can it justify ex post licensing demands by NPEs. Outside the pharmaceutical industry, NPE licensing does not look much like commercializa-

82 Id. (“Numerous firms have found ways to successfully commercialize open innovation, even making it the heart of a firm’s commercial activities in some cases.”). 83 Abramowicz & Duffy, supra note 69, at 340, 383 n.129 (noting that “second-movers” can have distinct advantage in that they do not have to bear cost of investing in creation or development of market, and they can copy first mover’s successes). 84 Samsung Elecs. Co. v. Apple Inc., 137 S. Ct. 429 (2016). 85 Ian Ayers & Lisa Larrimore Ouellette, A Market Test for Bayh-Dole Patents, 102 Cornell l. rev. 271, 276 (2017) (“[I]f conventional wisdom is correct that Bayh-Dole patents are justified only by their commercialization incentive, then a nonexclusive license is prima facie evidence that the invention ought not to be patented at all.”). Ayres & Ouellette suggest a market test for university patents, requiring them to offer a free nonexclusive license, and allowing them to grant an exclusive license only if no one takes them up on that offer. Id. at 279–80. That should smoke out any university patent cases that do in fact require exclusivity for commercialization. Id. at 280. 86 See generally Jacob H. Rooksby, Innovation and Litigation: Tensions Between Universities and Patents and How to Fix Them, 15 yale J.l. & teCh. 312 (2013). 87 Mark A. Lemley, Patenting Nanotechnology, 58 stan. l. rev. 601, 627 (2005) (“Ideally, universities will realize that enabling technologies are more valuable not just to society but even to their owners when many firms compete to exploit and improve them.”).

336 Research handbook on intellectual property and technology transfer tion theory would predict. There is evidence that NPE patents are asserted later in life,88 and almost always against independent inventors.89 Ex post NPE patent licenses do not transfer the technology to a party that can later make use of it.90 And NPE patent licensing demands essentially always seek nonexclusive licenses from multiple parties rather than an exclusive license from a single party. Indeed, NPEs commonly sue twenty or more defendants in the same industry at the same time, settling with each of them in exchange for a nonexclusive license.91 The evidence also suggests that NPEs are targeting already-successful commercializers, not facilitating new commercialization. Feldman and Frondorf studied fifty product companies that had initial public offerings between 2007 to 2012.92 The authors found that forty percent of respondents received patent demands during the periods around the time of the IPOs, with those demands coming largely from NPEs.93 The effects were even more pronounced for information technology companies, with almost sixty percent of respondents reporting patent demands around the time of their IPOs.94 Similarly, Cohen, Gurun and Kominers found that companies are more likely to be sued by an NPE following a large, positive, cash shock such as a funding event or an IPO and that a cash shock was a significant predictor of the number of times a company was sued by NPEs.95 Cohen, Gurun and Kominers also found that no other form of litigation has the same type of cash targeting behavior—not torts, contracts, securities, environmental, or labor law.96 Nothing but patent law. The results suggest that NPE demand behavior may be driven by the lure of deep pockets and the leverage opportunities afforded by an IPO period, rather than the meritorious representation of claims that a wronged inventor could not bring on its own. These studies also provide a reminder that any benefits of NPE activity should, at a minimum, be evaluated against the costs to innovation and society.97 All forms of commercialization and product-based theories have a final, surprising implication for NPE suits. If the reason we need a patent is not to induce invention but to induce commercialization of that invention, the law should prefer those who actually commercialize the invention over those who merely invent it but do nothing further. Thus, the owners of those patents have failed in their purpose if they have neither commercialized the invention themselves nor exclusively licensed the patent to someone who does. Such an approach, therefore, may justify a working requirement, something that is generally considered anathema to patent

88

Love, supra note 36, at 1312. Cotropia & Lemley, supra note 14, at 1423–4. 90 Feldman & Lemley, supra note 23, at 155–60. 91 Mark A. Lemley & A. Douglas Melamed, Missing the Forest for the Trolls, 113 ColuM. l. rev. 2117, 2125–6 note 41, 2128 (2013) (noting that certain models of NPEs usually want to enforce their patents against multiple defendants and others seek to enforce patents for licensing fees); Allison, Lemley & Schwartz, supra note 9, at 236 (noting that there is some evidence that NPEs assert low quality patents for nuisance-value settlements). 92 Robin Feldman & Evan Frondorf, Patent Demands and Initial Public Offerings, 19 stan. teCh. l. rev. 52, 53 (2015); see also Chien, supra note 68, at 463–6; Feldman, supra note 65, at 237. 93 Feldman & Frondorf, supra note 92, at 77. 94 Id. at 84. 95 Lauren Cohen, Umit D. Gurun & Scott Duke Kominers, Patent Trolls: Evidence from Targeted Firms 17–19 (Nat’l Bureau of Econ. Res., Working Paper No. 20322, 2016). 96 Id. at 20–1. 97 See generally Bessen & Meurer, supra note 16; Morton & Shapiro, supra note 62, at 482–3. 89

Is patent enforcement efficient? 337 advocates.98 Further, as between the NPE inventor who does not engage in technology transfer and the independent inventor defendant, commercialization and product-based theories should prefer the defendant, since it is the defendant, not the patentee, who has achieved the goal of the patent system.

V.

POLICY IMPLICATIONS: INDEPENDENT INVENTION AND PRIOR USER RIGHTS

Most other IP regimes, including copyright and trade secrets, exempt independent development from legal liability.99 Patent law, by contrast, punishes anyone who practices the claimed invention, even independent inventors.100 A number of scholars have suggested that patent law should adopt some form of independent invention or prior user rights defense.101 Others have worried that an independent invention defense might interfere with patent races or incentives to commercialize.102 Our analysis suggests that the patent system might sensibly require that a patentee show either that it practices in the market or that it has engaged in technology transfer (direct or indirect) to those who then put the technology into practice.103 A patentee who cannot show either would still be able to enforce its patent, but only against those it could show copied the invention from it, directly or indirectly. This hybrid approach tracks the legal justifications that have been offered for patents. An inventor who develops an idea others copy would be able to enforce the patent against those copiers given that copying is a form of technology transfer (and one we view as socially inferior to a license agreement). An inventor who ends up disseminating technology to the world, either by practicing the invention or by transferring technology to others who do, would be able to enforce the patent against both copiers and independent infringers. And a practicing entity would similarly be able to enforce patents against both copiers and independent infringers on a commercialization theory. Introducing even such a limited independent invention defense would require courts to evaluate disputed claims of copying in some cases. Some have worried that much independent

98 Michael B. Abramowicz, The Problem of Patent Underdevelopment (GW Law Faculty Publication & Other Works, Working Paper No. 231, 2005), available at http://scholarship.law.gwu.edu/cgi/ viewcontent.cgi?article=1215&context=faculty_publications [https://perma.cc/GN3U-N2SP] (last visited Oct. 17, 2019); Sichelman, supra note 69, at 345 (noting that patent scholars have typically opposed new forms of patent rights such as commercialization patents because they impose losses and are costly, difficult to implement, and needlessly complex). 99 Cotropia & Lemley, supra note 14, at 1423. 100 Id. 101 E.g., Carl Shapiro, Prior User Rights, 96 aM. eCon. rev. 92, 95 (2006) (arguing in favor of prior use rights for nearly simultaneous independent inventors); Samson Vermont, Independent Invention as a Defense to Patent Infringement, 105 MICh. l. rev. 475, 479–80 (2006) (arguing in favor of defense for independent inventors to infringement). 102 Mark A. Lemley, Should Patent Infringement Require Proof of Copying?, 105 MICh. l. rev. 1525, 1527–32 (2007). 103 While in an ideal world a court might want to determine whether there was transfer of non-patent know-how along with the patent, any legal rule requiring such a transfer would lead to sham transactions in which NPEs attach useless “know-how” to nonexclusive patent license in order to obtain more favorable treatment.

338 Research handbook on intellectual property and technology transfer invention is really copying in disguise,104 though others are skeptical.105 As Sam Vermont has observed, though, courts are quite good at resolving factual disputes of this sort.106 Independent development will tend to leave a paper trail. And the parties will have an incentive to collect and present evidence on the question. While unscrupulous parties may try to manufacture evidence, that is true in any sort of case, and courts tend to be good at ferreting it out.107 Further, we think courts can properly include indirect copying from an idea once it has been publicized by the patentee within the concept of copying.108 We might worry in the opposite direction, that if we require proof either of commercialization or of technology transfer to avoid independent invention, NPEs will engage in “token use” (offering to make a few customer products) or token technology transfer, insisting on making a licensee take know-how whether they want it or not. That is indeed a potential problem. Courts will have to resolve what is real, good faith commercialization or technology transfer in borderline cases. But we think they will generally be able to distinguish bona fide commercialization from token use, just as they do in trademark priority cases. A requirement that patentees who do not engage in any form of commercialization or technology transfer prove that the defendant copied from them should be paired with stricter penalties against those deemed to have copied the invention. It would be reasonable to require, not merely permit, treble damage awards and attorneys’ fees against those found to have copied. Increasing the penalties for those who opt to take technology from a patentee without paying, while eliminating the penalty imposed on innovators who do not benefit from patentee technology transfer, properly aligns the patent system’s incentives with the evidence and the array of theoretical justifications for patents. Focusing patent enforcement on cases in which the patentee has actually contributed something to society would return patent law to its utilitarian roots as a promoter of innovation. It would also mean that a large percentage of current patent lawsuits, and most (though not all) suits filed by NPEs, would disappear. Because those suits impose a pure cost on society without any corresponding benefit, eliminating them offers the promise of making patent enforcement efficient.

104

See Merges, supra note 57, at 8. See Lemley, supra note 13, at 711 (“[S]urveys of hundreds of significant new technologies show that almost all of them are invented simultaneously or nearly simultaneously by two or more teams working independently of each other.”). 106 See generally Vermont, supra note 101. 107 See id. at 502–3 (arguing that risk of fraudulent claims of independent invention exists but is unconvincing, and that courts could place higher evidentiary burden on independent inventors to provide corroborated evidence of invention in such cases). 108 Merges, supra note 57, at 29. 105

16. Reviewing inter partes review five years in: the view from university technology transfer offices Cynthia Laury Dahl

I.

INTRODUCTION

When the America Invents Act (“AIA”) was passed in 2011,1 university tech transfer offices (“TTOs”) were braced for big changes and potentially some fairly dire consequences. Together with other intellectual property-reliant entities, TTOs were worried that certain provisions of the AIA would directly attack the value of their patents and therefore chip away at their ability to pursue a business model based on protecting invention and licensing. Among the most sobering of the provisions were a series of new post-grant patent review proceedings. These proceedings provided several new mechanisms for third parties to challenge already granted patents and potentially invalidate them. Designed to “weed out” weak patents that should not have been granted, these proceedings nonetheless had much broader applications. Because the proceedings were designed to be less expensive than litigation and to grant quicker yet final results, the concern was that they would become an attractive option for any third party, including potential infringers, to employ to challenge any patent, weak or strong. TTOs and other patent owners feared that the net result of these post-grant review proceedings would be to strip them of their assets, or at the very least cost them money in terms of lower patent value, uncertainty, and cost to defend. They were partially correct. Seven years after implementation, these post-grant challenge proceedings and specifically inter partes review (“IPR”) have in fact completely changed the landscape of patent litigation, as well as invalidation challenges for defendants to a patent suit.2 Not only are the numbers of IPRs voluminous, but the likelihood that a patent subject to an IPR is declared at least in part invalid is very high.3 However, although the literature seems to indicate that IPRs have had a universally chilling effect on patents, the experience of several TTO offices reveals a different story. As relayed in their own words, TTOs report that assets have 1 Leahy-Smith America Invents Act (“AIA”), Pub. L. No. 112-29, 125 Stat. 284 (2011) (codified in scattered sections of 35 U.S.C.). 2 Jake Berdine & Matt Rosenberg, Creating Leverage: A Practitioner’s Guide to Inter Partes Review and Its Effects on Intellectual Property License Negotiations, 44 AIPLA Q.J. 13 (Winter 2016) (“The AIA-implemented post-grant procedures fundamentally changed the landscape for defendants who are approached by patent owners seeking royalty payments for potentially invalid patents.”). 3 Jesse Schwartz, Supreme Court Takes Up Inter Partes Review: Is Relief In Sight?, teCh transfer enews blog (Aug. 30, 2017), available at https://techtransfercentral.com/2017/08/30/supreme-court -takes-up-inter-partes-review-is-relief-in-sight/ (last visited Oct. 17, 2019) (“About 74% of IPRs with final written decisions have resulted in all claims being invalidated—more than 1,200 of them so far— and about 14% have invalidated at least some claims. Only 12% have resulted in challenged patents being fully upheld.”).

339

340 Research handbook on intellectual property and technology transfer thus far been mostly immune from IPR attack. This is due to a variety of reasons connected to the mission of TTOs and the nature of the intellectual property (“IP”) they protect. Although the specter of post-grant challenge proceedings has in fact cost TTOs some resources through increased uncertainty and the need to adapt to the new law, the overall effect has been far milder than expected. TTOs explain that post-grant challenge proceedings in general and IPRs in particular have not had a significant impact on TTO licensing, nor on their traditional approach to patent filing, patent enforcement, and budgeting. This topic is ripe for analysis. It has been seven years since the IPR procedure was enacted, and with the benefit now of two Supreme Court cases interpreting IPR, plus anecdotal information from the universities themselves, we can gather a much more accurate picture of IPR’s actual effects on TTOs.4 In addition, barring large legislative changes, these effects are unlikely to shift over the next several years, although throughout this Chapter I will discuss some interesting and still evolving issues that may change the analysis slightly. Finally, although TTOs by and large have not changed their approach because of these post-grant challenge proceedings, the seven years have given some TTOs enough time to adopt a very few small shifts in policy that might be instructive for others to learn about or perhaps emulate. These small shifts are helping those TTOs to inoculate themselves from even the minor tremors caused by the possibility of post-grant challenge. The first section of this Chapter provides a brief context for the AIA discussion, including what the AIA was designed to address, some of the changes that were implemented under the AIA, a brief description of the challenge proceedings, and why IPRs are the most relevant challenge proceeding to consider when discussing TTOs. The section also describes the methodology behind a survey I conducted of eighteen university TTOs to gather instructive anecdotal evidence presented below. The second section contains the survey comments. It discusses why IPRs are theoretically worrisome for patent owners, but explains why IPRs have not significantly affected TTO policies and procedures about what patents they file, their licensing practices, how they enforce their patents, and their budgeting. The final section looks forward from this point, highlighting some topics about IPRs that remain open and discussing some court cases, legislative initiatives and executive branch responses that—depending on result—may potentially have an impact on IPRs, and therefore TTO policy.

II.

AIA BACKGROUND AND THE IMPORTANCE OF IPR

The AIA was the first major overhaul of the patent laws in fifty years. Before its passage, conventional wisdom was that the United States Patent and Trademark Office (“USPTO”) was allowing too many broadly-worded patents to issue and that enforcing such patents was

4

For background information on the subject, see Robert MacWright, Three Years after the America Invents Act: Practical Effects on University Tech Transfer, 52 (3) Les nouvelles–J. lICensIng exeCutIves soC’y, June 2017, available at https://ssrn.com/abstract=2961434 (last visited Oct. 17, 2019); John Morgana &Veronica Sandoval, Pacific Northwest Perspective: The Impact of the America Invents Act on Nonprofit Global Health Organizations, 9 wash J. l. teCh. & arts 177 (2013).

Reviewing inter partes review five years in 341 stifling innovation.5 Especially large organizations in the computer/telecommunications, financial services and information technology industries were clamoring for stricter scrutiny of patents.6 Out of this history and given this perspective, Congress included a number of updates to patent law under the AIA. Among other changes, the AIA expanded patent review proceedings, both pre- and postissuance. Of particular concern to TTOs are three new post-grant procedures available for third parties to challenge the validity of patents that have already issued: Post Grant Review (“PGR”)7; Covered Business Method Review (“CBM”);8 and IPR.9 These new processes were designed to “produc[e] better patents and rectif[y] improper issuances” by expediting challenges.10 However, patent holders believe the overall effect has been that third parties may more easily challenge patents, regardless of their merits, and the procedures can result in significant delay and expense for patent owners.11 Which of the above procedures is appropriate depends on the party using the challenge, the timing of challenge, the basis and type of challenge, and the subject matter of the patent. PGR can only be used for a limited time (within the first nine months after a patent issues) but any third party can use PGR to bring any evidence to bear that successfully proves invalidity on many bases, including patentable subject matter, novelty, nonobviousness, and double patenting. CBM is only applicable against business method patents and the challenger has to have been sued for infringement under the patent. However, for the lifetime of the patent (after nine months after issuance) a third party can use CBM to offer any material to invalidate the subject patent on most any basis, including patentable subject matter, novelty, nonobviousness, and double patenting. IPR is a much narrower tool in that third parties may use only other patents or written material to invalidate the subject patent under Section 102 or 103 (novelty and nonobviousness) of the Patent Act. However, similar to CBM, IPR can be used throughout the lifetime of the patent (after nine months after issuance), but the patent may cover any subject matter. The IPR proceeding can either run in parallel to an active litigation or can be instituted outside of litigation. The district court will often stay the court case waiting for the IPR result.12 Of the several new proceedings, the most relevant to TTOs is the IPR. The other two proceedings either involve subject matter that is not usually patented by TTOs (for example the “practice, administration, or management of a financial product or service,” relevant to the patents subject to CBM), or are only relevant at a stage where a TTO patent is unlikely to be challenged (PGRs may only be filed during the first nine months of an issued patent’s life.)13 5

Mark A. Lemley, The Unsurprising Resilience of the Patent System, 95 tex. l. rev. 9, 10 (2016). MacWright, supra note 4 at 72. 7 35 U.S.C. §§ 321–29. 8 35 U.S.C. § 321. 9 35 U.S.C. §§ 311–19. The legacy Ex Parte Re-exam procedure still remains in effect after the AIA, but among other differences, does not involve the participation of the third party and proceeds before the USPTO examiner rather than the Patent Trial and Appeal Board (“PTAB”). 10 Morgan & Sandoval, supra note 4 at 7. 11 Christopher Arnold, The AIA and TTOS: How Technology Transfer Offices Can Best Handle the Changes in Patent Law Brought about by the America Invents Act, 41 J.l. & eduC. 417 (2012). 12 A chart comparing the requirements and characteristics of each proceeding is at USPTO, available at https://www.uspto.gov/sites/default/files/ip/boards/.../aia_trial_comparison_chart.pptx (last visited Jan. 23, 2019). 13 TTOs usually file patents over very early stage technology, so their patents will likely not be commercialized and thus ripe for an IPR attack during the relevant period for a PGR. 6

342 Research handbook on intellectual property and technology transfer IPRs, on the other hand, are valid against all kinds of patents, including the life sciences and physical sciences patents that make up the bulk of the TTO docket. The IPR proceeding is also important to consider because it was the subject of two different Supreme Court rulings in the 2017–2018 term, one of which specifically affirmed it as constitutional.14 The comments that provide the information in this Chapter come from telephone interviews with TTO offices performed in June and August of 2018. The subjects interviewed represented eighteen TTOs,15 from large and small public and private universities, a few of which were part of a larger university system. I chose my list of interviewees to include active TTO offices, which were recommended by other interviewees as being leaders in the TTO community, that could provide a sampling of different-sized offices, different-sized universities, and both public and private institutions. Although I did not plan the survey to be exhaustive or completely inclusive, the fact that the answers were vastly consistent across the sample indicate to me that the answers to my questions are likely illustrative of the answers of many other TTOs. In each case the employees at the TTO designated the person that I should speak to who was most knowledgeable about IPRs. Although the titles of each of the representatives I interviewed varied (as did the hierarchical structure of the TTO), generally I spoke to either the head of licensing for the TTO or to the highest ranking legal officer for the TTO. In a few cases I spoke to more than one person within the TTO—for example, when my questions crossed between the expertise of two people, or when the highest ranking person with the relevant information was relatively new to the office. I started each interview with the same list of questions, eliciting information about the TTO’s overall sense of worry about IPRs, how many IPRs the TTO had experienced (and under which circumstances), and if the TTO had changed (i) filing; (ii) licensing; (iii) budgeting; or (iv) enforcement policies, procedures, or strategy because of the possibility of IPRs. I also asked if they had noticed positive or negative changes in licensing success that they could attribute to IPRs. In each case, I typed written notes during the interview, taking down answers word-for-word whenever possible to allow for direct quotations. Some of the TTOs requested that their comments remain anonymous or that the statistics be reported in aggregate form. In most cases, this desire was because the TTOs had active IPR proceedings, or because

14 In Oil States Energy Services v. Greene’s Energy Group, 138 S. Ct. 1365 (2018), the Supreme Court considered whether it was a violation of Article III or the seventh amendment right to a jury trial for a patent owner to lose patent rights by having its invalidity determination adjudicated in an administrative PTAB proceeding rather than an Article III court, as would happen in a patent infringement lawsuit. Ultimately the Supreme Court found the IPR proceeding constitutional because the patent rights were “public rights” derived from common law and granted by an administrative body (the USPTO), which could maintain control over the decision of whether that grant was correct. However, the Supreme Court specifically stated that it was not ruling on the possibility that IPRs might be considered unconstitutional under the Due Process clause, the Takings clause, or that IPRs might not apply retroactively to patents that issued before IPRs were instituted. The other court case to consider IPRs was SAS Institute v. Iancu, 138 S. Ct. 1348 (2018), where the Supreme Court ruled that upon instituting an IPR, the USPTO was required to consider all claims that were challenged in the IPR, not just some of the claims. 15 I interviewed the TTO at Arizona State University; University of California, Berkeley; University of California, Irvine; University of California, Los Angeles; Columbia University; Duke University; University of Florida; Harvard University; Johns Hopkins University; Massachusetts Institute of Technology; University of Michigan; University of Minnesota; University of Pennsylvania; Stanford University; University of Texas at Arlington; University of Wisconsin-Madison; Washington University; and Yale University.

Reviewing inter partes review five years in 343 they were concerned about potential licensors or potential patent challengers learning about their attitudes toward enforcement or other policies. As a result, while the statements below are direct quotations from the interviews, I do not attribute the comments to the specific TTO that made them.

III.

HOW ARE IPRS AFFECTING TTO POLICIES AND PROCEDURES?

A.

IPRs Present More Risk to TTOs in Theory Than in Reality

Across the board, the TTOs I interviewed are concerned about the effects IPRs are having on their practice. They worry that IPRs leave their assets vulnerable to attack from better funded commercial players. They fear that IPRs will tilt the license negotiation in the licensee’s favor, because the licensee can threaten to file an IPR if they cannot get good license terms. And they claim that IPRs are contributing to a broad devaluing of patents across the board, because of the uncertainty they inject into the system where a patent can be declared invalid at any point in its life. They mention that the effect of IPRs is not what was intended by Congress; that instead of empowering parties to winnow weaker patents, IPR is being used as a litigation tactic to drive settlement and avoid having to pay for infringement. One summed up the situation by stating that IPRs “have added another burden to commercialization,” and another stated that IPRs are “part of a net negative … it really benefits everyone to have certainty on who owns patents and whether they are strong. Confusion makes it harder. All of these strategies and issues make it harder to get product out and to benefit the university.” However, curiously, the TTOs’ worry has not translated into significant policy or procedural changes. Overall, the survey comments showed that the expectation of deleterious IPR effects on TTOs was much worse than the reality. In response to even perceived risk, TTOs have implemented only small changes, if any at all. In addition, for a variety of reasons IPRs have not affected TTOs’ licensing success. This section will first address why TTOs might be worried about IPRs before postulating as to why the reality of IPRs does not match the expectation in the TTO setting, and therefore why TTOs can feel justified that they have made very few changes in policy. It makes sense to be intimidated by the prospect of IPRs for a variety of reasons. First, the exposure is comprehensive. No patent is immune from IPR challenge, so long as it has been issued for at least nine months. Any third party may file an IPR, regardless of whether they are an accused infringer and regardless of whether the patent owner has filed suit. Second, IPR procedures give distinct advantages to a patent challenger over a patent owner. It is substantially cheaper to file an IPR than for the patent owner to defend it.16 Moreover, the challenger can choose specific claims to attempt to invalidate, and estoppel only applies to claims for which the Patent Trial and Appeal Board (“PTAB”) issues a written opinion.17 This

16

Neal Solomon, The Problem of Inter-Partes Review, Ip watChdog, Aug. 8, 2017 (“In addition to the procedural issues, IPRs introduced an asymmetric component which particularly burdens the patent holder by requiring a very expensive ten-fold higher cost to defend the patent in the PTAB relative to the alleged infringer(s) cost of initiating an IPR.”). 17 35 U.S.C. §315(e)(1).

344 Research handbook on intellectual property and technology transfer means that the challenger may have several chances to challenge the patent (albeit different claims), and may in fact challenge the patent on other grounds, should the patent owner later file a patent infringement suit.18 Third, when an IPR challenge is used within the context of a litigation, the shift in the balance of power toward the patent challenger will affect the whole litigation strategy. If an IPR is timely filed, the judge in the patent infringement litigation will often stay the court case pending resolution of the IPR. Although the IPR will be resolved in no more than 12 months, this stay delays the adjudication of the infringement case and can also distract counsel and add greatly to legal fees. Although the effect of the USPTO’s rulemaking changing the review standard of patent claims in the IPR context remains to be seen, IPRs have almost always invalidated at least one claim of a challenged patent, and they are commonly successful in invalidating the entire patent.19 Because of the advantages granted to the challenger under an IPR, TTOs might think twice about enforcement, since any cease and desist letter might result in an IPR being filed. Although an IPR is always possible, it is ever more likely in the context of a patent infringement litigation. The possibility of an IPR can be quite intimidating. However, the reality for TTOs has been different. The reality is that the number of IPRs being filed against TTO patents is remarkably small. In the seven years since the IPR procedure was implemented, even as other patent owners have experienced an explosion in IPR proceedings,20 half of the TTOs interviewed (n = 9) had not yet had their patents subject to even one IPR.21 This is especially remarkable considering not only the vast numbers of patents filed by these TTOs, but also the fact that several of the TTOs reported that they were involved in an active litigation, where an IPR challenge would be more likely. Of the half of the TTOs familiar with IPRs, many had experienced only three or fewer proceedings. And for the few TTOs that had experienced more than three IPRs, many of those IPRs were clustered into either one or two large litigations. No TTO I interviewed reported that they had had more than four active cases that involved IPRs. The numbers may be particularly low because the chance of IPR success over TTO patents may be less than normal. TTOs must file early-stage patents both because they are covering the

18

The SAS Institute case (supra note 14) and recent USPTO rule making (effective for IPRs filed after Nov. 13, 2018) have leveled the playing field between patent holders and challengers substantially. The SAS case mandates that the PTAB review all claims challenged in the IPR, and the USPTO rule making now aligns the standard of review for claim interpretation and burden of proof used in IPRs with the standard used in the federal courts. This erases a former advantage for patent challengers, who were able to enjoy a broader claim interpretation and lower standard of proof for invalidity in PTAB procedures, and a narrower claim interpretation and higher standard of proof when fighting infringement in federal court. 19 Schwartz, supra note 3. 20 Mark A. Lemley, supra note 5, at 5–6 (citing Brian Love & Shawn Ambwani, Inter Partes Review: An Early Look at the Numbers, 81 u. ChI. l. rev. dIalogue 93, 94–7 (2014)); see Morgan Lewis, 2018 PTAB Digest: The Latest Trends and Developments in Post-Grant Proceedings, lexology, June 3, 2018, at 10, available at https://www.lexology.com/library/detail.aspx?g=0da05dd9 -8d41-450a-91af-4cae770d2ceb&utm_source=lexology+daily+newsfeed&utm_medium=html+email+ -+body+ -+general+section & ut m _campaign = aipla+2013+subscriber+daily+feed & utm _content = lexology+daily+newsfeed+2018-07-05&utm_term (last visited Oct. 17, 2019). 21 It is worth noting that in all the IPRs we discussed, the TTOs and their universities were on the defensive; no single TTO had launched an IPR offensively. This fits with the university mission as licensor of IP instead of commercial venture.

Reviewing inter partes review five years in 345 raw research subject matter of university labs and because TTOs often have to file patents early to beat an inventor’s disclosure deadline. In order to preserve patent rights, the university may need to file patents earlier than industry counterparts, sometimes before full data gathering is complete. These early-stage patents may have enablement risk, but they are so early stage that there is also less likely to be prior art that could invalidate them. Since IPRs invalidate through prior art, the early stage of the patented matter may be an advantage. An IPR challenge to a TTO patent is also less likely because of the TTO mission. Because TTOs focus much more on protection and licensing than on enforcement of their patents, many improbable conditions need to align for an IPR to proceed against a university patent. Patent challengers generally file IPRs in one of two scenarios: (i) the patent owner has accused the challenger of infringement and the IPR is a defense; or (ii) the patent presents a significant freedom to operate risk to the challenger’s commercialization, and licensing is not an option. The TTOs I interviewed enforce their patents very rarely; many of them consider enforcement outside the core mission of a TTO. As a result, the likelihood of a TTO filing an infringement lawsuit and sparking an IPR is low. Likewise, for a TTO patent to meet the second condition through creating a large freedom to operate risk, the patent would either have to cover valuable seminal technology, or the technology or science would have to be developed enough that a challenger would know they could not design around it when commercializing. In addition, the challenger would have to have been either unable or unwilling to license from the TTO, and yet must be so convinced of infringement risk that they are willing to invest thousands of dollars into attacking the patent prophylactically, thereby tipping their hand to the TTO to the probable infringement risk in the process. Given the early stage of most patented university technologies, these prerequisites all occurring together is extremely unlikely. And finally, even should a TTO face an IPR, they have two possible built-in defenses that keep the risk low. First, the TTOs of public universities may be able to completely shelter their patents from attack through a sovereign immunity defense. And second, most TTOs—from both public and private universities—face little or no risk of funding the IPR defense in all but the most unusual of conditions. Because the 11th Amendment protects states from federal suit, if IPRs are considered suits and state-run universities are considered arms of the state, public universities can invoke the defense of sovereign immunity to avoid IPRs completely. Two recent cases before the PTAB, namely Covidien LP v. Univ. of Fl. Research Foundation Inc.22 and NeoChord, Inc. v. Univ. of Md., Baltimore,23 each established that sovereign immunity properly shielded the university patents from IPR, even though (as in NeoChord) the motion to dismiss the IPR came after the IPR had been instituted,24 and importantly after the state university had participated in the IPR. There are still some questions about applying the sovereign immunity defense, among them whether participation in a full patent infringement lawsuit waives sovereign immunity25 22

Case No. IPR2016-01274, Paper 21 (PTAB, Jan. 25, 2017). Case No. IPR2016-00208, Paper 28 (PTAB, May 23, 2017). 24 Stephen Gardner, Nikia Gray, & Bryce Loken, Patent Trial and Appeal Board Dismisses Second Inter Partes Review of University Patent on Sovereign Immunity Grounds, Quarles and brady llp, May 25, 2017, available at https://www.quarles.com/nikia-l-gray/publications-and-presentations/ patent-trial-and-appeal-board-dismisses-second-inter-partes-review-of-university-patent-on-sovereign -immunity-grounds/ (last visited Oct. 17, 2019). 25 This question is at the heart of Regents of the University of Minnesota v. LSI Corp., case number 18-1559, on appeal to the United States Court of Appeal for the Federal Circuit, with arguments heard on March 11, 2019. 23

346 Research handbook on intellectual property and technology transfer and whether tribal immunity (which has been ruled not to be a defense to IPR) and sovereign immunity are analogous for these purposes or not.26 But for now, sovereign immunity is a valid defense that state universities can use to defend their patent assets from IPRs. Regardless of the outcome of the state sovereign immunity issue, universities still will fund IPR defenses in only the most unusual of circumstances. In most cases, a challenger will only file an IPR over patents with proven economic value, which usually means after the patent’s technology or science has been commercialized. Since the university does not commercialize patents itself, the patents ripe for IPR attack are almost always covered under license. The university license terms almost always mandate that the licensee cover the cost of enforcement, so even should the worst case IPR scenario happen, the university does not pay for defending its patent. So the cost of defense would only be an issue for the university under special circumstances—for example, if the patent at issue is unlicensed, or licensed nonexclusively such that no, one, licensee must bear the defense cost, or if the licensee could not afford the enforcement and the university wanted to step in to fund it. For all these reasons, the reality of IPRs might have been expected to not match the hype. But I was curious to know if just the possibility of more IPR proceedings being filed has nonetheless changed how TTOs do business. B.

TTOs Reveal the Effect of IPRs on Filing, Licensing, Enforcement, and Budgeting

After getting a general sense of the level of concern that TTOs had surrounding IPRs, and about how many IPRs they had experienced, I asked a series of questions to illuminate whether and how the possibility of IPRs had changed how TTOs were doing business, including whether they had changed their policies and procedures. I asked them to respond in four categories: whether IPRs had changed anything about their (i) patent filing strategies; (ii) licensing practices or results; (iii) willingness to enforce their patents; and (iv) budgeting. The answers to these questions are relayed below. 1. “Have IPRs caused you to change your patent filing strategies?” Given the possibility of IPRs, TTOs might be tempted to refocus their patent filing, limiting filing to areas where IPRs are less common.27 They might also perform more exhaustive prior art searches during prosecution to potentially inoculate their patents against IPR challenge. However, the survey responses showed neither to be the case. TTOs were deeply opposed to the first idea, and they dismissed the second idea out of hand. However, some TTOs did suggest they are making subtle changes to their filing practices to attempt to ready their patents for IPR challenge. Every TTO interviewed felt strongly that it was a critical part of their mission to protect all the patentable university inventions that might have a chance of being licensed and commercialized. They were therefore very opposed to making decisions on what to file or not file

26

The case deciding against tribal immunity defense was Mylan Pharmaceuticals Inc. et al. v. Saint Regis Mohawk Tribe IPR2016-01127, -01128, -01129, -01130, -01131, -01132. 27 Early reports stated that most patents being challenged (71.8%) were in electronics and the computer arts. Berdine & Rosenberg, supra note 2, at 17 (citing Christopher Douglas, et al., Inter Partes Review—One Year Later, alston & bIrd llp (Sept. 17, 2013)).

Reviewing inter partes review five years in 347 based on subject matter of the patent rather than on other factors. They were even somewhat offended at the suggestion: We are fortunate because we are healthy financially, and we can act as the service organization we are and attend to faculty needs, and we can patent when the idea is protectable, as opposed to worrying about if it might be subject to an IPR. We have money and resources and have always approached filings as more of a commitment to the technology transfer mission and less about tactics or profit margin. For us, when I picture 8 years ago, 2 years ago and now—IPRs have not changed filing or licensing decisions.

TTOs hedge their bets best when they file at least provisional patents on every invention that has enough data to support an application. Because TTOs never know where the next great commercial success will come from and do not have enough information at the start of the process to understand whether a patent will become valuable, they say they would never pick and choose among fields merely on account of an increased risk of IPR. Rather than avoiding certain subject matter to attempt to avoid IPRs, TTOs instead said they preferred to focus on ensuring that individual patent filings were as strong as they could be. For example, they file patents with as much supporting data as possible given publication constraints and first to file concerns.28 They avoid filing cover sheet provisionals whenever possible, and they make the claims of each provisional application as robust as possible. But they will also admit that these policies are not new. As one TTO explained, filing more complete patents is not a change brought about by the possibility of IPR, or even by the first to file provisions of the AIA, but is what TTOs have been trying to do for a while to protect their patents from attacks based on invalidity. The IPR and other AIA changes just make filing stronger patents even more of a priority: I could see how some TTOs might use IPRs as a factor when looking at whether to protect computer inventions, etc. because those patents might be more susceptible to attack. Maybe you file less aggressively in that area. But it’s rare that we make decisions not to file at all because of IPRs, both because we always file everything, and also because we have always handled all invalidation issues by just making sure the prosecution is as full as possible. IPRs haven’t changed how we feel.

As to doing more prior art searching, generally, TTOs do not believe that more extensive prior art searches will inoculate their patents from IPR challenge. First, since the patents they file are over such early-stage technology, the likelihood of there being any other prior art knocking out the patent on newness grounds is already low. And should the prior art exist, the fields of invention are so insular that the inventors would likely already know about any other potential prior art. And although it is not legally relevant, some TTOs depend on professional journals’ high standards for publication to give them insight into nonobviousness considerations. Since the TTOs are also generally strapped for resources, given the probable low risk of prior art citations, and given that some TTOs are worried about the results of a prior art search or a freedom-to-operate opinion (“FTO”) being discoverable, TTOs are not shifting to expend more money on prior art searches. Unless IPRs become a more pervasive problem for TTOs, this is unlikely to change.

28 Although this may not be as helpful in an IPR setting, because IPRs are based on Section 102 and 103 challenges, which are countered best by early filing.

348 Research handbook on intellectual property and technology transfer Further, the TTOs argue that not only would it be near impossible to judge which patents would be more vulnerable to IPR attack based on subject matter, but to make judgments about what to patent based on that analysis would be extremely unwise. They state that the better goal is to focus on filing inventions that will lead to strong patents, where there is (as best they can tell) a market to be licensed and commercialized: Standard operating procedure on how we evaluate IP is based on whether the invention as disclosed has good potential market value. It ends up being a much bigger question than patentability. Our budget is small—we can only afford to protect what has good prognosis for market. We look at what the tech is, who the players are in the field, what the product would be under the invention, whether it is something based on market research that people are likely to buy. A lot of black magic really—you do your best to figure out what the market would be. If it looks like a problem recognized in the field and that people would buy it, then we file a provisional. Then we market to companies we identify and others, or if it’s good tech for a startup, we license there. If there’s no interest at all, we might recommend not to pursue the non-provisional. Or sometimes we might think we have early technology, and the market is not ready; if we really think it could be important in market, we might file [a non-provisional application] anyway even without interest.

TTOs also make the valid point that they would never want to entirely avoid the risk of an IPR challenge by shying away from filing a patent that they think might be commercially successful. The patents that might attract an IPR are exactly the patents that TTOs should want to prioritize filing. Since there is a positive correlation between risk of IPR and commercial success, TTOs may favor filing the patents that might end up being most IPR-worthy: Frankly, anything that makes it to the commercialization stage, you are going to get IPR’d. We will deal with it when we get there—there WILL be an IPR. The likelihood of IPR is going to be based on whether [the patent] has commercial potential, not on the strength of the patent. Regardless of what’s been filed or not filed, they will attack. [IPRs are] like inventorship issues. When tens of millions of dollars are at risk, this [IPR] is just another tool.

Rather than the risk of IPR, when choosing which patents to file, TTOs are influenced much more by their concern over whether the patents will issue, given confusion around patentable subject matter and changes under the AIA that broaden what constitutes prior art and disclosure. TTOs describe IPR as being so sufficiently “downstream” that changing filing strategy based on a worry about IPRs seems a luxury when the patent will conceivably face so many challenges before it will even issue: We base our filing decisions on prior art, the path to patent, but because we have limited budgets, we can’t do a detailed patentability analysis. An IPR is so downstream, it doesn’t impact the upstream decision on what to patent. For the most part, we don’t really think about IPRs that much. We think about allowance and validity, but not specifically IPRs … Definitely feels like we are doing more provisionals, just to get that priority date, and preserve the international rights … It is not IPRs that are pushing us—we don’t think of it as that—tend to think about it as good prosecution. [IPRs are] just one more thing that we consider when we enforce and when we have our patents granted.

In sum, the risk of an IPR does not guide which patents TTOs choose to file. In fact, TTOs might be happy to have a patent in the portfolio that someday is deemed IPR-worthy, because it would be a strong patent with commercial value that is worth enforcing and challenging.

Reviewing inter partes review five years in 349 The one change in some TTOs’ filing strategy is not about what patents to file, but tactically how to maneuver during prosecution. Borrowing from corporate practice, some TTOs are now filing multiple continuations for some of their most valuable patents, or splitting claims across several patents to keep a set of patent claims alive. These claim sets are then able to be revised as necessary if challenged through IPR: Universities file very early in the product life cycle and the patent process; it’s really hard at that early point to position yourself against IPRs, so relying on continuations helps somewhat. If there were no IPRs, we would have a much thicker claim set. Continuations allow more flexibility to address the IPR; you can pivot to other claims to avoid the argument, if you keep the [patent] family alive … With IPR, we now take a better look at continuation practice. We used to file continuations much more sparingly.

This is of course not an option for every TTO, and certainly not an option for every patent family or even most families within a TTO’s portfolio. But continuation practice might be one helpful strategy to guard against challenges to patent validity, including but not limited to IPRs. This gives a patent owner a chance to fight back against what many consider a system that is slanted toward the benefit of the challenger. As one TTO stated: “Others are also doing the same thing in industry—we are taking their lead [in] that there are things you can do to protect yourself.” 2. “Have IPRs changed your licensing strategies?” The threat of an IPR might undermine the value of patents, which might carry over to patent licensing negotiations. On the other hand, depending on whether it is upheld, the state sovereign immunity defense to IPRs might shift the balance of power back in the other direction, since at least public universities might be able to offer a shield to the patents they license. If public universities can wield such a defense, that might in turn incentivize would-be licensees to license from public universities rather than their private university counterparts. All of these changes brought about from IPRs might conceivably force TTOs to alter their licensing strategies. TTOs do report that licensees’ attitudes toward patent value have in fact shifted. As a result, some TTOs are tweaking their licensing strategies around the margins to compensate. However, most of the TTOs attribute the shift in attitude more to their potential licensees’ business models than to IPRs, especially as concerns patents in the physical sciences. And while a state sovereign immunity defense seems helpful in theory, the reality is that it has not yet conferred a licensing advantage on public universities, or even tended to affect licensing discussions much at all. i.

Some licensees’ assessment of patent value has shifted, although probably not as a result of IPRs To start, most of the TTOs believe that attitudes about patent value have shifted. Some of the shift is blamed on IPRs, although most of the evidence for that connection appears to be anecdotal and second hand: We heard from other … campuses that prospective licensees can threaten to launch an IPR if they can’t get good terms as a negotiation tactic—this is awful! Whenever you start off a negotiation and

350 Research handbook on intellectual property and technology transfer someone is approaching it in such an aggressive way, it probably won’t get better as you go on! But we haven’t dealt with IPRs too much. There’s a lot of talk too about how high-paying licensees already paying royalties might file IPRs to get out of their license, but we have not found that to be the case.

But most of the blame should not be placed on IPRs. TTOs surmise there is a connection between how much value a licensee confers on the patent and why the licensee is seeking to license the patent in the first place: IPRs have impacted our ability to do business with established well-resourced businesses—they now increasingly look at IP through a slightly different lens. They look at assets as being helpful if they create an exclusive niche in the market, but if they are licensing just to prevent infringement, they are much less interested. It is even worse now, because they have a cheaper route to push back with IPRs. It has changed their perspective on the need to secure the patent, especially in the context of a freedom to operate.

So licensees that are hoping to sell a product based on the patents value those licensed patents more. Although IPRs are making a situation worse, perhaps only this certain subset of licensees—those that intend to commercialize the patent they are licensing—are really hoping for the patents they license to hold long-term value. Licensees that license in order to clear a freedom to operate hurdle do not mind if a third party challenges the patent they have licensed. Should the patent be proven invalid, provided the license allows termination upon such a finding, they can stop paying license fees.29 Since a sizeable percentage of the pool of licensees falls into the “freedom to operate” category, this could contribute to why licensees tend not to consider IPR risk when deciding whether to license and for how much. In addition, TTOs believe that licensing has become more difficult because in certain industries patents hold less value than in others. While a patent can serve as the entire basis for a life science company’s value, in the physical sciences, one patent is far less valuable. As a result, TTOs report that companies in the physical sciences industries tend to want to wait to be sued before they license a patent. Given that physical sciences companies create products that each could contain dozens if not hundreds or thousands of patents, infringement of any one patent (particularly a university patent which could be very early stage) is hard to prove, and since licenses to those patents tend to be nonexclusive anyway, a physical sciences company may make the decision to infringe rather than license: It is more common to have [licensing] problems in the physical sciences than in life sciences. This is because most deals in life sciences are exclusive licenses, which carves out a niche. In physical science, it is much more about cross licensing, not an exclusive license … It is not universal in the physical sciences, but especially in electronics, we are usually offering licenses to incremental tech in areas where they already have a foothold. They are just not that interested in licensing. A lot of large companies now seem to have a blanket policy not to license. [IPRs are] possibly part of it, but more probably it’s an aggregate lessening of value for patents. It is near impossible for us to get an injunction. It is hard for us to get large damages. Physical science companies are also getting savvy about whether patents matter to their industries. They have many more patents in any one

29

This assumes that the challenger is a third party. Many TTOs have provisions in their licenses that either terminate the license should the licensee challenge the patent, regardless of outcome, or keep the license active but automatically double or treble the license fees.

Reviewing inter partes review five years in 351 invention, and the patents are over incremental changes. The perception might be different maybe for life sciences, but physical science companies are taking the risk—“come and get me.”

All of these reasons add up to why many licensees of university patents may not ascribe as much value to the patents as they should. This has very little to do with the possibility of IPRs. ii.

Some TTOs have compensated for the reluctance to license through portfolio licensing and careful pricing Given that many TTOs are not willing or able to enforce their patents against infringers (for a variety of reasons discussed more in the next section below), this new reality of “efficient infringement” challenges TTOs to develop new licensing strategies. One inventive response— especially in the realm of the physical sciences—is to group patents together into portfolios to facilitate simultaneous patent licensing (especially in conjunction with enforcement after a patent has been infringed). This model acknowledges the reality that it is possible to build a life sciences company around licensing one particular patent, but that for a physical sciences company, licensing a portfolio of patents might be more enticing or even necessary: The licensing models for the two departments (medical school and engineering school) are different. For example, a small molecule patent—you can build a company around it—it is easier to commercialize. But engineering improvements are smaller; you wouldn’t build a company around them. We can get more revenue from a medical school patent because it is easier to build a company, easier to commercialize around it. It is more challenging for an engineering company. To deal with this, our new model is pooling engineering patents into a group and licensing as a group. We create a patent pool and license in a bundle. It would be great if we [life sciences licensing officers] could get a portfolio, but in [another licensing officer’s] space, it is much more of a concern. He works in magnetic resonance and wireless communications, and he is developing portfolios that he doesn’t just proactively license, it’s [also] … a litigation play.

And in addition to licensing portfolios of patents, TTOs are also carefully considering the point of view of more reluctant licensees when they price their royalties. Especially in the physical sciences where the perceived risk of enforcement is low, and especially when the licensee is interested in licensing in order to get freedom to operate, TTOs are looking to hit a “sweet spot” that is low enough that the licensee is still willing to pay to close off all risk of infringement: We do proceed with greater sensitivity in that we have modified what we would ask for in license fees. If the fees are too high, particularly in an FTO [freedom to operate] license situation, that raises the risk of IPR. We try to hit that sweet spot—especially with FTO—where we want them to pay just enough. Too much, and they may challenge the patent.

In this way, TTOs are trying to meet the licensee where they are, changing their policies somewhat to accommodate. But this has much less to do with IPRs than with the needs and expectations of the particular licensees.

352 Research handbook on intellectual property and technology transfer iii.

The effect of a sovereign immunity defense to an IPR is still uncertain, but so far it seems to be only minimally relevant as a negotiation point Is the prospect of sovereign immunity as a defense to IPR a valuable negotiation point for universities? When the Covidien case30 was first decided, TTOs of some public universities assumed not only that their patents could be shielded from this attack on validity, but that the shield could prove a valuable negotiation point with future licensees. Assuming the sovereign immunity defense could be imputed over to a licensee,31 the licensed patents could be completely inoculated from at least one form of attack. The public universities were not the only parties discussing this possibility; various commentators were mentioning the issue,32 and there was also a lot of industry buzz: Something that has gotten a lot of attention—not so much made an impact, but gotten a lot of attention—is around the sovereign immunity cases, where public universities are immune from IPRs unless they sue first … Some journalists were calling, law firms were offering congratulations, and the general counsel told us we might expect some attention.

Private universities also took notice. If IPRs were going to have a big effect on patent values and therefore on licensing success, a defense of sovereign immunity to IPRs could prove to be very important. The fact that only some TTOs could take advantage of the defense was a bit troubling, since it might also have created an imbalance in the community. But overall, although TTOs noted that any possible defense against IPRs would be helpful, they did not expect that sovereign immunity was going to be a driver for more successful licensing. For one thing, an IPR is only one method to invalidate a patent, and other methods are not subject to the defense. Even without the IPR process, there are still ways to invalidate a patent; [sovereign immunity] doesn’t foreclose the opportunity to invalidate, it just changes litigation strategy. IPR is still a strong tool, so [the defense] is still worth something, but not so much … In a nutshell, being able to foreclose an IPR option is helpful (valuable) but not dispositive.

In addition, the case law around sovereign immunity and IPR is still developing. There are many untested aspects of the sovereign immunity argument that may undermine its power as an IPR defense.33 For example, although sovereign immunity clearly applies to state universities as patent owners, it might be a personal defense.34 It is less sure that sovereign immunity can be commuted over to protect the licensee of a patent owned by the state university. It 30

Covidien LP v. University of Florida Research Foundation Inc., Case No. IPR2016-01274, Paper 21 (PTAB, Jan. 25, 2017). 31 This is an undecided question. As held in Reactive Surfaces Ltd, LLP v. Toyota Motor Corp., No. 1:14-CV-1009-LY, 2015 U.S. Dist. LEXIS 106796 (W.D. Tex. Aug. 13, 2015), a sovereign immunity defense cannot be imputed over to a patent co-owner; in that case, an IPR was allowed to continue against the co-owner even as it was dismissed against the state university. However as of this writing, it has not been decided in court if a licensee can be protected from IPR by virtue of the licensor state university’s sovereign immunity. 32 See, e.g., Gardner, Gray, & Loken, supra note 24. 33 Sovereign Immunity Growing as Inter Partes Defense, Effect on Licensing Terms Unclear, 11 teCh. transfer taCtICs 7 (July 2017), available at https://www.quarles.com/content/uploads/2017/07/ Technology-Transfer-Tactics-July-2017-Reprint-Sovereign-Immunity.pdf (last visited Oct. 17, 2019). 34 Id. (citing interview with Dan Venglarik).

Reviewing inter partes review five years in 353 is also not sure that sovereign immunity applies at all when the TTO is a nonprofit entity affiliated with a state university (as is the case with the University of Wisconsin) rather than an arm of the state university itself. Also untested is the question of when and under what conditions the state university loses whatever sovereign immunity defense it had; for example, whether the university loses its defense if it proactively files suit against the challenger for infringement.35 And finally, in a recent case, the United States Court of Appeals for the Federal Circuit (“CAFC”) determined that tribal immunity did not apply to shield a patent from an IPR proceeding.36 Although the state university TTOs can distinguish their missions easily from that of the Native American tribe in question in the case, that case might be seen to represent a weakening of the sovereign immunity defense in general. All of this uncertainty could undermine the value of the defense as a bargaining chip. In particular, should the state sovereign immunity defense be waived once the university files suit against the infringer, then the defense is worth less to the licensee, who can only use it so long as he does not enforce the patent: So at the PTAB, sovereign immunity can be used as a defense. But if you file suit, you waive those rights (potentially). So it becomes a difficult sell as a bargaining chip—you can’t bring suit, so how is it a good defense?

Sovereign immunity might also be inapplicable to any one license situation because it might not be possible to choose between licensing opportunities. Most technologies are unique to the university. While a defense of sovereign immunity offered by a public institution might be compelling, if the technology the licensee needs was created at a private university, that is where the licensee will execute the license. As one TTO explained, different offices are not necessarily in competition with each other. Licensees will license the patents that they think they can commercialize, regardless of who owns them: We don’t think about other universities as our competitors. If you want our tech, license. You might not be able to get the same tech somewhere else. CRISPR is at X—if you want it, go there. When a company looks at three competing technologies, they are not deciding on who offers the best deal, and even if they choose the best deal, it is probably not about the IPR. It’s about “can they commercialize or not?”

And finally, since IPRs only come into play later in the life of the patent when the technology has become a successful product, for licensees, potential defenses to IPRs may seem very distant and at the very least lower in priority. The chance of IPR seems so far removed from the present that IPRs are not on the minds of even the savviest and most forward-thinking licensee. As TTOs explained, most licensees are primarily focused on whether the patent is going to grant at all:

35

In Ericcson v. Regents of the University of Minnesota IPR IPR2017-01186 -01197 -01200 -01213 -01214 -01219, the PTAB denied the University of Minnesota’s motions to dismiss the pending IPRs on sovereign immunity grounds because U. Minnesota had sued Ericcson for patent infringement. This case is pending on appeal before the CAFC. 36 Mylan Pharmaceuticals Inc. et al., supra note 26.

354 Research handbook on intellectual property and technology transfer IPR does not drive [licensing]—I haven’t heard a licensee say, “we would like to partner with you, but we are concerned about withstanding a challenge.” Much more the question—“is this thing going to issue?” IPR is typically not an issue unless you are dealing with an exceptional patent that is productized. It is not high on the list of priorities for a licensee, because IPRs only hit when the patent and product are hot.

And as TTOs from public universities hungry to use sovereign immunity as a selling point say is a sobering thought, most of their licensees do not even know what an IPR is, let alone how sovereign immunity might defend their patent asset. A large percentage of TTO licenses are executed with start-up companies, which are often unrepresented or have counsel that do not know much about IP. And even if they know that IPR can affect their patent’s value, often smaller companies are licensing their own inventions back from the university, so they are confident in the newness and nonobviousness of their invention. They view licensing as a means to a commercialization ends rather than an investment in the value of the patent. As a consequence, the sovereign immunity defense does not seem relevant to them. Far from being a differentiator, in a license negotiation, TTOs are finding that licensees do not even understand the concept: But in reality, startup licensees are not even thinking about substantive rights when they license— they are thinking: what does investor want to see, and what will help me make money. It’s almost as if the licensee is checking the box. Do they have the ability to practice and a clean chain of title? They might not actually look at the patents, let alone get into the weeds of the rights. They are small companies—they don’t understand patent law and how it works—they understand they need patents, but don’t understand the sovereign immunity or IPR process or what that means. It is surprising to find that small company licensees just aren’t very sophisticated regarding IP and don’t use outside attorneys, and even if they do, those attorneys don’t understand IP issues. I was surprised when I came—coming from large pharma, I was used to people that understood IP. With these little companies it doesn’t seem to have as much value.

In sum, public university TTOs report that they will sometimes bring up the concept of IPRs and a sovereign immunity defense to IPRs. But the topics are seldom brought up by the licensee, and never become important to bargaining in the license negotiation. Licensees will occasionally question patent value, but that negotiation point is more focused on whether a patent will issue, and its value to overall commercialization. Public universities may hope that the sovereign immunity defense to IPR will take on greater importance in the future, potentially as licensees become more aware, and the case law settles. But for now, sovereign immunity has not resulted in additional licensees, nor sparked more successful negotiation results: We have never had a person bring that up. It’s a theoretical argument, like we are usually playing checkers with licensees and this argument is from 3D chess. Licensees have to have a lot of things in a row, they are thinking about a lot of stuff. [An IPR defense] would be a lot to consider. I haven’t seen any budge in licensing, but 90% of our licensees probably have no idea about that aspect of the law. It’s a pretty rare case that we are licensing to a sophisticated and large entity. The big name companies probably [understand]. But there could be a disconnect between their business development folks and their legal team. I am not sure that the awareness of legal has percolated. Similarly, we are aware of the issue, but we haven’t brought it to the forefront of strategy in our business dealings. I have heard other speakers at conferences mention that third parties have raised the issue with them in meetings. I wouldn’t be surprised if it came up in the future for us. I have talked to some licensees about sovereign immunity issues. Having IP in the name of the university rather than in name of licensee if we are doing a sponsored research agreement might be

Reviewing inter partes review five years in 355 helpful to them. Mostly have [only] used it in the negotiation, but I am not sure of the effect of it. Some licensees still insist they have to own the IP no matter what—some say they want to jointly own. So [sovereign immunity] is probably not instrumental in completing deals, but we are bringing it up.

iv. Sovereign immunity is similarly not driving any shift in business model Because the defense of sovereign immunity could protect against IPRs, looking at the Saint Regis Mohawk Tribe facts for inspiration, it seems like there might be an opportunity for public universities to develop a cottage industry as a patent holding company. Since the Saint Regis case was narrowly construed to interpret tribal as opposed to state immunity in the context of an IPR defense, it might be possible for state universities to take assignments to third party patents in order to shield them from IPR attack. One TTO explained that they had in fact been approached, but that they refused to go down that path, based on a conflict with the TTO mission: People [have] called me up asking to assign patents to [university]—I said no—we don’t want to be in that business. Some faculty thought this was a good monetization idea. But our administration said no; we are for research and education. Our outcome is sometimes patents, and we will defend sometimes, but we didn’t want to shift to this model.

So even if courts do not extend the Saint Regis tribal immunity holding to state sovereign immunity actors, TTOs reported they are not interested in shifting their licensing policies and procedures in this way. v.

IPR is causing some TTOs to redraft their licensing documents to ensure that licensees pay for IPR One universal truth across the TTOs is that whenever possible, someone else should pay to defend against any IPR. This topic is discussed in more detail in the section related to budgeting below, but many TTOs have in fact changed their licensing policies in one way because of IPR: they have redrafted their license agreements to more clearly state that if a patent is licensed, the licensee is responsible for paying to defend any IPR proceeding.37 Most TTOs have both prosecution and enforcement sections in their license agreements. Many TTOs have added IPRs specifically to one or the other section, usually in a way where the licensee should front the cost for the defense instead of reimbursing for it. Although the TTOs realize that if the patent is licensed to a start-up company that they may not be able to pay that cost, like any other costs associated with litigation, the beginning bargaining position is that it will be the licensee’s responsibility:

37 Interestingly, forcing the licensee to fund the defense of an IPR can potentially set up a conflict of interest if the license is for freedom to operate purposes. It is likely that such a licensee may want the IPR to succeed, and thus may not want to defend it, if the IPR would invalidate the university patent and obviate the need for the licensee to pay license fees. David Schwartz, Universities Face An Increasing Likelihood Of Inter Partes Reviews, teCh. transfer enews blog, teasdale (Oct. 8, 2014), available at https://techtransfercentral.com/2014/10/08/universities-face-an-increasing-likelihood-of-inter-partes -reviews/ (last visited Oct. 17, 2019).

356 Research handbook on intellectual property and technology transfer We had to adjust our licensing language—when we license to a company, we put the possibility of this expense under the “prosecution” section rather than litigation so that the company pays for it. Generically licensee pays for all patent costs (directly or indirectly), so we assume that that includes IPRs. But we did add language after the AIA to explicitly call out the IPR process as being under prosecution. Yes, we have changed the language in the prosecution section to cover IPRs as patent-related expenses that the licensee is directly reimbursing. But we are also realistic—if [licensee is] not a big company, as soon as the expenses start to hit, regardless of what’s in the contract, it’s going to be an issue.

Some TTOs did not change their license language because they felt it was already sufficiently broad to cover IPR (“No changes to contracting—all locked up already.”). But they were sure that the expense would be covered. One final TTO is waiting to change its license language to specifically cover IPR until the sovereign immunity cases are resolved. Because sovereign immunity might turn out to be a complete defense to all IPR challenges, this TTO did not feel ready to assume that any party involved with its patents would be responsible for funding the defense of an IPR. 3. “Have IPRs changed your patent enforcement?” Since an IPR proceeding attacks a patent’s validity and costs money to defend, TTOs might think twice about enforcing a patent, lest the enforcement attract an IPR. Especially because IPRs are most commonly filed against university patents that have been commercialized, through enforcement, the TTOs might be opening up their most valuable patents to potential challenge. The choice is further complicated by the fact that most of the time the patents that are enforced have been licensed to a third party, which controls and will pay for the enforcement. So although the patent is the university’s asset, and the TTO will likely weigh in on the decision to enforce, it is the licensee that is calling all the day-to-day shots on the enforcement, and on defending any filed IPR. The interviews revealed that most TTOs do in fact specifically consider the risk of IPRs when they consider enforcement, although some TTOs weigh the possibility much more heavily than others. One of the more concerned TTOs stated: It used to be that if you believed there was an infringer, you could send a licensing request and at most risk a declaratory judgment, but you now risk an IPR. The financial barrier for a company [to file an IPR] is much lower than in litigation, so [IPR] is almost a certainty … [IPR is implicated in] a limited number of cases, but when it’s relevant, it’s very relevant.

As a result, at least one TTO reports they are extra careful during prosecution to make their “assets as impervious to IPRs as possible … meeting our duty to disclose, seeing that references get properly considered and things like that.” However, even that TTO acknowledged that the standard is likely not that different from before because they “were always careful because patents could always be challenged.” In addition, other TTOs bemoaned their quandary: they wanted to be able to file patents that could resist IPR, but one realized that they “can’t do prior art searching for all of our applications; it’s too expensive,” and the other wisely stated that all patents remain vulnerable, especially university patents due to their early stage: Inherent in how early you have to file, and how early the technology is, and how often people disclose before they file, university patents are vulnerable … We want to file the best patent, but we can’t always. You can’t control patent quality like you can in corporations.

Reviewing inter partes review five years in 357 Another TTO concerned about IPRs specifically consults with its licensees whenever it is considering enforcement, and specifically counsels them to guard against IPRs by running an extensive prior art search on the patent before taking any steps toward enforcement. They explain that this extra step is designed to find any art that was not discovered before and allow both the licensee and the TTO to more accurately assess the risk of losing an IPR attack on the patent. However, these TTOs are in the minority. For the vast majority of TTOs, the risk of IPR is not at all an important factor as to whether they enforce their patent. The reasons are economic, cultural, and logistical.38 Simply put, for most TTOs, there are so many other competing and more important considerations that go into the analysis about whether to enforce a patent such that the threat of IPR does not significantly affect the ultimate decision. First, TTOs execute a complicated economic analysis to decide whether it makes sense to enforce a patent against an infringer. IPRs are only a small part of that analysis, both because usually a licensee is footing the bill, and also because even if the university is paying, the IPR is only a small percentage of the whole. TTOs cited the relatively low cost of an IPR as compared to the total litigation expense. As one TTO stated, remembering the analysis it went through before enforcing a patent: “Given the costs of cases, the cost of IPR would still be relatively inconsequential … We were well aware that IPRs were possible when we filed,” and another TTO: “Is it a pebble on the scale? I suppose, but it is a pebble.” For these TTOs, concerns other than IPRs weighed much more heavily in the economic analysis. For starters, as one TTO reported, it must consider the unique challenges facing a university enforcement that make enforcement more difficult: since universities do not make and sell a product, an injunction is not a sufficient remedy for a university, and further, because the patents are so early stage and may be one of several patents being used in the infringing product, it is a riskier and more expensive proposition for a university plaintiff to prove infringement, let alone damages. TTOs attempt to consider all aspects of the case that may weigh into the overall chance of success, and the IPR may have only a small role to play: Litigations are always a time and money sink. We have to evaluate whether [the whole case] is worth it on the merits, how long it will last, who the adversary would be, and whether we can come to a settlement agreement first. [We] always have had to do this—IPRs are just part of the analysis now.

The analysis of whether to enforce a patent is at least as much if not more about whether the patent is strong, valuable and infringed, and even the kind of technology the patent is over, than about the risk of IPR, even if losing the IPR means losing the asset altogether. The economic analysis is multifaceted and complex. Beyond economic reasons, TTOs also must address cultural influences that prevent them from pursuing patent enforcement. Many TTOs claim that they face philosophical hurdles to trying to enforce their patents that take precedent over any risk of IPR as a deciding factor. TTOs confess a common institutional reluctance to enforce patents at all because of a real or perceived conflict with the TTO mission to simply protect and license (but not enforce against infringement of) inventions. For example, one TTO explained: “Every patent owner knows if

38

For an excellent discussion of the complex set of considerations behind a TTO’s decision to enforce its patents, see Jacob H. Rooksby, When Tigers Bare Teeth: A Qualitative Study of University Patent Enforcement, 46 akron l. rev. 169 (2013).

358 Research handbook on intellectual property and technology transfer you do not enforce sometimes, you have no credibility … [but] our mission is tech transfer. The only time we would enforce a patent is when the technology is significant and widely practiced, and licensing overtures have gone ignored.” Another TTO questioned whether enforcement is ever appropriate for a TTO, but also acknowledged that a culture of nonenforcement creates a tension that makes serving different constituencies difficult: We have one case where we are exploring enforcement. There are bad faith facts, the inventors are worked up, so we have some buy-in to at least investigate. But that’s not what we articulate as our mission. We are supposed to be about getting the technology out into the world. If someone is already using [the technology], is that mission accomplished? But our client is also a series of inventors— they want attribution, and the money means something to them; it is hard to tell them “don’t worry about it.”

Relatedly, TTOs also want to ensure that their actions uphold the good reputation of both the university and its inventions. Depending on the philosophy of the TTO, this can either mean no enforcement (if enforcement is seen to be counter to the university’s mission) or stronger enforcement (so as to reinforce the integrity of their licensing and to show that the patents have value). The first approach is illustrated by this quote from an institution that does not often enforce its patents: The issue is that [university name] doesn’t want to go off and sue people. We often get requests to sue from our PIs [principal investigators]. We go through the motions, but we are just not going to send the mean letter. We at most are going to send the “Gosh, are you aware of this patent?” letter.

But another TTO has the exact opposite approach since they believe in stronger enforcement: We are one of the highest [patent] filers in the nation. We have valuable IP assets with economic impact and a mission of developing IP. It is important we are seen to be protecting our IP. We don’t want to be frivolous, and we don’t pursue everything, but we need to maintain our reputation. IPRs are one of several factors we consider [when we enforce]; we want our patents to stay valid.

A final TTO summarized the tensions well: [It’s a] fine line that you walk—people have to know you would enforce, but you cannot go after everyone. We try to be fair; reputationally it’s both about the strength of our patents and also about our conduct as partners and in the industry. And it’s about our goals. We are not revenue driven, like a company. We are part of the university.

Finally, TTOs may not be considering IPRs heavily in their enforcement analysis because they are overwhelmed by administrative hurdles. The TTOs are only one part of the university. One TTO lamented that even when the case for enforcement was clear, the enforcement calculus was still complicated because it depended on the players involved: “the politics are hell; we might find ourselves suing a company that has a founder that will fund a [university] building.” Another TTO discussed the many ways that the university was connected to the community and the inherent conflicts of interest that created: The TTO represents one aspect of the university, but not all of [university name]. You don’t make friends this way [by enforcing patents]. There are donors, partners, etc … And overall, it’s not like our office brings in more than the development office, so we have to listen to them … If [an

Reviewing inter partes review five years in 359 infringer] wants to give money to research or hire our students, we’re just not going to do it [enforce a patent] … Relationships are now far more complicated because students are going to work at [companies that are infringing our patents], they are starting their own companies, professors and alumni are on boards, it’s a very complicated mine field you have to navigate. We have to be pretty positive about the validity and the strength of the patent and that someone’s product infringes before we would consider it … It is so much more complicated when you are a university rather than a corporation.

Before getting approval to file a patent lawsuit, TTOs often face layers of bureaucracy, such as getting sign-off from the highest levels of the university, or even of the state attorney general’s office in the case of a public university (when a state university enforces a patent, technically it is the state that is the plaintiff and the case is funded by taxpayer dollars). As one TTO explained the process at her university: You have to get it approved by the Office of General Counsel, you then go to a subcommittee of Regents who have to decide if it’s a gamble worth taking, and litigation counsel might have to convince them; then you have to structure a fee schedule, possibly look at patent strength scores, evaluate who the defendant is … In sum, it’s a last resort.

The analysis becomes even more complicated when the university is part of a larger system, where the possibility of a conflict multiplies across the many related schools. As a result, enforcement of a university patent is exceedingly rare and does not depend heavily on the possibility of IPR. Many of the TTOs interviewed did not think that their office had ever enforced a patent, and not one of the TTOs currently has more than four active litigations. In sum, there are many conditions that go into whether a TTO enforces a patent. The possibility of IPR is one consideration that TTOs might keep in mind, but it is not near the top of the list. 4. “Have IPRs caused changes to your budgeting?” Because increased risk often equals increased cost, the new IPR proceedings might be forcing TTOs to allocate money differently to ready themselves for any eventual IPR challenge. I asked the TTOs to consider whether they had changed their budgeting, both to support new filing techniques designed to strengthen the quality of their patents, as well as to set aside money to address IPR defense, either within or apart from enforcement litigation. Those who have implemented slightly modified filing policies to improve their patent quality at least in part because of IPRs, have modified their budgets accordingly. One TTO is paying for more continuation practice to keep claims alive should the need come to hedge bets on a patent under attack through IPR. Although continuations are not terribly expensive, this TTO has had to increase the budget for its patent filing. Similarly, another TTO is taking steps to improve the quality of its patents in the face of potential IPRs by sending its provisionals out to be drafted by outside attorneys, instead of handling all its provisional application filing in house. That university is filing fewer patents now, but also was able to reallocate the filing budget to accommodate the change in policy. However, as to budgeting specifically for IPR defense, although some TTOs seemed to have some misgivings about it, not one TTO reported changing its budget to accommodate any additional risk presented by IPRs. Some specifically mentioned that they had not experienced enough IPRs to worry about allocating funds:

360 Research handbook on intellectual property and technology transfer No—when we set our budget, we go through various questions about what costs may come up this year … An individual issue may cause us to adjust our overall budget for the year, but IPRs are not a separate line item. We don’t anticipate them and don’t budget for them because we haven’t had many.

The consensus across the TTOs is that budgeting specifically for IPR defense is unnecessary given that the risk of exposure is so low. They cite that a rare set of circumstances would have to occur for an IPR to be filed. TTOs also complain that they barely have enough money to cover the cost of the necessary patent filings they hope to do within the office. Although they might like to have some money put aside for IPR emergency or for any sort of enforcement, they are already barely meeting their current costs: I’ve been thinking about a buffer—do we need one? If a patent is valuable, should we have a war chest—to enable us to go after a third party if needed? For example, if a licensee says we don’t owe you a royalty, because we are not using the patent any more, we have no budget to make this inquiry. We don’t have money set aside to do this. We haven’t even progressed to asking where [that money] would come from! I’ve been with the office for 13 years, and our budget has been essentially the same. We pay all costs for licensed and non-licensed patents. To think about a reserve is hard when we don’t even have enough [money] to patent.

Besides the low risk of occurrence, TTOs also explain that they have decided not to budget for IPRs because in most cases the university would not be responsible for funding the defense. Since most IPRs occur against licensed patents, TTOs assume that in most cases the licensee— not the university—will have to pay for the defense. And even in the smaller set of situations where the university would be responsible for paying, such as when the IPR challenges an unlicensed patent, a patent is nonexclusively licensed to several parties (such that any one licensee is not responsible for paying); or when the licensee cannot afford to defend the patent, the TTOs can rely on traditional as well as possibly some new options to finance the defense. One TTO described her relationship with a law firm on a fee arrangement that has helped with —and even encouraged—enforcement over the past ten years: Occasionally the firm has done some evaluation of some patents, has come to us proposing enforcement. That’s how these last two cases came about. We convinced the licensee to go ahead with enforcement. We may decide to do a few others … As a whole, the university is risk averse in terms of IP; they try to have relationships with companies, try not to sue those we have relationships with … But coming from industry, I feel you should enforce. It is a way to get value out of your patents if you can’t license. Frankly we have a lot of patents on the books just sitting there.

And other TTOs explain a new model for financing enforcement and IPR defense that involves a separate litigation financier providing the money for the proceedings: The universities generally don’t fund the litigation themselves; they look for a law firm on contingency, or they engage with a litigation financier. There is more of that happening in the industry. It is expensive, but these are investment firms investing in IP through financing the litigation. It is basically debt financing; they advance the money for the litigation. If it is successful, they recover a multiple of what they fronted (through damages after the university pays back expenses) and then a percentage of a share of royalties going forward.

Several TTOs mention the litigation financier option as possibly increasing TTO willingness to enforce patents (and defend against IPRs) because it erases the stigma of working with

Reviewing inter partes review five years in 361 a contingency law firm and “feeling like a troll.”39 Because under the financier scenario the TTO can choose to work with its preferred law firm and control the litigation, it may allow TTOs to enforce their patents in a way that feels more within the mission of the university. In fact, these firms seem to be specifically targeting universities as a viable option: What’s my budget for enforcement? Zero. None. But we almost don’t need it; litigation funders are a pretty big part of the litigation scene now. They are like VC [Venture Capital] firms sitting on funds and [they] will buy a piece of litigation. You hire the firm, they will pay the fees, in exchange for 40% of revenue. This is available to universities. A lot of these firms were at AUTM [annual conference of the Association of University Technology Managers] to pitch their services.

Should working with a law firm or a litigation financier not be possible, the TTOs explain that they would have to take those situations as they come. Each describes a process whereby they would analyze whether it was worth defending the challenged patent, and if so, where they would go to request funding. One TTO surmised the extra funding would come from litigation funding the TTO always receives from the home school of the affected patent; another TTO said it would seek funding thorough the general counsel’s office that would be allocated by a committee, and several said that they would have to find the money elsewhere in the TTO budget or from a different university department: I do think about it—what would we do if and when IPR were to happen? We would look at the status of the case—how much is it worth it to us to fight. For each case we analyze risk—we consider whether we file overseas, divide the patent, etc. There is always a cost benefit analysis. So we would look at all facts—how old is the patent, what is its track record of licensing, is it a slow market that will be important even if not yet important … I have a hunch that if they are filing an IPR [especially if outside a litigation], it is probably an important patent. If it is that important, we probably would find money to manage the cost. We would reach out to the university to support the effort, or reallocate some costs.

Regardless of the ultimate decision, TTOs—at least at this stage—are content to maintain current budget policy and wait to deal with an IPR defense only when and if the problem should arise.

IV.

WHAT DOES THE FUTURE HOLD FOR TTOS CONTEMPLATING IPRS?

Even though it has been more than seven years since the IPR procedure was implemented, the effects of IPRs have not been what might have been expected. Whether by virtue of the early-stage technology that TTOs patent, their philosophy of limited enforcement, or use of a structure whereby patents that are commercialized are usually licensed, TTOs have not

39 See Rooksby, supra note 38, at 187–91 (discussing how hiring contingency law firms—as opposed to litigation financiers—feels more “like a troll” to certain TTOs); see also Jacob H. Rooksby, University Involvement in Patent Infringement Litigation, 47 les nouvelles 8, 17 (2012) (“However, some technology transfer professionals who indicate their university has engaged law firms on a contingency fee basis in the past have stated they view these risk-sharing arrangements as perilous or even inappropriate for universities to accept.”).

362 Research handbook on intellectual property and technology transfer yet had to systemically change their filing, enforcement, licensing, or budgeting policies to accommodate IPRs. But the TTOs I interviewed were withholding judgment about what the next five years would bring. Looking forward, there may be changes in case law that would encourage TTOs to react. Although IPR proceedings do not yet drive TTO decision making, case decisions could still settle uncertainty and give TTOs finality on how to proceed. For example, even as sovereign immunity is not providing much of a licensing advantage to public universities yet, it would still be efficient to resolve the uncertainty around its applicability. Knowing at what point the state university patent owner waives the sovereign immunity defense because of involvement in a litigation proceeding will definitely help plan litigation strategy for even the few numbers of enforcement proceedings going forward. Similarly, it is hard to use sovereign immunity as a bargaining chip in a licensing negotiation if it is not clear if the defense can shield the patent licensee. TTOs should continue to monitor IPR cases until that question regarding licensees is resolved. It will be interesting to see if the CAFC’s decision on tribal immunity will have implications for state sovereign immunity cases.40 And finally, there are still questions left open by Oil States about the constitutionality of IPRs under the Due Process and Takings clauses, and whether IPR should be able to apply to patents that issued before the AIA. It is not inconceivable that the CAFC or even the United States Supreme Court might have to consider those questions in the terms ahead. Even though IPRs have not affected TTO policy significantly, it would still be a relief if IPR proceedings were completely shut down on constitutional grounds. TTOs must also keep informed on executive branch and legislative activity related to IPRs. The PTAB is now grappling with how to retroactively apply the results of SAS Institute to IPRs that were incorrectly decided. This holds huge implications for the few IPRs to which TTOs are party, as to whether the PTAB needs to consider the additional claims, and what the estoppel effect of these IPR decisions will be on any infringement lawsuits. In addition, USPTO Director Iancu has been outspoken about the IPR process, and he acknowledges that industry blamed IPRs and associated uncertainty about patent value as contributing to the US falling from first place to twelfth on the United States Chamber of Commerce annual Global IP Index patent list from 2017 to 2018.41 Following Director Iancu’s announcements that he would like his office to focus on “increasing the reliability of the patent grant,” specifically addressing IPR proceedings, hoping to “balance rights-holder’s and rights challenger’s interests,”42 which led to the USPTO rulemaking that harmonized standards of review between the PTAB and district courts, in 2019 the US returned to a second place ranking.43

40

Or if in fact the case may be overturned. As of this writing, the case is on petition for certiorari with the United States Supreme Court. 41 U.S. Chamber International IP Index, global InnovatIon polICy Center (6th ed., Feb. 2018), at 35, available at http://www.theglobalipcenter.com/wp-content/uploads/2018/02/GIPC_IP_Index_2018 .pdf (last visited Oct. 17, 2019). 42 Andrei Iancu, Director, at U.S. Chamber of Commerce Patent Policy Conference (Apr. 18, 2018), available at https://www.uspto.gov/about-us/news-updates/remarks-director-andrei-iancu-us-chamber -commerce-patent-policy-conference (last visited Oct. 17, 2019). 43 U.S. Chamber International IP Index, global InnovatIon polICy Center (7th ed., Feb. 2019), available at https://www.theglobalipcenter.com/wp-content/uploads/2019/02/023593_GIPC_IP_Index _2019_4Pager.pdf (last visited Oct. 17, 2019).

Reviewing inter partes review five years in 363 IPRs have also caught the attention of Congress. Legislation introduced in 2018 in the House, with a parallel bill reintroduced in the Senate was designed to address some of patent owners’ most salient criticisms of IPR proceedings. Its sponsors are hoping to reintroduce the bill in the 2019 Congress. Entitled the Support Technology & Research for Our Nation’s Growth and Economic Resilience (“STRONGER”) Patents Act of 2018,44 the bill, among other things sought to: align the PTAB and district court claim construction and burden of proof standards; impose a standing requirement to bring an IPR; give deference to prior court adjudications of validity and limit the number of challenges that can be brought against the same patents and claims; allow a patent owner to more easily amend claims during an IPR; and create separate panels of judges to decide on instituting an IPR and ruling on the merits. The bill also addressed a few issues to make infringement lawsuits easier to file and win, namely revising the standard for inducing infringement, making it possible to infringe even if manufacturing of the infringing product occurs overseas, reinstating the presumption that an injunction is appropriate if a patent is found valid and infringed, and giving the Federal Trade Commission the authority to file suit against parties that send demand letters issued in bad faith for purposes of extortion. Some form of this bill has been introduced for several sessions. Congress is responding not only to the numbers of invalidations, but also to some egregious misuse of the IPR system, as with hedge fund managers like Kyle Bass shorting stocks of biopharmaceutical companies and then launching IPRs to drive stock prices down.45 If IPRs continue invalidating patents at the current pace, and the numbers of IPR proceedings stay high, and stories like what is happening with Kyle Bass continue to pique the interest of Congress and the public, the legislation may gain traction.46 IPRs have transformed the patent litigation landscape. They will conceivably continue to play an important role for years to come. Although survey answers confirm that IPRs have not yet affected technology transfer to the point where TTOs must make significant policy changes, IPRs are an important consideration as TTOs make decisions about how best to fulfill their missions. TTOs should continue to follow the latest court decisions and executive and legislative proposals on IPRs. At some point IPRs may become a more significant part of the TTO landscape, particularly if shifting ideas about patent value, licensees with different licensing needs and philosophies, and new limits on patentable subject matter force TTOs to grapple with broader visions of mission that encompass increasingly more patent enforcement.

44 H.R. 5340, 115th Cong., available at https://www.govtrack.us/congress/bills/115/hr5340 (last visited Mar. 28, 2019). Analysis of its salient points is at Steve Brachman, STRONGER Patents Act Introduced in House, Seeks to Strengthen Crippled Patent System, Ip watChdog, Mar. 28, 2018, available at https://www.ipwatchdog.com/2018/03/26/stronger-patents-act-house/id=95188/ (last visited Oct. 17, 2019). 45 See Gene Quinn, Post Grant Patent Challenges Concern Universities, Pharma, Ip watChdog, Apr. 1, 2015, available at https://www.ipwatchdog.com/2015/04/01/post-grant-patent-challenges -concern-universities-pharma/id=56351/ (last visited Oct. 17, 2019). 46 Many of the ideas set forth in the STRONGER Patent Act have been suggested separately by commentators as being good for patent owners and specifically small businesses. See Melissa Cerro, Navigating a Post America Invents Act World: How the Leahy-Smith America Invents Act Supports Small Businesses, 34 J. nat’l ass’n adMIn. l. JudICIary, 225, 226, 229 (2014); Neal Solomon, Solutions for Inter Partes Review: Restoring Patent Rights and Respect for the Presumption of Validity, Ip watChdog, Aug. 10, 2017, available at https://www.ipwatchdog.com/2017/08/10/solutions-inter -partes-review-restoring-patent-rights-respect-presumption-validity/id=86680/ (last visited Oct. 17, 2019).

17. Data governance and the emerging university Michael J Madison

I.

INTRODUCTION

As data become more central to university research practice, data governance policies implicate questions that are more fundamental than those supplied by framings grounded in existing technology transfer programs. Should data be “open” or “closed”? Are data “basic” knowledge or “applied” knowledge? Neither question suits data and data practices themselves. Governing data suggests re-thinking the character of universities as research institutions. What should universities do with research data collected and generated by their researchers, and why? The question has both positive and normative attributes. Positively, the questions concern the complexities of data as an information resource in a specific setting. Normatively, the questions concern a novel way of illuminating continuing challenges associated with governance of the university itself, as a knowledge-based enterprise. “What should be done with data?” is a mode of asking “what are the purposes of universities?” This Chapter weaves these two themes together. Combined, they represent both a forward-looking inquiry into a still-critical institution and a case of addressing broad and abstract social questions via understanding their institutional contexts. Positively and pragmatically, any modern research university has at hand vast quantities of data and datasets, much of them collected and generated by the university’s own faculty and students and stored managed locally, some of them accessed by the university’s researchers but stored and maintained elsewhere. Data constitute critical research resources across multiple domains of inquiry and practice, rather than niche or specialized tools. Recognizing their cross-cutting significance, in Europe, in the United Kingdom, and increasingly in the United States and elsewhere, universities are preparing informal guidance and formal policies to assist researchers and others in storing, curating, securing, and sharing and distributing research data.1 Funders, regulators, and scholarly publishers may expect universities to share data and may also expect universities to secure data. Researchers themselves often want both to share data and to control it. Amid a diverse university environment, what should data governance policies consist of? Normatively, those pragmatic concerns overlap with advocacy among many researchers and information policy professionals of principles animating the ideas of Open Science and Open Data.2 That advocacy takes some form of the following arguments. The enormous volumes of data that underlie contemporary research constitute a massive knowledge resource that ought

1 See, e.g., Martin Donnelly, Five Steps to Developing a Research Data Policy, dIg. CuratIon Ctr. (Jan. 2014), available at http://www.dcc.ac.uk/resources/policy-and-legal/five-steps-developing -research-data-policy/five-steps-developing-research (last visited Oct. 17, 2019). 2 See, e.g., Open Science by Design: Realizing a Vision for 21st Century Research, nat’l aCads. of sCI., eng’g, and Med. (2018); Final Report—Science as an Open Enterprise, royal soC’y (2012), available at https://royalsociety.org/topics-policy/projects/science-public-enterprise/report/; OECD

364

Data governance and the emerging university 365 to be openly accessible to researchers everywhere. It is argued that data constitute knowledge. To make knowledge useful, in various modern senses, knowledge ought to be open, and knowledge ought to be shared.3 In the abstract, those propositions are nearly incontrovertible, but advancing and implementing them, and addressing complementary and sometimes opposing goals, means approaching them institutionally, in programs and practices rather purely as concepts. In what specific respects should data governance embody or reflect aspirations toward “openness”? This Chapter sets those questions in the broader contexts of the university as an institution and of data as university-based resource. Highlighting that context implicates intersections and overlaps between emerging practices and goals regarding research data in universities, on the one hand, and established practices and expectations regarding intellectual property (“IP”) in the university, particularly university-based inventing, patenting, and technology transfer, on the other hand. Much as the Open Science movement aims to encourage researchers, administrators, funders, publishers, and regulators to establish institutional frameworks to promote data sharing, at the same time existing institutions of university-based IP production may push, instead, toward embedding university research data in closed, proprietary frameworks and practices. One purpose of the Chapter is to anticipate and describe that possible conflict using the relatively well-trod pathways that distinguish between open and proprietary frameworks in addressing how to manage research and scholarship in the university, as related modes of knowledge production. Those pathways include conceptual divides between the production and dissemination of basic or generalized knowledge, on the one hand, and useful or specific applications of knowledge, on the other hand. Those conceptual divides are traced institutionally via university-based forms and practices of IP law, namely patents and copyrights. Basic or generalized knowledge is documented and disseminated via scholarship. Scholarly works are governed by copyright law and are subject to dual normative imperatives: broad distribution and no expectation of reward or return to the researcher. Useful or specific knowledge is disseminated via being encoded into inventions, governed by patent law, and their dissemination is linked to rewards to both inventors and their university employers via technology transfer systems. In short, “open” knowledge production corresponds roughly to university institutions associated with scholarly production and to copyright; “closed” knowledge production corresponds roughly to university institutions associated with inventions and patent. In this scheme, where does data fit? The Chapter argues that like inventions (in part) and copyright works (in part), data and datasets constitute both modes of useful knowledge and of basic or general knowledge. Shoehorning data into patent-like technology transfer practices risks closing off access to and use of important, basic, research resources. But the alternative approach, given the either/or “open v. closed” premise, is unworkable for different reasons. Assigning data rhetorically and practically to the more open institutions and practices associated with copyright and scholarship risks harm to data-intensive research itself, by imposing a “thing”-like character on data governance when research practices often assume or rely on data’s “flow”-like character.

Principles and Guidelines for Access to Research Data from Public Funding, OECD (2007), available at http://www.oecd.org/science/sci-tech/38500813.pdf (last visited Oct. 17, 2019). 3 See, e.g., Joel Mokyr, The Gifts of Athena: Historical Origins of the Knowledge Economy (2004).

366 Research handbook on intellectual property and technology transfer A second purpose of the Chapter is to introduce data as a perspective on the evolving character of the research university itself, highlighting conflicts between openness and proprietary claims that build on a less common perspective. Universities today have many purposes, but central to almost all descriptions of the university is the production, collection, curation, and distribution of knowledge.4 Traditionally and conventionally, the university exercised those functions through teaching, collecting physical specimens, and engaging in research and publishing scholarship. In other words, for centuries universities generated and shared knowledge through defined pathways of experience: via interactions among humans, via interactions between humans and objects, and via text.5 Data-based research constitutes each of these, in one sense, and none, in a different sense. As the university continues to evolve, the new salience of data in the research enterprise illustrates some central conflicts over what the university is and what it should be. Perhaps, at its core, the university is still a knowledge institution. If the university is now a data-oriented institution, as it now appears to be, then one might conclude that data are knowledge. That’s a functional claim rather than an ontological one. As a basis for governance, it falls short, because for the reasons suggested earlier, the “open v. closed?” framing of university-based knowledge production suits data poorly. Perhaps the foundational assumption of modern research universities, that they should distinguish between basic or general knowledge and useful or specific knowledge, should be re-thought. Perhaps, instead, the university is no longer primarily a knowledge institution; data themselves supply the university’s new central organizing principle, and entirely new governance principles should be developed. Organizationally, the Chapter begins by laying out briefly its chief pragmatic concern: the rise of data-intensive research and the role of data and datasets in the modern university. The problem to be described, as it were, is how the university ought to manage those resources for maximum social benefit and minimal social and private harm. The Chapter takes a pragmatic approach rather than a first principles approach. General questions about the virtues and drawbacks of data collection and of data sharing and data security are largely left for others to explore. Here, questions relate primarily and more concretely to what should be done as data governance in a specific institutional setting, and how, and with what implications. Having described the character of modern data-intensive research, the Chapter turns to a brief account of the university itself, tracing themes of openness and the character of university-based knowledge through history, to the present. The Chapter then combines themes offered by the review of data-based research and the history of universities. Claims of openness attached to emerging data-based research practices in the 21st-century university stand in contrast to the proprietary institutions of patent law and technology transfer carried over from the late-20th-century university. But that statement of “open v. closed?” affirms a case of institutional continuity as much as it offers a case of contrast. The university has wrestled for centuries with conceptual and practical demands that applied knowledge research practices be accommodated as complements to basic knowledge research practices; the modern research university is a dynamic amalgam of open and closed

4

See Jaroslav Pelikan, The Idea of the University: A Reexamination (1992). See, e.g., Michael J. Madison, Brett M. Frischmann, & Katherine J. Strandburg, The University as Constructed Cultural Commons, 30 wash. u. J. law & pol’y 365, 381–6 (2009). 5

Data governance and the emerging university 367 governance.6 The 20th-century university has been criticized on the ground that its open, knowledge-driven mission is unduly influenced by demands that knowledge be controlled and proprietary. Emerging interest in data governance may push toward a renewed focus on openness. But trying to situate data in that framing may do more harm than good. Instead, the Chapter offers the concept of the data-intensive university, a fundamental reframing of institutional purpose that permits asking questions about data governance that neither assume nor reject “open v. closed?” (or “basic v. applied?”) as foundational premises but instead place governance questions firmly in new institutional context.

II.

THE MEANINGS AND APPLICATIONS OF DATA-INTENSIVE RESEARCH

Data challenge conventional distinctions between basic or general knowledge and applied or useful knowledge. Data have long been understood to consist of important inputs into research programs, as well as research tools or instruments. Today, data may also constitute important research outputs. That plural character challenges the application of standard governance frameworks grounded in IP and other information law. Discussions of data and data governance should begin with some history and some cultural and legal context. The rise of research data and the problem of knowledge governance concerning data are woven from threads of different colors. This Part addresses both established and novel attributes of data in research, and it outlines relevant frameworks in law. A.

Data: What’s Old, What’s New?

Scholars and scholarship begin with data. They always have. Historians organize and interpret events in the world. Geologists organize and interpret attributes of the earth. Biologists and psychologists organize and interpret attributes of life forms. What has changed in recent decades is the power of computing technologies. Creating and interpreting massive digital datasets is now within the reach of both specialized fields and disciplines that are organized around those computational problems and also pre-existing fields and disciplines that have begun to adopt and explore computational techniques. Gathering, processing, and interpreting super-sized datasets is becoming the norm in many fields, not just a few. “Data” should be defined, initially but broadly, as evidence, or generally, as “representations of phenomena in the world,” natural, physical, material, and cultural, used in connection with research and scholarship.7 That is a functional definition. Data are gathered, created, and used by researchers as informational inputs into interpretive processes, practices, and technologies that themselves are useful as means of generating and sharing semantically meaningful results. Data may be infrastructural resources, in the sense that they provide bases for multiple applica-

6 Donald E. Stokes, Pasteur’s Quadrant: Basic Science and Technological Innovation 26–57 (1997). 7 See Christine L. Borgman, Big Data, Little Data, No Data: Scholarship in the Networked World 17–29 (2015).

368 Research handbook on intellectual property and technology transfer tions, interpretations, and other uses in complex systems.8 As inputs, data may be regarded as kinds of evidentiary raw material. As infrastructural resources, data may be regarded as kinds of multi-purpose research tools, or instruments. By convention, we often treat “data” as a foundational resource and then call the interpreted results “information” or “knowledge,” implying a typically linear or hierarchical relationship. That conventional linearity or hierarchy has attributes of prestige as well as function. In both senses, the convention has an ancient lineage. At least beginning in the 17th century, and perhaps as long ago as Greek antiquity, the fruits of observation, investigation, and study were divided between basic, or general, or eternal principles governing the world, considered to be knowledge, the province of philosophers; and applied or useful information, the province of mere engineers and technologists. “Data” and the singular “datum” emerged during the 17th century as part of the syntax of Baconian science, but its plural uses gave it life in both basic science and applied technology domains. The phrase “data processing,” now somewhat out of fashion, captured a mid-20th-century sense in which “data” were considered to be “raw material” for higher-order industrial analysis.9 Even in traditional framings, there has rarely been a hard or fast divide in practice between “data” and “knowledge.” Rather than thinking of data as parts of linear pathways leading to knowledge, both historians of science and scholars of information believe that it is more appropriate to understand data as parts of overlapping processes of evidence collection and interpretation. Even if data are in part form that contribute sequentially to the production of information and knowledge, data are equal part flow, in which evidence of phenomena lead to interpretation and possible use, perhaps being documented and disseminated in various forms, and those forms of knowledge and information in use provide further evidence for further interpretation, and so on. For example, in many fields, sociologists of science teach that a sense of data as solely or primarily “raw material” will recede from explicit consideration via long processes of embedding observational details in domain-specific concepts and categories. A “wave form” was “data” once, to physicists, but today it may simply be a wave form, whose meaning and function are self-evident within the field. Data cease being things or objects in themselves. They are codified—black boxed, sociologists might say—as part of disciplinary practices.10 Data are not “data”; they are research. Datasets and data collection as such are increasingly viewed in some scholarly domains as credit-worthy research and scholarly activities in their own rights. Some both inside and outside the university setting argue that production and publication of a complex dataset ought to be treated as a meritorious knowledge producing activity, eligible for consideration in recruitment, retention, and promotion decisions for students and researchers.11 Interventions in those debates vary from field to field, as one would expect. Not everyone

8

See Paul N. Edwards, et al., Knowledge Infrastructures: Intellectual Frameworks and Research Challenges (Report of a workshop sponsored by the National Science Foundation and the Sloan Foundation, University of Michigan School of Information), May 2012, at 25–8; Brett M. Frischmann, Infrastructure: The Social Value of Shared Resources 225–6 (2012). 9 See, e.g., JoAnne Yates, Control Through Communication: The Rise of System in American Management (1989). 10 See Bruno Latour, Science in Action: How to Follow Scientists and Engineers Through Society 2–3 (1987). 11 Borgman, supra note 7, at 47–53.

Data governance and the emerging university 369 agrees that publishing a novel dataset should be treated by a field as equivalent to publishing a meritorious journal article. Newish phrases such as “data curation” and “data carpentry,”12 and practices of data visualization,13 express related sentiments. Data are not simply “out there” waiting to be found, but have to be assembled and shaped, as cultural phenomena. Data and datasets are interpretation as much as representation. Judith Donath offers the phrase “data portraiture” as an alternative to “data visualization” to call attention to the possibility that “visualization” suggests a practice that might be regarded, inaccurately, as free from professional judgment, or the data, like human identity, does not express itself. Efforts to interpret and represent data (themselves, representing some other phenomena) have foundations in the arts as well as in more conventionally technical fields.14 Data and datasets may be handmade. Data and datasets may be generated by algorithms. In either case, they nearly always represent disciplined choices about how material in the world is gathered, combined, made salient (or not), stored, and communicated. They are themselves modes of basic or generalized knowledge, meaning that data constitute research results and products generated either by domain-based researchers or by specialized researchers in the field increasingly known as data science. The proposition that data are simultaneously input, tool, and generalized knowledge is hinted at by the related idea of data as flow. That concept has now been institutionalized by scholars via the information data lifecycle, which rejects the idea of data as solely “raw material,” “stuff,” or fixed “object,” leading to interpreted knowledge, in favor of an ecological approach.15 Data are representations not of a particular state of the world (as a scientific researcher might consider them), but instead representations of intersections among evolutionary processes that define the relevant natural or social world and the ecological processes by which researchers identify, create, curate, and disseminate data. Some designers of health care information technology systems build on that premise by promoting “data liquidity”: ensuring that the right data is accessible to the right person, at the right time.16 Exploring data as flow as well as form takes on new significance in the research enterprise by virtue of the rise of data-intensive research, sometimes referred to as data-intensive science or “Big Data.”17 Whatever the label, the stakes of data governance are greater and more explicit than ever before. Data-intensive science has its roots in the fact that beginning roughly twenty-five years ago, a handful of specialized scholarly disciplines began to grow up specifically to deal with opportunities to explore super-large digital datasets, particularly in particle physics, astronomy, and astrophysics, on the one hand, and genomics, on the other hand. In each case but in different

12 See Tracy K. Teal, et al., Data Carpentry: Workshops to Increase Data Literacy for Researchers, 10 Int’l J. dIgItal CuratIon 135 (2015), doi:10.2218/ijdc.v10i1.351. 13 See Edward R. Tufte, Visual Explanations: Images and Quantities, Evidence and Narrative (1997); Edward R. Tufte, Envisioning Information (1990); Edward R. Tufte, The Visual Display of Quantitative Information (1983). 14 See Judith Donath, The Social Machine: Designs for Living Online 209 (2015). 15 See, e.g., Alex Ball, Review of Data Management Lifecycle Models, REDm-MED Project Document (U. Bath, 2012), doi: redm1rep120110ab10. 16 See Paul K. Courtney, Data Liquidity in Health Information Systems, 17 CanCer J. 219 (2012), doi:10.1097/PPO.0b013e3182270c83. 17 See, e.g., Data-Driven Innovation: Big Data for Growth and Wellbeing, OECD (2015), available at www.oecd.org/sti/data-driven-innovation-9789264229358-en.htm (last visited Mar. 28, 2019).

370 Research handbook on intellectual property and technology transfer respects, customized information technologies were developed to process exceptionally large collections of observations of the world, at super large scales (astronomy and astrophysics) or at super small scales (particle physics and genomics). In difference respects, researchers linked observational capabilities, high speed communications and computer networks, and storage and image processing technologies, sometimes including high capacity systems for converting information resources from analog (material) to digital (virtual) forms. Notably, the datasets themselves were not the province of university-based researchers operating alone. The scale of the research enterprise meant that complex partnerships to produce and manage datasets developed and shared among academic researchers, researchers in other settings, philanthropies, national governments, and publishers.18 Calling additional attention to contributions from industry, Caroline Wagner calls this cluster of developments “the new invisible college,” describing science and technology “as an emergent networked system rather than as a national asset.”19 In the new invisible college, the idea of data as flow takes on a set of concrete institutional forms. Moreover, abstracted from these specific domains, powerful new computational resources and their descendants—technologies for sensing and observing, information processing and storage capabilities, network communications facilities, and related interpretive techniques— have been introduced gradually to and accepted in other research domains, such as the so-called digital humanities. Ever-increasing swaths of the university now fall under the general heading of “data-intensive” science, discovery, research, and collaboration,”20 because they collect or generate data or create complex models for analyzing and presenting data. Researchers of many stripes are now data scientists in both name and function. Given its technological depth and breadth, the data lifecycle increasingly needs and gets a lot of higher status—that is, research-based—technological and financial support. Research data management now has its own acronym: RDM.21 Commercial interests may occupy critical positions in the data lifecycle. A key source of bioinformatics research data in medicine and public health comes from clinical medical practice. Universities with medical schools and allied medical providers as partners may rely on those institutions as sources of research data. Providers’ and insurers’ interests in expense and revenue streams, and in commercial opportunities relevant to clinical practice, may collide with researchers’ and scholars’ interests in that data. Universities may structure technology transfer practice in ways that prioritize claimed proprietary interests in the data over research interests in sharing the data.22 In sum, data have now fully lost their conventional resonance as so-called “stuff that scholars interpret” and have acquired a new, broad, and distinctively modern technological resonance as both resources for interpretation and resources that result from interpretation, and in which a great deal of specialized time and equipment are invested. In a sense, data remain research tools. But as research tools they are critical knowledge resources throughout the research enterprise, not only in laboratories, and they are important research products as 18

See Open Science by Design, supra note 2, at 78–100. Caroline S. Wagner, The New Invisible College: Science for Development 9 (2009). 20 See The Fourth Paradigm: Data-Intensive Scientific Discovery (Tony Hey, Stewart Tansley, & Kristin Tolle eds., 2009); Borgman, supra note 7, at 31. 21 See, e.g., Research Data Management, onlIne CoMputer lIbrary Center (“OCLC”), available at https://www.oclc.org/research/themes/research-collections/rdm.html (last visited Mar. 28, 2019). 22 See Jacob S. Sherkow & Jorge L. Contreras, IP, Surrogate Licensing, and Precision Medicine, 7(2) Ip theory 1 (2018). 19

Data governance and the emerging university 371 well as research infrastructures. Data, datasets, and related computer software and hardware are no longer the special province of a handful of researchers. They are the keys to the entire university. Its research data resources dwarf its potentially patentable resources, in technical scale and social significance.23 B.

Data and the Law: Conflicts with IP Premises

Lawyers, legal scholars, administrators, and policymakers focus on law, regulation, and governance of data and datasets. Governance, as that term is used in this Chapter, means opportunities for and limitations on data collection, production, storage, and use by the university and its researchers, including both formal rulesets imposed from the outside (the state, funders) and the inside (the rules and policies of the university), plus informal norms and expectations emerging from various sources (disciplinary expectations and practices both inside and outside the university). Data governance, which might be regarded as a species of knowledge governance or information governance,24 embraces this multiplicity of interests, as a complex, changing system that answers the questions: who uses and should use data, and how? One critical point is how data governance is colored heavily by IP concepts, in general and specifically within the university. Concretely, that means patent law and copyright law. The IP origin story matters here. In society and commerce as a whole, that is, from perspectives external to university and scholarly practice, many lawyers, policymakers, and commercial interests have long advanced the conceptual argument that “creative” and “innovative” intellectual production, or modalities of knowledge defined over time by relevant bodies of law, are reducible to conceptual things: copyrightable works and patentable inventions. In the language of the law, those things are protected from improper appropriation and are tradeable in markets because of their expected economic value. IP law “thing-ifies” or codifies intellectual production precisely in order to stimulate the production of more, different, or better intellectual “things.”25 Sometimes that “thing-ification” instinct leads to good results in both large and small senses, and sometimes it leads to poor ones. How do those IP concepts, and the thing-ification instinct, intersect with data and datasets? The logic is straightforward. Generating or collecting data is a mode of intellectual production, because data is either a mode of information production or a step in knowledge production, or both. According to one typical line of reasoning, data are or should be “thing-ified” in the eyes of the law, from an IP standpoint, if and to the extent that “thing-ification” would help society achieve what it wants: more data (intellectual production), or new data, better data, or more valuable data.26 This would give intellectual “things” status akin to the property law protection usually given to manufactured objects. A different line would deny legal “thing” status to data as such, and would assign data to an intellectual public domain, if doing so would promote the

23 See Borgman, supra note 7, at 125–202 (describing data-related research practices in the social sciences and humanities). 24 See Michael J. Madison, Reconstructing the Software License, 35 loy. u. ChI. l.J. 275 (2003). 25 See Dan L. Burk, The Role of Patent Law in Knowledge Codification, 23 berkeley teCh. l.J. 1009 (2008); Michael J. Madison, Law as Design: Objects, Concepts, and Digital Things, 56 Case w. res. l. rev. 381 (2005). 26 Within information science, discussion of information-as-thing has had a different character. See Michael Buckland, Information as Thing, 42 J. aM. soC. Info. sCI. 351 (1991).

372 Research handbook on intellectual property and technology transfer production of more or better inventive or creative things based on that data. In an important sense, this would treat data as found, not made. In most respects, IP law favors treating data and facts as un-owned, open, and freely shareable.27 Property rights in data and datasets as such strike many people, particularly scientific researchers, as anathema to the belief that truths about the world are, as United States Supreme Court Justice Louis Brandeis once wrote, “free as the air to common use.”28 Normatively, facts and data about the world are too useful and meaningful to too many people and for too many purposes, and their ontology tends too much toward “basic” and “found” rather than “applied” or “made,” for proprietary claims to facts and data to be justified. IP law takes that reasoning to more specific levels, holding to the premise that data both are and should be presumptively “open” but also offering some nuance and some basis for narrow claims that IP law might “protect” data under some circumstances. Data may appear to be “things” in form, but their thing-ness typically is linked to legal openness. Data are often regarded in law as facts about the world, including natural phenomena in the world. So long as that is the case, then data are not subject to IP “protection.” In copyright, data and facts lack the creativity needed to justify legal coverage.29 If, however, facts are interpreted or “created” in a copyright sense, because they are assembled or collected in a “creative” way (even a modestly creative way), then copyright may apply.30 Claims of proprietary patent right relative to data and datasets are likely to be rejected on one or more statutory grounds. Datasets do not constitute “inventions” under Section 101 of the US Patent Act because they are mere “abstract ideas,” or they may not be “useful” in the sense that the law requires. If, however, a data structure is “inventive,” particularly in the context of computer programs, they may be patented.31 In genomics, datasets may consist wholly or partly of genetic information. Data consisting of genetic sequences as such are usually considered to be unpatentable “products of nature,” but synthesized genomic products may be patented.32 Framing data as open and unprotected in the IP context relies heavily on a conventional sense of facts as basic “raw material,” matter that is foundational to interpreted knowledge products. Both that conventional sense and its embodiment in IP practice resist the earlier characterization of data as “flow” rather than “form” or “thing.” That is one sense in which IP law is a poor match for data governance, even if IP law shares a “data openness” sensibility. There is a second sense in which IP law handles data governance poorly, in this case from a “proprietary data” sensibility. The conventional understanding of data as unownable “raw material” runs contrary to the idea, now common in modern scholarship, that facts may be

27 See Pamela Samuelson, Mapping the Digital Public Domain: Threats and Opportunities, 66 l. & ConteMp. probs. 147 (2003); Jerome H. Reichman & Pamela Samuelson, Intellectual Property Rights in Data?, 50 vand. l. rev. 52 (1997). 28 Int’l News Serv. v. AP, 248 U.S. 215, 250 (1918) (Brandeis, J., dissenting); see also Yochai Benkler, Free as the Air to Common Use: First Amendment Constraints on Enclosure of the Public Domain, 74 nyu l. rev. 354 (1999). 29 Feist Publ’ns, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340 (1990). 30 See Michael W. Carroll, Sharing Research Data and IP Law: A Primer, plos bIology 13(8): e1002235 (2015); Justin Hughes, Size Matters (Or Should) in Copyright Law, 74 fordhaM l. rev. 575 (2005). 31 Enfish, LLC v. Microsoft Corp., 822 F.3d 1327 (Fed. Cir. 2016). 32 Ass’n for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576 (2013).

Data governance and the emerging university 373 interpreted knowledge products in their own right,33 and to the instinct in the industrial sector that datasets may constitute commercially valuable, ownable products. Some legal actors have been encouraged to figure out a better match for data governance than IP law, without lasting success in the structures of the law or in practice. First, data may be gathered and bundled in ways that make datasets amenable to treatment as assets that are subject to contract law, offering some form of legal protection against unauthorized data access or data use even to commodified collections of data as legal “things”—even while these “things” are excluded from IP coverage as a legal matter. That practice is increasingly accepted in industry and finance and in commercial domains, even if (and sometimes because) contract law offers a pragmatic yet fragile end-run around the public policy that dictates no IP coverage for data.34 Second, in the context of biological product and drug development, statutory “data exclusivity” may attach to the results of testing new products independent of the patent status of the products themselves.35 Third, in Europe, in 1996 the European Union adopted a Continent-wide Directive on protection of databases that was aimed at preventing “misappropriation” of data collections produced with meaningful investments of time, money, and/or expertise.36 That instrument has had little practical effect. The Directive now faces possible revision.37 Despite this patchwork of formal governance instruments, in US law and elsewhere, the relative ineffectiveness of these strategies belies the fact that they reflect an instinct that is often common both to organizational life and to personal experience in the world, that data may be a thing. Not an unowned, open, “raw” thing as described above, but rather an asset—a manufactured thing, an object of possible economic, social, and/or cultural value, and a thing that is ripe for enclosure. That instinct co-exists with the instinct common in the research worlds that data is and ought to be regarded as part of a lifecycle, or flow. The next Part situates these sensibilities and conflicts about data and data governance in related conflicts in the university.

III.

DATA AND THE CHANGING UNIVERSITY

Contemporary university-based technology transfer institutions, grounded in research funding practices and IP regimes, encode a long-standing distinction in the university: between useful knowledge, which is controlled via the patentable system, and basic knowledge, which is disseminated widely via the copyright system, as scholarship. In that framing, data may constitute

33 See Mary Poovey, History of the Modern Fact: Problems and Knowledge in the Sources of Wealth and Society (1998). 34 Stacy-Ann Elvy, Commodifying Consumer Data in the Era of the Internet of Things, 59 b.C. l. rev. 423 (2018); Stacy-Ann Elvy, Paying for Privacy and the Personal Data Economy, 117 ColuM. l. rev. 1369 (2017). 35 See Erika Lietzan, The Myths of Data Exclusivity, 20 lewIs & Clark l. rev. 91 (2016). 36 Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the Legal Protection of Databases, 1996 O.J. (L 77) 20. 37 See European Commission, Staff Working Document and Executive summary on the Evaluation of the Directive 96/9/EC on the Legal Protection of Databases, Apr. 25, 2018, available at https://ec .europa.eu/digital-single-market/en/news/staff-working-document-and-executive-summary-evaluation -directive-969ec-legal-protection (last visited Oct. 17, 2019).

374 Research handbook on intellectual property and technology transfer both useful knowledge and basic knowledge. Yet data resist characterization by analogy to copyrightable scholarship or patentable inventions. Data governance requires a new framework, one that extends current conceptions of research in the university. This Part introduces data and data governance as modes of knowledge production in the research university. In light of a foundational framing of the purposes of the university through history, it explores the inadequacy of the university’s current parallel governance systems of open knowledge and closed or proprietary knowledge as applied to data. It considers the social and institutional risks of applying the current dual institutional structure too quickly to data governance questions. It suggests that the way forward includes reconsidering either what knowledge means, in the university setting, or what the university means, with respect to data and knowledge, or both. A.

An Abbreviated History of the University

Long prior to and independent of concerns about data and datasets, the university has been characterized primarily by a commitment to knowledge, driven not by partisanship, bias, utility, or financial reward but instead by the desire for general understanding, and perhaps, by an interest in improving both human and natural conditions. In a strong sense, the university is still defined by the production of knowledge and by its dissemination throughout society.38 But that definition is a puzzle, because knowledge production in the modern research university has a dual character. There is both research and scholarship, on the one hand, and also practices of invention and technology transfer, on the other hand. The modern university does not resolve that puzzle. Instead, it institutionalizes the duality via parallel practices of knowledge governance. Basic research, knowledge, and scholarship are meant to be driven by an ethos of curiosity, creativity, and experimentation, and their results are meant to be open and widely shared so that additional researchers can build on them in the same spirit. Applied research, knowledge, and invention are expected to be motivated by interests in solving specific technical or social problems, and their results are meant to be distributed via propertization and transfer from the university into market institutions. The institutional duality may be merged into an overarching commons governance framework, where commons refers to plural institutional practices of knowledge sharing.39 That duality blends with the university’s dual institutional identity as a stable, changing entity and as a dynamic, adaptable one. In the 21st century, it is both imperative, in a sense, and inappropriate, in a different sense, to speak of “the university” as a singular, integrated institution. Imperative in the sense that “the university” retains both rhetorical and functional significance as a conceptual category. The university as an institution continues to occupy a distinctive place in society as the site of meaningful and deep research-based investigation into the world. The university offers important and powerful institutional complexity, richness, and autonomy for itself and for its researchers, teachers, and students. The university remains an institution whose core purpose is the production and distribution of knowledge,40 even while its economic scale often links those purposes to many others—economic devel-

38 See Alfonso Borrero Cabal, The University as an Institution Today: Topics for Reflection, unesCo publ’g and Int’l dev. researCh Ctr. (1993). 39 See Madison, et al., supra note 5. 40 See, e.g., Jonathan R. Cole, Toward a More Perfect University (2016).

Data governance and the emerging university 375 opment, employment, athletics and entertainment, and more. The thousand-year history of the university suggests a continuity of tradition and practice in all of these senses, even if the antecedents of the modern research university are intellectually and culturally diverse. That description of the university as a singular institution with a continuous history is likewise inappropriate. The institutional character of the university evolved over many centuries. It was not given. The evolution continues. The research university, as we know it today, evolved relatively recently. The idea that an institution might be designed to produce new knowledge, in addition to documenting and sharing existing knowledge, had to wait for the moment that enabled the development of modern science as a research-based enterprise. In its pre-Enlightenment forms, the university denoted an institutional arrangement for re-producing knowledge, embodied in the body of the university’s faculty and in its graduates. Post-Enlightenment, with the emergence of German research-based universities in the early 1800s and the adoption of German forms in the US toward the latter part of the 19th century, the modern research university came into focus. In this brief re-telling of that continuous yet contingent history, the Chapter focuses on three themes. One is the centuries-old conceptual framing of the knowledge now embedded in the university environment: basic or pure knowledge as distinct from applied or useful knowledge. Two is the changed institutional implementation of that dualistic framing in modern universities. Three is the contemporary governance modality that now characterizes and advances that dualism. B.

Knowledge Through the Centuries

Universities produce knowledge. But what sorts of knowledge, and why? From their beginnings in Europe in the 12th and 13th centuries, until the epistemological revolution wrought by the Enlightenment and practiced by members of the so-called “Republic of Letters” through the mid-18th century,41 universities and university researchers emphasized the study and sharing of pure, basic knowledge—the concept being traceable to its ancient Greek roots. University knowledge brought forward ancient interests in epistêmê, referring to timeless, unchanging, truths about both physical and spiritual worlds, rather than technê, referring to craft and practice. Enlightenment epistemology redefined that duality, characterizing basic knowledge as grounded in empiricism rather than reflection.42 German universities were the first to implement that shift institutionally, organizing and re-organizing themselves around scientific research grounded in observations of the world and around open sharing of research results, principles that are associated with the writing of Francis Bacon. The University of Berlin, established in the early 19th century on the initiative of Wilhelm von Humboldt, is most famously associated with establishing the foundations of the modern university as a home for research, and for the systematic investigation and production of fundamental knowledge. Wissenschaft is the well-known German term for this mode of research and institutional design, aligning the ancient interest in basic truths with modern understanding that truth is the product of investigation, not reflection. As that pairing expanded beyond Germany, Cardinal Newman captured it in his classic The Idea of a University in 1852. He described a proper 41 42

Joel Mokyr, A Culture of Growth: The Origins of the Modern Economy (2018). The historical review in this section is based on Stokes, supra note 6.

376 Research handbook on intellectual property and technology transfer university as a place of teaching universal knowledge, a community of thinkers, seeking truth and understanding judgment via both teaching and research, that is, via broad and inclusive educational practices rather than via vocational pursuits or spiritual improvement.43 German Wissenschaft was distinguished both conceptually and institutionally from technical training, or applied science. German universities were unlike the German polytechnical institutes, the Technische Hochschulen. When researchers and higher education leaders in the US looked to Europe for research practices to borrow and build on in advancing science in the US, they looked to the former and not to the latter. In the later 19th century, US recipients of German research degrees took up academic teaching positions on their return to North America; German-style university research programs were models for new US research universities (particularly John Hopkins University, the University of Chicago, and Stanford University), for new PhD programs at existing US colleges, such as Harvard and Yale, and for the expansive visions adopted by certain universities that benefitted from Morrill Act “land grant” status (particularly the University of California, at Berkeley, and the University of Wisconsin). Basic knowledge based on research, rather than useful or applied knowledge, was the priority and the domain of authentic modern research universities. C.

Knowledge in the Modern University

That conceptual and organizational principle carried over throughout the 20th century, but institutional practice—bureaucratic, professional, and financial—rendered the distinction between basic knowledge and applied knowledge far blurrier on the ground. Professional schools expanded alongside PhD programs. Harvard University created the first modern law school in 1870. The University of Pennsylvania opened the Wharton School, for management training, in 1881. Stanford University launched its School of Engineering in 1925. Yale University founded its School of Music in 1894 and its School of Art in 1869. US universities that were known primarily as engineering institutes (such as the Massachusetts Institute of Technology, founded in 1861) or as homes to agricultural and mechanical research and training (such as many “land grant” universities), expanded into research in social sciences and the humanities. In Germany, a relatively bright institutional line divided basic research and knowledge (universities) from applied research and knowledge (technical institutes). In the US, that line was drawn less distinctly, because it was drawn inside each university. Within that university, basic research and knowledge was to be open and shared by scholars; applied and useful research was to be put to use in preparing the professions and supporting industry. D.

Contemporary Knowledge Governance in the University

Bringing that history up to the present means highlighting two further critical developments, each of which both relied on and advanced the distinctions just traced back through time.

43

John Henry Newman, The Idea of a University Defined and Illustrated (1852); the centrality and durability of the vision of the university as a home for basic knowledge are illustrated by the fact that both themes are integral to Jaroslav Pelikan’s updating of Newman 140 years later. See Jaroslav Pelikan, The Idea of the University: A Reexamination (1992) (updating and examining John Henry Newman, The Idea of a University Defined and Illustrated (1852)); Pelikan, supra note 4.

Data governance and the emerging university 377 The first development was production toward the end of World War II of the report titled Science—The Endless Frontier by Vannevar Bush, Director of the Office of Scientific Research and Development.44 The report proposed a massive scaling-up of federal sponsorship of basic research to be conducted in the country’s universities and research institutes, de-linked from military purposes, and allowed to advance almost exclusively under the supervision of autonomous research experts themselves. Bush’s suggestions for institutionalizing this vision in the federal bureaucracy, particularly the idea that the research would be conducted in a spirit of free and unfettered inquiry, almost entirely free of government oversight, was less successful than its conceptual framework, which has animated US government sponsorship of university research ever since. Bush clearly and unambiguously distinguished basic or pure research and general knowledge, the domain of scientific researchers to be supported by federal funding, from applied science and useful knowledge, including its commercial applications, which flowed from basic research in a kind of linear progression and which was the domain of industry.45 He argued that a well-supported and well-structured domain of basic knowledge production was essential to American health, prosperity, and security, particularly as the Cold War dawned, specifically because (he argued) basic research led to applied knowledge and technology. The US grant-making apparatus that eventually emerged after World War II, distributed across the National Science Foundation, the National Institutes of Health, the Department of Defense, and other agencies, relies heavily on Bush’s distinction. Basic knowledge was no longer a good in itself (Wissenschaft), but instead an input into a critical pathway toward useful knowledge. The second development was enactment of the federal Bayh-Dole legislation by the US Congress, in 1980.46 As a matter of formal law and institutional practice, the Bayh-Dole Act standardized US policy regarding ownership of patentable inventions produced by recipients of federal research sponsorship. Previously, US policy was not standardized; different federal agencies used different approaches, leading to concern—in an era of generalized economic anxiety about American economic competitiveness relative to reindustrializing Japan—that federal support for American scientists was not producing significantly commercially useful results. The policy solution was to permit universities and university researchers to retain patent rights to their inventions, even if the research was conducted with federal support. The Bayh-Dole Act and successor legislation almost single-handedly gave rise to modern university-based technology transfer practice, as universities came to rely on their patent interests to obtain patents, build portfolios of licensed technology, and try to benefit from equity and income from spin-off companies. Technology transfer offices in research universities operate today as boundary worlds, translating ideologically and practically between the applied research practices of the university and the product development and financing practices of the commercial market.47 44

Vannevar Bush, Science—The Endless Frontier (1945) [hereinafter, Bush, Science]. Bush was an engineer, former Dean of MIT’s School of Engineering, and president of the Carnegie Institution for Science. He is also famed for his authorship of a visionary essay on communications technology. See Vannevar Bush, We May Think, in The Atlantic (July 1945). 45 Bush, Science, supra note 44, at 75. 46 Patent and Trademark Law Amendments Act, 96 P.L. 517, 94 Stat. 3015 (Dec. 12, 1980). 47 Brett M. Frischmann, Michael J. Madison, & Katherine J. Strandburg, “Governing Knowledge Commons” in Governing Knowledge Commons 1, 26 (Frischmann, Madison, & Strandburg eds., 2014) (citing Patrick L. Jones & Katherine J. Strandburg, Technology Transfer and an Information View

378 Research handbook on intellectual property and technology transfer The Bayh-Dole Act and modern technology transfer reified the long-standing divide described earlier between basic knowledge and applied or useful knowledge, assumed as a premise that the importance of basic knowledge lies in its capacity for producing useful knowledge, and baked both arguments into the structure of modern IP law. Basic research and knowledge were to be financed largely by the government as forms of what economists call public goods; applied and useful research were to be distributed through society as patented private goods, in the marketplace. Virtually all US research universities now pair their technology transfer organizations with university-wide governance policies that distinguish between copyright law and patent law. Details vary from university to university, but in broad terms, scholarship is the domain of copyright law, and the normative structure of copyright law encourages openness and broad distribution of scholarly knowledge. Invention is the domain of patent law and proprietary markets. In practice, that means that scholarly works, such as journal articles and monographs, are governed by copyright but are typically remitted by university governance frameworks to the control of their researcher authors. Governance is largely normative. Researcher scholars are expected to disseminate their research results as widely as possible, carrying on the long-standing “Republic of Science” tradition of open sharing. In practice, many if not all scholarly publishers are for-profit enterprises that demand that researchers turn over their copyrights as a condition of publication, giving rise to substantial conflict about proprietary systems and Open Access alternatives. But the normative premise is clear: basic knowledge should be open knowledge. Useful knowledge, known in modern culture as technology or inventions, may be located within research results or may derive from them. In either case, useful knowledge is governed by patent. Within the university, governance frameworks for inventions are largely dictated by the requirements of the Bayh-Dole Act, which means that federally-supported inventions must be disclosed by researchers to the university so that the university may choose whether or not to claim ownership of them via the patenting process. Governance is highly bureaucratic. Bayh-Dole, technology transfer, and patent law combine to pull researchers toward applied research.48 Researcher inventors are expected to (and the more entrepreneurial researchers often volunteer to) limit dissemination of their work product other than through the closed mechanisms of the market. The normative premise is uncontroversial: useful knowledge may be closed knowledge. The foregoing summary is necessarily crude, and readers will see that it is more effective as a description of intellectual history than as an accurate account of scientific practice. Scientific research specifically, and empirical investigation of the world generally, often do not assert or build on clear distinctions between basic and applied knowledge, no matter what the Greeks believed, or on a linear relationship between the two.49 But ideas matter, and these ideas— contrasts between basic knowledge in open institutions, and applied knowledge in controlled institutions—have important implications. Two of these are explanatory, having to do with institutional diversity and with contemporary critiques of the university. The third brings the discussion back to data, datasets, and data governance.

of Universities: A Conceptual Framework for Academic Freedom, IP, Technology Transfer and the University Mission (unpublished manuscript) (Feb. 22, 2010)). 48 See Brett M. Frischmann, The Pull of Patents, 77 fordhaM l. rev. 2143 (2009). 49 Thomas S. Kuhn, The Structure of Scientific Revolutions (3d ed. 1996 (1962)).

Data governance and the emerging university 379 One implication is that institutional pluralism in modern universities owes a lot to assumptions and practices built on distinctions between basic and applied or useful knowledge, open and proprietary research results, and applications of copyright law versus patent law. Distinctions between “public” and “private” US universities today are often less significant, in terms of their knowledge functions, than their classification as “research-intensive” or other. Similarly, “elite” and “non-elite” designations are often trumped by the scale of a university’s research program and research funding. In turn, differences among universities on those dimensions explain a great deal of the practices of specific universities with respect to investment in research infrastructures and the scope and types of field-specific research practices, needs, and goals. Outside the US, the role of this distinction in institutional pluralism is even clearer, because scientific research has continued to rely on funding practices and institutional arrangements that are more closely linked to national research programs and scientific institutes and less directly linked to the university. In short, the knowledge production and distribution functions of each university are embedded in specific, local institutional cultures and programs. No two universities are precisely alike. But the synthesis of humanities researchers and scientific researchers in specific universities, and how the university provisions them, within the university’s umbrella of unitary institutions, is explained largely by the character of the university’s commitment to independent knowledge production. As to governance, the university makes possible the creation of disciplinary communities and fields, sometimes referred to as communities of practice.50 Self-governance of the university at the local level is often one of its most critical defining features. What counts as knowledge and scholarship within a given discipline is always changing, at least modestly, but is fundamentally a question of expert determination within the discipline. As to provisions, functionally, within its subsidiary schools, colleges, departments, and faculties, and across those units from institution to institution, the university supplies various material infrastructures for documenting, distributing, and collecting knowledge products across all domains, primarily theses, papers, books, and the contents of lectures and tutorials. Capital-intensive research programs depend on expensive devices, laboratories, and related research environments, much of which is funded by blends of internal and external support. University libraries, university presses, and university-based journals support scholarship across multiple fields. In total, as convener and coordinator, the university offers useful and efficient organizational and physical means for scholars of similar stripes to gather in a single place and for teachers to gather with students. Gathering is equal part gatekeeping. The distinctiveness of the university depends, often, on how it imposes admission and membership requirements on both researchers and students. The second implication is that the university is apt to be called to account by critics either for its failure to adhere appropriately to its ancient anchoring in pure knowledge, or for its failure to abide by the basic/applied knowledge binary described earlier, or both. The university has come to be valuable socially, culturally, economically, and even politically in all sorts of complex ways. That value, or values, entail delicate contemporary blends of openness and public benefit, on the one hand, and privatization and enclosure, on the other hand, and it compromises historic blends. The first broad, clear modern effort to call attention to conflicts 50

Paul Duguid, “Community of Practice Then and Now” in Organizing for the Creative Economy: Community, Practice, and Capitalism (Ash Amin & Joanne Roberts eds., 2008); Andrew Abbott, Department and Discipline: Chicago Sociology at One Hundred (1999).

380 Research handbook on intellectual property and technology transfer between the university’s culturally public character and demands to privatize it in different respects came from Clark Kerr, Chancellor of the University of California Berkeley and later President of the University of California. He argued that the 20th-century US university had been charged with so many different missions—economic development for the community, entertainment and social advancement and jobs for graduates, symbolic capital for alumni (often related to athletics)—that he came to reject the “university” label and semi-seriously suggested substituting the word “multiversity.”51 More recent critiques charge the university with “academic capitalism,”52 because of the university’s current alignment with the expectations of private sector partners and funders. Technology transfer practices and the demands and expectations of IP law in the modern economy are often at the core of the complaint; the university is accused of having abandoned its historical character as a home for the production of pure (or mostly pure) knowledge and having sold out to market merchants.53 Related critiques look less to engagement between the university and commercial interests in the marketplace and more to entanglement between the university and the state. Here, the charge is often not that the university’s basic knowledge character has been polluted by commerce but that the university has gotten entangled in the blurring of boundaries between the market and the state. Critics charge the university with both reflecting and advancing creeping neoliberalism, as both private and public funders monitor and assess the university and its citizens and subordinate poorly-defined public goals (“knowledge”) to the dictates of the market.54 Once a semi-autonomous institution that was largely culturally exempt from regulatory oversight associated with private activity, the university is turned into a legible and regulable environment,55 often to promote or protect important individual or collective interests (such as access for students), but at the cost of eroding some of the university’s distinctiveness. The critics, in short, do not reject the complex knowledge-based foundations of the university or the dual knowledge governance frameworks that have been built on those foundations. Instead, both university celebrants and skeptics largely reinforce and amplify what has come before. E.

How Data Illuminates Changes and Conflicts About the Mission of the University

That judgment raises the stakes of the third and most important implication of the knowledge-based model of the modern research university. It illuminates the challenges associated with data, datasets, and data governance. As noted earlier, virtually all research universities today have technology transfer practices and formal policy instruments that

51

Clark Kerr, The Uses of the University (5th ed., 2005 (1963)). See Sheila Slaughter & Gary Rhoades, Academic Capitalism and the New Economy: Markets, State, and Higher Education (2009). 53 For a thorough account of how the cultural and practical imperatives associated with IP law have compromised the traditional cultural identity of the university, see Jacob H. Rooksby, The Branding of the American Mind: How Universities Capture, Manage, and Monetize IP and Why It Matters 178 (2016). 54 See Henry A. Giroux, Neoliberalism’s War on Higher Education (2014). 55 See James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (1999). 52

Data governance and the emerging university 381 address allocation of IP interests in research products and scholarly works across the institution. Technology transfer practices manage the boundary between presumptively open, shared knowledge practices in the scholarly setting and presumptively controlled, proprietary knowledge production (i.e., patent-based practices) in the industrial or commercial setting.56 Inside the university, this is a fragile peace but a mostly workable one. Most of those formal university policies and governance declarations, especially in the US, do not address interests in data or datasets. Where and how might data fit in university knowledge governance? One way to look at the question is to start with the long-standing university commitments to research in basic knowledge. Those practices are closely aligned with normative and pragmatic policies promoting openness and sharing of research results. So long as data gathering, production, and analysis are fairly characterized as basic research, leading to generalized knowledge and shareable results, then historical norms align relatively easily with modern expectations. Designing a corresponding governance instrument ought to be relatively straightforward. But there are other considerations. Data and datasets are not necessarily so readily and uniformly characterized as modes of basic research and knowledge. They may not be knowledge inputs or knowledge products; they may be knowledge tools and knowledge flows. Modern universities have existing formal and normative knowledge governance frameworks, but those are largely directed to knowledge products. University technology transfer practices and IP policies appear to define the full scope of the relevant governance landscape. As knowledge products, data and datasets might be squeezed into those policies and practices, or closed analogous policies and practice might be developed. So, to the extent that data and datasets have technological utility or even some commercial value, then in some respects they might be candidates for governance treatment by analogy to governance of patentable inventions—even if datasets are not patentable themselves. Or, independent of their scholarly character, data and datasets might resemble intellectual and scholarly works of the sort that are typically governed by copyright. To the extent that contemporary knowledge governance expresses an “open v. closed” framing, that outcome might be normatively preferable. Alignment between datasets in the university and university copyright policies for scholarly works might advance the interests of those who advocate for greater openness and sharing of university-based knowledge. But that outcome would come at a potential cost, because it would treat data resources for governance purposes as taking fixed forms that may conflict with how data are generated and used in “flow” or “lifecycle” practices and processes, as described earlier. Attaching copyright interests to things to which they ordinarily do not attach, even in the interests of openness, raises both conceptual and practical concerns with respect to the purposes of IP law and broad questions of social, cultural, and economic benefit.57 In sum, one can analyze questions about data governance in the university in a relatively straightforward way by relying on the well-established conceptual framework that institutionalizes practices regarding research products in two overlapping and usually consistent ways:

56

See Katherine J. Strandburg, Curiosity-Driven Research and University Technology Transfer, in University Entrepreneurship and Technology Transfer: Process, Design, and IP 93 (Gary D. Libecap, ed., 2005). 57 See Niva Elkin-Koren, What Contracts Can’t Do: The Limits of Private Ordering in Facilitating a Creative Commons, 74 fordhaM l. rev. 375 (2005).

382 Research handbook on intellectual property and technology transfer along a spectrum that runs from basic, general knowledge to applied, useful knowledge, and along a spectrum that runs from copyright and openness to patent and closure. Those two spectra largely define the institutional character of the modern research university. Using that approach, questions of data governance align poorly with the standard spectra. Data are not obviously or always basic knowledge or applied or useful knowledge. In many respects normative considerations suggest that data should be open and shared, but the typical governance hook for that approach is unsatisfying at best and risky at worst. The Chapter offers the hypothesis that by questioning data governance, we open the door to the possibility that the centuries-old knowledge-based framing of the university has been exhausted, and the university’s knowledge governance framework grounded in technology transfer and IP law has reached the limits of its utility. Data and datasets are neither useful things nor scholarly things; in many respects data are not things themselves. But neither are data and datasets completely not things. The university needs to begin again, in part taking data as a central part of its knowledge mission and in part re-thinking that mission itself. It needs to address data governance questions from the ground up, in institutional context.58 What is the emerging data-intensive university for?

IV.

DATA GOVERNANCE AND THE EMERGING UNIVERSITY, RE-FRAMED

The future of the university is a broad, conceptual topic. The design of data governance is a concrete, institution-specific problem. This Part bridges the two, by specifying a series of domains where they can be assessed concurrently and progress made toward possible solutions simultaneously, without necessarily anchoring the analysis either in attachment to forms of knowledge as such (a legacy conceptual framework) or in technology transfer and IP institutions (a legacy institutional framework). The neologism “data-intensive university” is itself a speculative response to the hypothesis that the centuries-old knowledge-based framing of the university has been exhausted, and the university’s knowledge governance framework grounded in technology transfer and IP law has reached the limits of its utility.59 Fleshing out that argument, the Chapter argues that what may unify the emerging university is data itself, practices and programs of collecting, documenting, understanding, using, and caring for evidence, across all fields, domains, and disciplines of inquiry, both existing and new, including arts and culture, science and technology, social sciences and the professions. “Understanding the world and stewarding and improving its condition” would take the place of “universal and then useful knowledge” as the governing paradigm for institutional practice. This is meant to be a generous and broadly inclusive definition. “Data” are not products or processes only in natural, physical, or biological sciences. Data, as evidence, matter in all fields of research. In some, as noted earlier, the rhetoric of “data” may be displaced by processes

58

See Henry Plotkin, Darwin Machines and the Nature of Knowledge (1997). The neologism is not entirely novel, though other recent uses of the phrase appear to be motivated by interests related to those that animate this Chapter. See David M. Berry, The Data-Intensive University, stunlaw (Sept. 14, 2018), available at http://stunlaw.blogspot.com/2018/09/the-data -intensive-university.html (last visited Oct. 17, 2019). 59

Data governance and the emerging university 383 of embedding practices in different vocabularies and syntax. What counts as evidence in any particular domain may be contested and may be judged in various ways. Moving from “knowledge” as a central narrative of the university to “evidence” as a central narrative is therefore largely a rhetorical adjustment, rather than one motivated by a strong interest in including new programs in universities or excluding existing ones. But the rhetorical advantage would be this, in institutional terms: the gradual demise of the priority long given to distinctions between basic knowledge and applied knowledge and corresponding implementations in knowledge governance practices that embed choice between “open” and “closed.” Evidence, meaning data and datasets, would be subsumed within a university governance framework that blends interests in data forms and data flows in institutionally- and contextually-appropriate norms governing data sharing and data exclusivity. On the ground, data governance policies and practices would be anchored in a fundamental commitment to evidence, rather than anchored in the practicalities of a fundamental commitments to one sort of knowledge or another. Is this possible? Is it desirable? What difference(s) might it make? Both as process and as result, consider the transition via brief reviews of three intersecting perspectives, each of which illustrates both opportunities and barriers to be expected. The first confronts IP itself as a knowledge governance paradigm. The second is a direct exploration of normative claims regarding openness and data. The third examines the purpose of the university as a product of stakeholder interests. A.

The University as an IP Institution

As noted earlier, knowledge governance in the university is most often expressed institutionally in terms of the two default regimes of IP law: copyright and patent. Research universities typically have adopted formal policy instruments that address allocation of IP interests in research products and scholarly works across the institution. The point to be elaborated here is how that governance framing tends either to exclude data and datasets, making the dominant IP frameworks either significantly under inclusive with respect to a huge and critically important knowledge resource, or misleading, if those frameworks are deployed in ways that might be stretched to include data. As to patent law, patentable inventions created by US university researchers are ordinarily expected to be assigned to the university for consideration for patenting and eventually transfer into commercially marketable technologies and related companies. Under the Bayh-Dole Act and related federal regulations, that expectation is reinforced by legal obligations to disclose and assign interests in inventions with respect to work supported by federal research sponsorship.60 But universities may and sometimes do impose disclosure and assignment obligations on researchers (including students) as a matter of university policy, exceeding what is required by law. Universities have sustained their investments in technology transfer practices in part 60 In 2011, the US Supreme Court held that inventions produced by university researchers with federal research funding were not assigned by law to the university. Rather, the university only acquired that title via an express, signed assignment from the inventor. See Bd. of Trs. of the Leland Stanford Junior U. v. Roche Molecular Sys., 563 U.S. 776 (2011). That decision led to a scramble at TTOs nationwide to revise their policies on invention disclosure and assignment by researchers and to ensure that relevant researchers had assignment agreements in place. In 2018, the federal Code of Federal Regulations was revised to require that recipients of federal research funding have assignment obligations in place. 37 C.F.R. § 401.14(f)(2) (2018).

384 Research handbook on intellectual property and technology transfer based on an ideological commitment to the idea of the university as a participant in private markets for technology innovation.61 As to copyright law, copyrighted works created by university employees, even scholarly works produced with US federal support, are not required or expected by law to be transferred to the university for possible commercial exploitation. (Computer software, or code, represents an unclear middle ground, because both copyright interests and patent interests may apply.) Under US law, faculty-created copyrighted works might be considered to be university-owned “works made for hire.” Still, many universities have adopted policies and practices that formally or informally waive university copyright claims, grant copyrights in faculty-generated works to their authors, or simply announce that faculty authors are copyright owners with respect to their works, and that the university wants little further to do with them. Some universities go farther, disclaiming copyright interests not only in faculty-generated scholarship but also in much faculty-generated teaching materials. The results are far from perfectly clear or consistent,62 but the overall tenor is to distance the university from copyright in scholarship and to promote open inquiry and open sharing of research results, apart from the commercial market, just as the tenor in patent law is to bring the university into close proximity with it. The earlier review of legal frameworks applicable to data make it clear that data have no good home in this scheme, nor should they, because they lack the creativity or inventiveness that IP law looks for. Where do data belong? In closed governance frameworks, or open ones? In this “open v. closed” setting, should data be more “like” patented inventions or more “like” copyright works? Specifically, because the logic of IP is well-established in the university, and because stakeholders invested in university-based IP governance are apt to be stakeholders in data governance as well, there may be an understandable tendency in the university to approach questions about university-based data governance by analogizing data to IP and by analogizing data governance to IP governance. “Who owns the data?” is a tempting question,63 as a convenient IP shorthand for asking whether datasets and associated tools amount to “scholarly works” or “scholarly products,” whose use and disposition lie in the hands of faculty researchers, or potentially commercializable inventions, whose use and disposition is determined according to university ownership rules and market imperatives that usually dominate technology transfer practice. There is no correct or even best answer to the question of how to apply IP frameworks in the university by analogy to data. Instead, take note of how IP law is introduced as its overarching governance logic, often presuming that governance begins by asking, “who owns the information?,” or “who owns the data?” The question that began as “open v. closed?” from a normative standpoint is revealed as a question that assumes, in both cases, a premise about unilateral ownership and control. That conclusion is the implication of both the general statement earlier that “knowledge” as an umbrella normative paradigm for the university does not advance contemporary understanding of governance of the university’s information resources, and the specific statement earlier that IP-based scholarly openness and technology transfer as institutional practices have

61 See David C. Mowery, et al., Ivory Tower and Industrial Innovation: University-Industry Technology Transfer Before and After the Bayh-Dole Act (2004). 62 See Rooksby, supra note 53, at 178–205. 63 See Barbara J. Evans, Much Ado About Data Ownership, 25 harv. J.l. & teCh. 69 (2011).

Data governance and the emerging university 385 been exhausted as a governing logic for university knowledge products, even by analogy, when questions are raised about data. Both statements take as conceptual given that knowledge is a kind of thing, a basic thing or a useful thing, to be transferred into the broader world beyond the library or the laboratory. Data are different. If data are a thing (and commercially, especially, they may be), they may be owned and controlled. Data may not be things at all. The data lifecycle and the idea of data as flow, described earlier, make data resources ill-suited to the ownership-and-control premises of university knowledge governance aligned with IP law. IP law is quite agnostic about data—even skeptical of it. Industry practice now teaches that data may be subjected to contractual regulation, as a pragmatic matter. Data is not ownable nor owned, usually, in research settings, but may be controlled, in part, in commercial settings. If data and datasets are things, they are third sorts of things, neither obviously owned nor unowned, and in many respects they are not things at all. As information resources that are fundamental to the practices of the entire research university and that are shared, as flow, in many respects, ownership and control may simply be the wrong question. B.

The Normativity of the Data-Intensive University

The better and more appropriate governance question in the research university, at least, takes data-as-evidence as comprehensive premise and asks how to advance that premise through contextually-appropriate rhetorical and institutional practice. One cannot hope to do that in the abstract. One can only hope to do it on the ground, exploring what works and is effective, and what is not. In the research university, there has emerged not only the fact of data-driven and data-intensive research but also a widely shared instinct that much data and dataset practice should be open and shared. Is it possible to institutionalize that instinct, as matters of practice and policy, while not unduly disrupting inherited practices associated with traditional university knowledge products? The question is what to do about data as an overarching principle and practice, while respecting continuing, critical investments in both scholarship and in technology transfer itself. Organizing the questions in that way implicates all sorts of institutional practices, from recruitment of students to appointment of faculty, researchers, and teaching staff; to professionalized, administrative bureaucracy; to formal policies; to cultural systems of prestige and normative systems of value. These things can be imagined; can they be done? Formal policies, one of the questions with which this Chapter began, is an area where it is being done, where institutional practice is being reshaped by efforts to express anew the norms associated with research production. Formal policies and practices may exist at the level of the university, so that they transcend particular fields or disciplines; at the sub-university level (within faculties or departments); and also at the trans-university or trans-research institution level (within disciplines, and/or among research sponsors and funders). The norms, however, are largely shared: data accessibility; data sharing; research reproducibility and research transparency; data accuracy and integrity; data security and data privacy; data retention. Those so-called “top level” norms are accompanied by a cluster of related second-order goals, including administrative sustainability for data practices, in terms of enforceability and financial and technical resource allocation; education and cultural anchoring of data-related practices in institutional norms; and compliance with relevant positive law. Taken together, two generalized, dominant normative visions emerge. “Share data widely and openly” is the first. “Secure data appropriately,” against invasion of privacy interests,

386 Research handbook on intellectual property and technology transfer misattributions of origin and analysis, and misappropriation by non-researchers, is the second. The edict “make the data accessible outside the university” is paired with the edict, “keep the data secure inside the university.” Ensure the free flow of research and scholarship, and do not interfere with effective commercialization of inventions. The historical framing of the university in terms of different forms of knowledge is recast, normatively, in terms of the multiple roles played by data. Three brief examples illustrating these clusters of concepts may be wrangled into institutional practices and policies that have concrete application. Implementing data as the normative premise of the research university may be a product of organizing from below, inductively, rather than determining an optimal strategy from above, deductively. The data-intensive university may turn out to be an evolutionary adaptation of the Baconian research paradigm. Example one is the effort documented by European researchers to advocate for the adoption of the “FAIR” guiding principles for scientific data management and stewardship. “FAIR” is an acronym; in English, it stands for making data and datasets findable, accessible, interoperable, and reusable.64 The point of the principles is that data governance strategies in particular research institutions should express those values relative to digital assets stewarded by those institutions, and they should express those values as institutional policy and as practice. FAIR information practices are slowly gaining adherents and implementation in Europe, in the UK, and in the US as a set of standards and practices adopted voluntarily. Example two is the Open Science movement, a community of researchers and allies who are building on successes of the Open Access movement for scholarly publishing to advocate for building and implementing institutional infrastructures that promote open, nonproprietary sharing of research results in repositories and elsewhere.65 The Open Science movement, like the FAIR movement, explicitly adopts a normative stance, and like the FAIR movement it directs its normative claims to institutional actors like universities and funders. A major component of Open Science advocacy is supporting institutions in building and operating repositories for datasets. Example three is the BD2K initiative at the National Institutes of Health.66 BD2K stands for Big Data to Knowledge, and it consists of a large-scale, long-term funding program administered by NIH with the express aim of developing a trans-disciplinary infrastructure for managing and making the best uses of Big Data datasets. BD2K grants encourage recipients to adopt and practice FAIR principles, illustrating different ways in which normative arguments may influence the shapes of data governance ecologies. The US experience so far contrasts with the approach taking in the United Kingdom. Researchers who receive funding provided by research councils united under the umbrella United Kingdom Research and Innovation (UKRI) are expected to provide data management and data sharing plans with all grant and fellowship applications.67 Guidance from research councils is aligned with a set of common

64

See Mark D. Wilkinson, et al., The FAIR Guiding Principles for Scientific Data Management and Stewardship, 3 Scientific Data, Article number: 160018 (2016), doi:10.1038/sdata.2016.18. 65 See Open Science by Design, supra note 2. 66 See Office of Strategic Coordination—The Common Fund, Big Data to Knowledge, nat’l Inst. of health, available at https://commonfund.nih.gov/bd2k (last visited Oct. 17, 2019). 67 Funders’ Data Plan Requirements, dIg. CuratIon Ctr., http://www.dcc.ac.uk/resources/data -management-plans/funders-requirements (last visited Oct. 17, 2019).

Data governance and the emerging university 387 principles on data sharing provided by UKRI.68 Many UK universities have voluntarily affirmatively adopted comprehensive data management strategies. In the US, no equivalent governance framework has been adopted across all government-sponsored research funders, though NIH and NSF requirements that researchers submit data management plans with their proposals, described earlier, affect broad domains of university-based research. US universities have not yet moved as a group to adopt institution-wide data management policies. US practice to date consists in part of development of such policies69 and in part of guidance that individual researchers should comply with funders’ requirements.70 To be sure, each of these efforts is advancing in bits and pieces rather than consistently at the level of the university itself, and not each of them consistently avoids or successfully addresses potentially problematic attributes of data governance in the university setting. Data governance should be generous, pluralistic, and inclusive with respect to present and future disciplinary scope and research practice; should be open and inclusive with respect to the professional status of researchers and related data science, information science, and informatics professionals; and should avoid replicating problematic bureaucracies and prestige and power hierarchies in universities and in other research institutions. Shifting from “knowledge” in the research university to “data” and “evidence” in the research university should not be an excuse to suppress or bureaucratize curiosity-driven research or innovative research or researchers. Nor should any of the foregoing be read as altering normative and institutional commitments to university students. C.

The University As the Sum of Its Stakeholders

Addressing those questions, and building upward and outward from the examples given and from others to come, entails a host of complex local questions and conversations. What is best for a particular university and its citizens? Some university policies are quite broad, brief, and largely aspirational.71 Some are more focused, or longer, or more conversational than prescriptive.72 University-based research is heterogeneous and changing, in field, discipline, and community terms. Universities vary widely in the scope of their research program, funding, research culture, and management sophistication. Some research fields are well-known to be based heavily on collecting and analyzing large datasets (biomedical research, public health,

68 Common Principles on Data Policy, uk researCh and InnovatIon, available at https://www .ukri.org/funding/information-for-award-holders/data-policy/common-principles-on-data-policy/ (last visited Oct. 17, 2019). 69 Research Data and Materials Policy, yale u, Aug. 31, 2017, available at https://provost .yale.edu/sites/default/files/files/Research%20Data%20Policy%2006-07-2018.pdf (last visited Oct. 17, 2019). 70 Penn Libraries, Data Planning and Management, u. penn., available at https://guides.library .upenn.edu/data-management (last visited Aug 3, 2018). 71 Research Data Policy, u. bath, Apr. 9, 2014, available at https://www.bath.ac.uk/corporate -information/research-data-policy/ (last visited Oct. 17, 2019). The University of Bath policy is accompanied by particularly clear and useful practical guidance, Guide to Research Data Management at the University of Bath, u. bath, http://www.bath.ac.uk/research/data/ (last visited Mar. 28, 2019). 72 Retention and Maintenance of Research Records and Data: Principles and Frequently Asked Questions (“FAQS”), harv., Apr. 12, 2017, available at https://vpr.harvard.edu/files/ovpr-test/files/ research_records_and_data_retention_and_maintenance_guidance_rev_2017.pdf (last visited Oct. 17, 2019).

388 Research handbook on intellectual property and technology transfer astronomy and astrophysics), practices that impose management and security and privacy problems that may be several orders of magnitude greater than problems associated with the smaller or even hand-curated datasets more typical of much social science research. Some researchers may work with datasets so large and complex that they require commercial storage services and are processed via supercomputers. Others may work with datasets maintained in simple spreadsheets stored on laptop computers. Some may have full-time professional data scientists to manage their data needs and to facilitate collaboration with other researchers. Some may rely on time borrowed from graduate students. Some may partner with centralized or specialized library services. Researchers in some fields may even be unaware that the material they work with, such as the contents of art collections, may be classified as data for governance purposes. Within a given institution, thoughtful governance requires inventorying the data practices of the university and understanding the university’s needs and goals relative to data and related information resources and the roles that data, data governance, and technical support for data are expected to play. What resources should be treated as data for governance purposes? Whose practices, needs, and goals matter to answering that question? It is common to observe “spillovers” or externalities of data production, storage, or use. Many spillovers are good; some are harmful. A substantial portion of relevant spillovers involve collaboration, scholarship, and social value that crosses boundaries between one university and another, and/ or between one or more universities and research institutes, and/or even between one or more universities and private enterprises. The rhetoric and practice of openness, regarding data and other information resources, is not limited to scholarship and datasets. In technology transfer practice, the phrase “Open Innovation” captures knowledge sharing practices across company research programs and between industry and the university.73 At the same time, but in distinct contexts, data sharing can be deeply problematic, even harmful, particularly if data sharing has commercial motivations or impacts, or if it implicates government surveillance and security. What is good for and by university researchers may not be good for or by Facebook or other 21st-century information platforms.74 Thoughtful governance also requires policies and directives that speak intelligibly to multiple stakeholders and that are capable of implementation in multiple places and via multiple methods. Within the university, answers may be expressed centrally, in a formal policy or in one or more statements to the effect that compliance with relevant mandates is expected, or locally, in the expectation that schools, departments, laboratories, other units, and individual researchers will manage their data appropriately, and/or in some combination of both. It is increasingly common, for example, for a data management plan with respect to a research project to nominate a “data steward” or “data custodian” who is responsible for complying with both relevant university policies and third-party obligations and expectations.75 73

Henry Chesbrough, Open Innovation: The New Imperative for Creating and Profiting From Technology (2003). 74 In 2014, Facebook was criticized heavily for enlisting users involuntarily in an “emotional contagion” experiment. Evan Selinger & Woodrow Hartzog, Facebook’s Emotional Contagion Study and the Ethical Problem of Co-opted Identity in Mediated Environments Where Users Lack Control, 12 res. ethICs 35 (2016). 75 See, e.g., Sara Rosenbaum, Data Governance and Stewardship: Designing Data Stewardship Entities and Advancing Data Access, 45 health servICes res. 1442 (2010), doi:10.1111/j.1475-6773 .2010.01140.x.

Data governance and the emerging university 389 In short, data governance requires contextual understanding and institutional practice. These are partly empirical questions, and they are also partly normative and aspirational questions. They point to ways in which data governance concerns data as a shared resource shared across specific communities and organizations, rather than data as a knowledge resource inside a specific university. Researchers collaborate across institutions and sometimes collaborate across disciplines. Researchers change jobs. In what respects does researcher mobility intersect with data governance requirements and aspirations? There is no single answer to that question. A data management policy may describe the extent to which the university and its researchers should anticipate and plan for relevant technology contingencies, including ownership of and access to both computer hardware and code.76 One may say that the problem of data governance in a particular university is nested within the problem of data governance in scholarly research generally.77 One should not assume that governance strategies at one nesting level automatically translate with equal effect to other nesting levels. Nor is the university necessarily prone to claiming undue control over data while third parties are necessarily advocates for broader sharing. The data-intensive university does not automatically mean reducing or eliminating technology transfer strategies. “Who owns the data?” is not an irrelevant prompt for interrogating contextually-appropriate governance strategies, even if it is an inappropriate overarching prompt. As described earlier, technology transfer in the university can be understood as a contextually-appropriate mode of information sharing, and its detailed implementation can be described institutionally as a part of a data governance framework. Data governance at ground level entails understanding the intersections of governance strategies and interests at multiple levels. In both conceptual and pragmatic senses, data is often managed as commons via strategies of openness, sharing, and polycentricity,78 but with contextually-appropriate elements of proprietary management and exclusivity.

V.

CONCLUSION

This Chapter has two purposes. One, it extends ongoing critiques and assessments of the modalities of knowledge production, transmission, and stewardship that characterize the modern research university, highlighting conflicts and contradictions between assumptions and practices that point toward openness, sharing, and the public benefit, on the one hand, and assumptions and practices that point toward closure, proprietary practice, and private interest, on the other hand. Two, it re-orients those critiques and assessments via exploration of emerging interest in data governance in the university. Building on globalization of research networks and rapid advances in information technology, data-intensive research increasingly dominates many areas of university research practice, so much so that the research university traditionally grounded in the pursuit of knowledge may now be understood more accurately as 76

See Borgman, supra note 7, at 224–37. See Elinor Ostrom and Charlotte Hess, “A Framework for Analyzing the Knowledge Commons” in Understanding Knowledge as a Commons: From Theory to Practice (Charlotte Hess & Elinor Ostrom eds., 2006); Joseph E. Stiglitz, “Knowledge as a Global Public Good” in Global Public Goods: International Cooperation in the 21st Century (Katherine J. Strandburg, Brett M. Frischmann, & Michael J. Madison eds., 1999). 78 Context-specific case studies are documented in Governing Medical Knowledge Commons (Katherine J. Strandburg, Brett M. Frischmann, & Michael J. Madison eds., 2017). 77

390 Research handbook on intellectual property and technology transfer the data-intensive university, in which data are described broadly and pluralistically to include modes of evidence production, collection, analysis, and curation throughout the research enterprise. Given that re-characterization of the university, standard governance framings in terms of open vs. closed frameworks, often aligned with typical expressions of IP law, are no longer descriptively or normatively sufficient. Open sharing of information resources, in practices of Open Science, Open Access, FAIR information practices, and proprietary distribution of research products via technology transfer (including Open Innovation practice) are now better understood both conceptually and practically as parts of the ecologies and practices of data in modern research. So, at a practical level, while data governance policies and practices are being drafted, reviewed, and implemented, questions and challenges arise at several levels simultaneously. Some of those derive from the mostly “ordinary” set of challenges that apply when overlapping and potentially inconsistent interests and goals need to be integrated into a policy instrument in a complex organization. For data governance, what is relevant data? Who is covered by a policy instrument? What are relevant actors entitled to do, or expected to do, with data? In each case, why and how, and how will any associated burdens be absorbed by the university or otherwise accounted for in terms of money, time, and expertise? Those mostly “ordinary” challenges expose some broader, more open-facing questions about the character of the university’s institutional setting and mission and about background expectations, needs, and goals concerning intellectual resources. Policies and practices concerning data are still emerging. The Chapter focuses on concern that data will be “thing-ified” if they are characterized by analogy to patentable inventions and copyrightable works and therefore subjected presumptively to thinking and practice derived from IP law. Data governance throughout the university offers opportunities to promote a thoughtful reimagining of the purposes and practices of research institutions and a broader understanding of the roles of institutions in information policy generally.

18. “Free data?”: open science in the age of personal data protection Paolo Guarda

I.

OPEN SCIENCE AND DATA-DRIVEN SOCIETY

A.

Premise

Current society gives extreme importance to information in all contexts of human activity. Terms such as “information society” or “data-driven society” are recognized as describing scenarios where information is no longer just functional to the realization of other principal services and activities; it becomes the main core, the prime mover of new business models. Even more evocative expressions are used to describe this concept, such as “new oil.”1 Lawmakers, both nationally and internationally, have put great emphasis on this kind of concept for years. It is not always clear, however, what kind of tools are intended to be adopted to fully realize this free and easy use and dissemination of data. The emphasis is, indeed, very often on traditional tools that highlight the importance of existing intellectual property rights (“IPRs”) in order to tighten the rules and consequently strengthen the position of rights holders. It is difficult to harmonize this kind of intervention with other approaches and paradigms that have emerged over the years. From the perspective of the focus on IP in university technology transfer, two different visions are facing each other: the overprotection approach, based on strict rules and strong protection of the rights holder; and, on the other side, an open one, based on a more flexible and dynamic approach to the management of the author’s rights as well as on some users’ rights. The first is affirmed and reiterated in the various documents and regulatory provisions that govern the subject at international and national level, with an approach that is influenced by the origin and the historical and technological context in which those rules have taken shape. The second one is, on the other hand, claimed, within the Western countries, from many sectors of the society (e.g. free software developers, academic scientists, etc.) and represents an essential tool to develop “Open Science” (OS). This approach makes flexibility its “watchword”, and is based on lowering economics, technology and legal (IPRs and contract) barriers to access information as well as on democracy, transparent, pluralistic and inclusive approach to science and innovation stimulating free flow of ideas and information across borders.2 Thus, the concept of “Open Science”, in all its “forms” (Open Access, Open Data, Open Source Software and so on), identifies a totally different approach

1 As far as it was possible to determine, Clive Humby, a UK mathematician and architect of Tesco’s Clubcard is widely credited as the first to coin the phrase: “Data is the new oil. It’s valuable, but if unrefined it cannot really be used. It has to be changed into gas, plastic, chemicals, etc. to create a valuable entity that drives profitable activity; so, must data be broken down, analyzed for it to have value.” 2 See Roberto Caso & Paolo Guarda, “Copyright Overprotection Versus Open Science: the Role of Free Trade Agreements” in Free Trade Agreements 35 (Lilian Corbin & Mark Perry eds., 2019).

391

392 Research handbook on intellectual property and technology transfer to the management of information than the classical “proprietary” one. The emphasis is on the importance of information sharing, the free use and reuse of data, interoperability and openness. The term “Open Science” refers to a movement that has as its central goal the sharing of research results (publications and data) on the Internet. Therefore, free access with reuse rights represents the heart of it. The core of the revolution of OS lies in the full exploitation of the Internet as an instrument of dialogue and sharing.3 From this perspective OS identifies itself as Open Access (“OA”) to the results of scientific research (i.e. publications and data).4 The date of beginning of OA is uncertain. It is a phenomenon that arose spontaneously at the dawn of the Internet. Some initiatives in the 1960s and 1970s preceded its birth. In the late 1980s and early 1990s, before the Internet was dominated by commercial interests and the Web was taking its first steps, the first scientific journals and OA archives were published.5 OA is, in fact, a movement that is close and, in many respects, connected to Free and Open Software and to the cooperative development of the Internet. It means granting for free the rights of use to the public, such as copying rights, the right to modify the work and to process derivative works, the right of distribution, the right of communication to the public. Furthermore, there are those who would like to include in the notion of OS also the concept of Open Innovation (“OI”): the form of production of new technology that is based not only on internal resources of the organization—typically, the company—but also on external ones. Around the core of Open Science, the idea of a more democratic, transparent, inclusive, fair and interdisciplinary science develops.6

3

The OS expression is recent. In the past, the term Open Access (“OA”) was used: see Peter Suber, Open Access (2012). The conceptual foundations of OS lie in that part of Western philosophy that has exalted the public use of reason, Immanuel Kant, Beantwortung der Frage: Was ist Aufklärung?, berlInIsChe MonatssChrIft (1784), at 481–94, and in the modern scientific revolution that put at the center of its own mission the public character of communication. See Robert K. Merton, Science and Technology in a Democratic Order, 1 J. legal & pol. soCI. 115 (1942); Paul A. David, The Republic of Open Science. The Institution’s Historical Origins and Prospects for Continued Vitality, stan. sIepr dIsCussIon papers 13 (2014). There are new interdisciplinary theoretical developments connecting OS and OA with the theory of common goods of knowledge: see Charlotte Hess, & Elinor Ostrom, “A Framework for Analyzing the Knowledge Commons” in Understanding Knowledge as a Commons: from Theory to Practice (2005). For further analysis see Roberto Caso, Scienza aperta, Trento LawTech Research Paper nr. 32, 2017, available at http://hdl.handle.net/11572/183528 (last visited Oct. 17, 2019); Thomas Margoni, Roberto Caso, et al., “Open Access, Open Science, Open Society” in Positioning and Power in Academic Publishing: Players, Agents and Agendas 75, 75–86 (Fernando Loizides & Birgit Schmidt eds., 2017). 4 According to one of the most accredited definitions, the OA literature is digital, online, with free access and without the major restrictions arising from copyright and contractual license; see Suber, supra note 3. For a review of the literature on OA see Giancarlo Frosio, Open Access Publishing: A Literature Review, CREATe (working paper (2014)), available at https://www.create.ac.uk/publications/open -access-publishing-a-literature-review/ (last visited Oct. 17, 2019). For the framing of the scientific communication system in the theory of common knowledge goods see Charlotte Hess & Elinor Ostrom, Ideas, Artifacts, And Facilities: Information As A Common-Resource Resource, 66 l. & ConteMp. probs. 111, 134 (2003). 5 For further details on OA archives, see Paolo Guarda, “Open Access to Legal Scholarship and Open Archives: Toward a Better Future” in From Information to Knowledge—Online access to legal information: methodologies, trends and perspectives 143 (Maria Angela Biasiotti & Sebastiano Faro eds., 2011). 6 For a review of possible meanings that may refer to the values of democracy and transparency see Benedikt Fecher & Sascha Friesike, “Open Science: One Term, Five Schools of Thought” in Opening

“Free data?”: open science in the age of personal data protection 393 To make it more complex, the discipline concerning personal data protection in some cases needs to be taken into consideration. It evidently happens only when data that are the object of sharing and/or possible use are “personal data.” The United States and the European Union have, as is known, a different approach to this problem and have come up with different solutions.7 However, it is widely known how much of a global impact the new European General Data Protection Regulation has had and has nevertheless engaged discussions on this issue regardless of the (geographical) context of reference. Information privacy from this perspective can represent a further obstacle to the free circulation of data and to the potential of the business models conceived around it. This contribution addresses the issue from the perspective of Open Science, seen as a model of sharing and use of information that is not only ethically correct for all the possible stakeholders (public bodies or private corporations) but also and above all “necessary” for the public bodies engaged in the creation and dissemination of information (universities, research centers, and so on). The Open Data scenario will be taken as a benchmark, since it represents a privileged field of study with reference to the interaction of rules and interests relating to the management of information and of possible personal data. From a methodological point of view, I will move to the analysis of the detailed regulatory rules when needed; instead, I will tackle the several issues by presenting the legal institutions involved at the general level. I will, therefore, reference the operating rules only when precise information appears to be functional to the description and explanation of a potentially critical and peculiar aspect of the main topic. Following this introduction, the first section will focus on the general aspects of open science, describing opportunities, interests and needs related to the adoption of these open models of use and circulation of information. The second section will provide a brief overview of IPRs involved in Open Data issues and in the management of information. The third section will be devoted to analyzing data protection regulation, focusing, in particular, on the Fair Information Principles, the OECD Guidelines and the new European General Data Protection Regulation; a sub-section will deal with the research activity: taking the Horizon 2020 research project funding as a paradigmatic example for testing the impact of approaches based on open logic in research contexts involving the use of personal data. Finally, the concluding part of the Chapter will draw some general considerations on how to balance the interests involved along the lines of “specific purpose v. any purpose” dichotomy. B.

Open Data: Opportunities, Interests and Needs

One of the possible developments of Open Science is represented by the Open Data (“OD”) movement, the application context I will take as a paradigmatic example in this Chapter to analyze the relationship between the values of openness, IPRs and data protection regulation.8

Science 17 (Sönke Bartling & Sascha Friesike eds., 2013); see also Roberto Caso, The Darkest Hour: Private Information Control and the End of Democratic Science, Trento LawTech Research Paper 35, 2018, available at http://hdl.handle.net/11572/208881 (last visited Oct. 17, 2019). 7 See Daniel J. Solove & Paul M. Schwartz, Reconciling Personal Information in the United States and European Union, 102 CalIf. l. rev. 877 (2014); Rossana Ducato, “La crisi della definizione di dato personale nell’era del Web 3.0. Una riflessione civilistica in chiave comparata” in Il Diritto e le Definizioni 145 (Fulvio Cortese & Marta Tomasi eds., 2016). 8 For a complete study with references to the OD issue see Lucie Guibault & Andreas Wiebe eds., Safe to Be Open: Study on the Protection of Research Data and Recommendations for Access and

394 Research handbook on intellectual property and technology transfer OD are data that can be freely accessed, used, modified and shared for any purpose and are subject, at most, to requirements that preserve provenance and openness. A possible definition of OA provided by the “Open Knowledge International” is based on the following main features:9 ● “availability and access”: the data must be available as a whole, for a price not exceeding a reasonable cost of reproduction, preferably by downloading from the Internet. The data must be available in a convenient and modifiable form; ● “reuse and redistribution”: data must be provided under terms that permit reuse and redistribution, including mixing with other datasets. The data must be machine-readable; ● “universal participation”: everyone must be able to use, reuse and redistribute—there should be no discrimination against fields of endeavor or against persons or groups. For example, “non-commercial” restrictions that would prevent “commercial” use, or restrictions of use for certain purposes (e.g. only in education), are not allowed. Finally, the fundamental rationale behind the meaning of “open” is “interoperability”: the ability to combine a database with others. Dealing with OD, you may also refer to the discipline of “Open Government Data,” that is a doctrine according to which public administration should be open to citizens, both in terms of transparency and direct participation in the decision-making process, also through the use of new information and communication technologies. Therefore, Open Government Data is a philosophy and increasingly a set of policies that promotes transparency, accountability and value creation by making government data available to all. While performing their functions, public bodies produce huge quantities of data and information. By making their datasets available, public institutions become more transparent and accountable to citizens; by fostering the use, reuse and free distribution of datasets, governments promote business creation and innovative, citizen-centric services. Policy makers and governments are paying close attention to the topic and OD has developed from being a concept especially dedicated to scientific research areas into a central and fostering idea carried into public-policy.10 A focus on removing technical restrictions to reuse can be found in the 2013 G8 Open Data Charter,11 Open Data guidelines of the Organisation for Economic Cooperation and Development (“OECD”),12 US President Barack Obama’s Executive Order,13 and Canada’s Action Plan on

Usage, 2013), available at https://www.univerlag.uni-goettingen.de/handle/3/isbn-978-3-86395-147-4 (last visited Oct. 17, 2019). 9 Open Knowledge International, available at https://okfn.org/opendata/ (last visited Mar. 28, 2019). 10 See Frederik Zuiderveen Borgesius, Jonathan Gray, & Mireille van Eechoud, Open Data, Privacy, and Fair Information Principles: Towards a Balancing Framework, 30 berkeley teCh. l.J. 2108, 2073 (2015). 11 g8 open data Charter, 2013, available at https://www?.gov?.uk/?government/?publications/ ?open-data-charter/ g8-open-data-charter-and-technical-annex. 12 See Barbara Ubaldi, Open Government Data: Towards Empirical Analysis of Open Government Data Initiatives, OECD, 2013; see also OECD Recommendation of the Council for Enhanced Access and More Effective Use of Public Sector Information, OECD, 2008. 13 Exec. Order No. 13642, 3 C.F.R. § 13642 (May 9, 2013)

“Free data?”: open science in the age of personal data protection 395 Open Government.14 In Europe, a structured statutory example of this trend is represented by the Public Sector Information (“PSI”) Directive15 that addresses the transparency requirements of the PSI as a means of economic activity by the private sector. It is supposed to facilitate and encourage the reuse of PSI produced and stored by public bodies (for example state, regional or local authorities) and is aimed at creating an up-to-date common basis throughout the EU for the use of public sector data, encouraging public bodies to make data available electronically for free or, if they choose not to, for a charge no greater than the combined cost of collecting, storing, processing and making the data available. In conclusion, it is worth citing the wide variety of social and political goals that OD are expected to achieve.16 The first goal pursued through OD policy is fostering political accountability and democratic participation. The idea of open government is tied to the ideal of transparency of governments’ decisions and activities. A second objective pursued is related to innovation and economic growth. Open data initiatives generally highlight the potential of enabling the reuse of public sector information in order to create new businesses and innovative services and products. From this perspective OD policies are increasingly becoming the favorite way to unlock the value of public sector information. Business models based on it may be diverse: financial services providers using official statistics as input; companies in the meteorological sector using weather data to provide highly specialized services; postal codes used as identifiers, and so on. Finally, a common “mantra” of OD is “efficiency”: it should help to save resources and improve (public and private) services. The digitalization and automation of processes and workflow is often conceived by policy makers as a recipe for inefficiencies in the analog context and above all as a way to save economic resources by improving the final service.

II.

OPEN DATA AND IPRS: A BRIEF OVERVIEW

IPRs come into play when addressing the topic of OD. For the sake of completeness and although by no means attempting to be exhaustive, this section will provide the basic notions concerning the protection of databases through IPRs and their possible exploitation through licenses.17

14

Open Government Canada, July 25, 2018, http://open.canada.ca; Government of Canada: Ottawa, On, Canada, Canada’s Action Plan on Open Government, Can. gov., 2015, available at http://data.gc .ca/eng/canadas-action-plan-open-government; see generally Amy Conroy & Teresa Scassa, Promoting Transparency while Protecting Privacy in Open Government in Canada, 53 (1) alberta l. rev. 175 (2015). 15 Directive 2013/37/EU of the European Parliament and of the Council of June 26, 2013 amending Directive 2003/98/EC on the reuse of public sector information. Pending the publication of this chapter, this Directive has been replaced by the Directive (EU) 2019/1024 of the European Parliament and of the Council of 20 June 2019 on open data and the re-use of public sector information: Member States have until 16 July 2021 to transpose the new directive into national law. 16 See Borgesius et al, supra note 10, at 2080–7; for further analysis see Jonathan Gray, Towards a Genealogy of Open Data, General Conference of the European Consortium for Political Research in Glasgow, Scotland (2014) . 17 See Guibault & Wiebe, supra note 8, at 19–92; see also J. H. Reichman, Pamela Samuelson, Intellectual Property Rights in Data, 50 vand. l. rev. 49 (1997).

396 Research handbook on intellectual property and technology transfer The database is protected by copyright when the criteria for the selection and arrangement of data are creative and reflect an original choice of the author.18 The US Copyright Act (art. 17 U.S.C. §101) and the European Database Directive (art. 3 of Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the legal protection of databases) provide the same operational rules, with some obvious differences in the legislative drafting and its application. Non-original databases that select or present information in a trivial order are not protected by copyright: for example, following a chronological or alphabetical order for a list of subscribers of a telephone company.19 Copyright protection arises automatically as soon as the created object has some material form. The work does not need to be registered or published in order to get copyright protection. The first holder of copyright is “the author,” which in the case of a database is the person who creates the database. In summary, the copyright holder has the exclusive right to: copy the work; issue copies to the public; rent or lend to the public; communicate to the public; make an adaptation or do any of these other acts in relation to an adaptation. In addition to that, a database may also qualify for a specific right introduced in 1996 within the EU: the so-called Sui Generis Database Right (“SGDR”).20 This right, established by Directive 96/9/EC, subsists in a database if there has been a substantial investment in obtaining, verifying or presenting the contents of the database (even if the contents and/or structure of the database are not original and therefore do not attract copyright). Investment is construed widely and covers financial, human and technical resources. Like copyright, the database right arises automatically. The “maker,” i.e. the person who makes the substantial investment, has the right to prohibit the extraction or reuse of a substantial part of the database. This right, if the conditions are met, can exist simultaneously with copyright on the same database. Copyright and SGDR do not directly affect the content of the database itself. To identify the legal regulation applicable to the content of the database, it is necessary to evaluate the specific application scenario and the specific contractual provisions for accessing and using the database. By way of example let us consider the case of: mere facts (e.g. a database relating to meteorological surveys) for which there is no legal form of ownership; works protected by copyright (e.g. photographs, scientific articles, and so on) for which copyright may be involved; personal data for which reference should be made to the regulations on personal data protection (and, therefore, it will be necessary to define who the data controller is).21 One of the typical activities of the research scenario connected to the use of databases and facing evident difficulties with the application of IPRs is the so-called “Text and Data Mining” (“TDM”). Data mining is the set of techniques and methodologies that have as their object the extraction of useful information from large amounts of data (e.g. database, data warehouse and so on) through automatic or semi-automatic methods. In order to extract the data, it is necessary to copy the articles or data contained in a database on a machine. This may constitute a violation of IPRs. TDM is emerging as a powerful tool for harnessing the power in data

18

See Estelle Derclaye, The Legal Protection of Databases: A Comparative Analysis (2008). See the famous case Feist Publications, Inc., v. Rural Telephone Service Co., 499 U.S. 340 (1991). 20 See P. Bernt Hugenholtz, “Something Completely Different: Europe’s Sui Generis Database Right” in The Internet and the Emerging Importance of New Forms of Intellectual Property 205 (Susy Frankel & Daniel Gervais eds., 2016) 21 A very interesting and thought-provoking contribution on possible forms of data propertization is available in Lothar Determann, No One Owns Data, 70 hastIngs l. J. 1 (2018). 19

“Free data?”: open science in the age of personal data protection 397 by analyzing datasets and content at multiple levels in order to discover concepts and entities in the world, patterns they may follow and relations they engage and on this basis annotate, index, classify and visualize such content.22 From a legal point of view, these datasets and content (e.g. data, alphabetic or numerical entries, texts, articles, papers, collections of words) can indeed receive different types of protection. For the purpose of most TDM activities two forms of protection can represent a real barrier: copyright on the elements of the database (protecting the single elements of the database when these are original works of authorship (e.g. scientific papers, drawings, images)) and, in the EU, the SGDR. TDM often requires making a temporary copy of the datasets or works to be mined. If we consider the scenario where both the above described legal titles operate, i.e. the European Union, we note that the EU legal framework is based on the assumption that authors deserve a high level of protection (see Recital 9 of Directive 2001/29/EC of the European Parliament and of the Council of 22 May 2001 on the harmonization of certain aspects of copyright and related rights in the information society (“InfoSoc Directive”)23) which has led to the formulation of very broad definitions of protected rights. On the contrary, the set of rules intended to balance this exclusivity has been drafted in very loose terms and accordingly exception and limitations to copyright and to the SGDR are exhaustively listed in the InfoSoc and Database Directive but are not made mandatory (except for art. 5.1 of InfoSoc).24 Given this framework TDM most likely infringes copyright and/or the SGDR, lacking a specific nationally implemented exception (to date only France and the United Kingdom have created a TDM exception limited to non-commercial purposes). In other legal systems, like the US one, TDM and web mining have been considered a transformative use covered by fair use.25 It is worth to be cited that the lately enacted art. 3 of Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019 on copyright and related rights in the Digital Single Market and amending Directives 96/9/EC and 2001/29/EC establishes a new EU copyright exception for TDM, but only for “research institutions” and “for the purposes of scientific research”. To fully implement OD and, therefore, to make the default rules on IPRs more flexible, Open Licenses such as the Creative Commons Public License (“CCPL”) version 4 are a technically viable alternative to the lack of proper legislative intervention in this field. CCPL v4.0 addresses both copyright and SGDR in the licensed work. In particular, by applying a CCPL v4.0 to a database such as a website or a repository of journal articles the licensor (the person who applies the license and who needs to be the right holder or be authorized by the right holder to do so) is giving permission to reuse: (a) the SGDR in the database; (b) copyright in the database in the limited cases in which copyright applies to its structure; and (c) copyright and/or related rights in the elements (works such as journal articles and original photographs) composing the database. While other open content licenses may also achieve the same results,

22

See Caso, et al., supra note 3. Infosoc Directive was enacted to implement the WIPO Copyright Treaty and to harmonize aspects of copyright law across Europe, such as copyright exceptions. 24 Of the 21 exceptions listed in art. 5 of InfoSoc only one is mandatory, while the remaining 20 are implemented at the discretion of each of the 28 European Member States. This situation is clearly unsatisfactory in terms of legal certainty and even though some countries (such as the UK) have shown foresight by creating a dedicated TDM exception, the presence of a non-commercial limitation still represents a competitive barrier if compared to other more dynamic legal systems (e.g. the United States). 25 Authors Guild, Inc. v. Google Inc., 954 F. Supp. 2D 282 (S.D.N.Y. 2013), aff’d sub nom. Authors Guild v. Google, Inc., 804 F.3d 202 (2d Cir. 2015). 23

398 Research handbook on intellectual property and technology transfer the convergence towards one or a few licenses that can be recognized as a de facto standard could be not only desirable but also essential in order to lower the transactional costs associated with license compatibility and therefore to facilitate use and reuse of resources (even for goals such as TDM).26

III.

OPEN DATA AND DATA PROTECTION REGULATION

A.

Data Protection Regulation and Critical Issues

Open Data can be beneficial for a range of activities and contribute to a wide variety of social, political and economic goals. When databases subject to dissemination contain personal information, critical issues increase and need to be carefully considered. Among others, public bodies traditionally manage an enormous amount of personal data. This trend can only increase in the light of the many new projects (i.e. “smart cities”) that aim to process personal data of citizens to improve, for example, mobility services; moreover, the availability of more and more online services exponentially increases the amount of personal data potentially generated through these processes. Therefore, it is worth weighing and balancing the rights of people involved with regards to the processing of their personal data. On a global level, the right to privacy is protected by various declarations and legislative documents: the United Nations Declaration of Human Rights,27 the International Covenant on Civil and Political Rights,28 the US Fourth Amendment, the European Convention on Human Rights,29 the European Union Charter of Fundamental Rights,30 national constitutions and other laws protecting privacy. Managing the dissemination of a database containing personal information and adopting the open logic therefore presents a series of critical issues determined by the interaction with privacy principles and citizens’ expectations. Following an approach proposed by the doctrine, we may identify three general categories of possible clashes.31

26 See Lucie Guibault, “Licensing Research Data Under Open Access Condition” in Information and Knowledge: 21st Century Challenges in Intellectual Property and Knowledge Governance (Dana Beldiman ed., 2013); Lucie Guibault & Thomas Margoni, Analysis of Licensing Issues published in Lucie Guibault, Safe To Be Open. Study on the Protectiono of Research Data and Recommendations for Access and Usage 143–160 (Andreas Wiebe ed., 2013). There are interesting research projects dedicated to this issue. Among others, it is worth quoting “OpenMinTeD”: a European project of 16 partners that sets out to develop a sustainable infrastructure for text and data mining, working on a platform that brings together text and data, TDM tools and services, training, support and guidelines: open MInted, available at http://openminted.eu (last visited Mar. 28, 2019). 27 Universal Declaration of Human Rights, art. 12, G.A. Res. 217A (III), U.N. Doc. A/810, at 71 (1948). 28 International Covenant on Civil and Political Rights, art. 17, Dec. 16, 1966, S. Treaty Doc. No. 95-20, 6 I.L.M. 368 (1967), 999 U.N.T.S. 171. 29 Convention for the Protection of Human Rights and Fundamental Freedoms, art. 8, Nov. 4, 1950, 213 U.N.T.S. 222. 30 Charter of Fundamental Rights of the European Union of the European Parliament, arts 7–8, 2010, O.J. (C 83) 2, 1. 31 See Borgesius et al., supra note 10 at 2087.

“Free data?”: open science in the age of personal data protection 399 First of all, a sort of “chilling effect:” people may fear that their information will be stored or made public. This could affect their propensity to use, for example, services offered where there is doubt as to whether personal information remains truly confidential. The uncertainty with regard to the correct processing of personal data can ultimately have a negative impact on the quality of the services offered (both on a private and public level). Furthermore, it is clear that if the information is released in OD, people lose control over their personal data. The release in OD implies, in fact, that the information can be reused for an indefinite number of times and for any purpose. Privacy stresses the importance of individuals’ control over their personal information and emphasizes the importance of the principle of purpose specification in the processing of personal data. There is another effect: with the increase of (public or private) databases released in OD there is more potential for re-identification of information considered anonymous. This leads to an exponential boost in risks to citizens’ privacy and a reduction of control over their personal information. A third issue related to privacy is represented by the fact that OD could be used as an input for social sorting and discriminatory practices, thus increasing the misuse of personal information and potential invasions of citizens’ privacy (see the use of personal data packages for direct marketing purposes, credit scoring or screening job applications). It is therefore pivotal to find out the correct balance between the shared needs that encourage the dissemination in OD with the rights that data protection regulation recognizes to individuals, in the last instance as an essential protection of their dignity. The detailed discipline on personal data protection may change depending on the country of reference. To study the topic, however, we may start from some general principles. The “Fair information practice principles” (“FIPPs”) or “Fair Information Principles” (“FIPs”) are the outcome of the United States Federal Trade Commission’s inquiry into the manner in which online entities collect and use personal information and safeguards to assure that practice is fair and provides adequate information privacy protection.32 They are guidelines containing widely accepted concepts concerning fair information practice in an electronic marketplace. Several organizations and countries have their own regulations or guidelines including a version of the FIPs. The most globally influential version of these principles is found in the Guidelines Governing the Protection of Privacy and Transborder Flows of Personal Data,33 from the OECD.34 This document has two main goals: the protection of privacy and individual freedoms and the promotion of free flows of information between the member countries of the OECD. The Guidelines set “minimum standards:” they are not legally binding and represent mere recommendations to member countries about their adoption. When the OECD Guidelines were adopted in 1980, only about one third of the member states had

32

See Neil Richards, Intellectual Privacy: Rethinking Civil Liberties in the Digital Age 162 (2015); Robert Gellman, Fair Information Practices: A Basic History, BobGellman.com (last visited Mar. 28, 2019); Marc Rotenberg, Fair Information Practices and the Architecture of Privacy (What Larry Doesn’t Get), 1 stan. teCh. l. rev. 1, 4 (2011); Ann Cavoukian, “Evolving FIPPs: Proactive Approaches to Privacy, Not Privacy Paternalism” in Reforming European Data Protection Law 293 (Serge Gutwirth & Ronald Leenes & Paul de Hert eds., 2015). 33 Recommendation of the Council Concerning Guidelines Governing the Protection of Privacy and Transborder Flows of Personal Data, OECD, 2013, available at http://www.oecd.org/sti/ieconomy/ 2013-oecd-privacy-guidelines.pdf (as amended on July 11, 2013). 34 OECD is an intergovernmental economic organization with 36 member countries, founded in 1961 to stimulate economic progress and world trade.

400 Research handbook on intellectual property and technology transfer adopted a data privacy law. Now, almost every OECD member state has a data privacy regulation with the FIPs at its core. The Guidelines were updated in 2013, but the essence of the principles remained unchanged. These principles are: ● Collection Limitation Principle—There should be limits to the collection of personal data and any such data should be obtained by lawful and fair means and, where appropriate, with the knowledge or consent of the data subject.35 ● Data Quality Principle—Personal data should be relevant to the purposes for which they are to be used, and, to the extent necessary for those purposes, should be accurate, complete and up-to-date.36 ● Purpose Specification Principle—The purposes for which personal data are collected should be specified no later than the time of data collection and the subsequent use limited to the fulfillment of those purposes or such others as are not incompatible with those purposes and as are specified on each occasion of change of purpose.37 ● Use Limitation Principle—Personal data should not be disclosed, made available or otherwise used for purposes other than those specified in accordance with the Purpose Specification Principle except: (a) with the consent of the data subject; or (b) by the authority of law.38 ● Security Safeguards Principle—Personal data should be protected by reasonable security safeguards against such risks as loss or unauthorized access, destruction, use, modification or disclosure of data.39 ● Openness Principle—There should be a general policy of openness about developments, practices and policies with respect to personal data; there should be readily available means of establishing the existence and nature of personal data, and the main purposes of their use, as well as the identity and usual residence of the data controller.40 ● Individual Participation Principle—An individual should have the right to: (a) obtain from a data controller, or otherwise, confirmation of whether or not the data controller has data relating to him or her; (b) have communicated to him or her data relating to him or her within a reasonable time; at a charge, if any, that is not excessive; in a reasonable manner; and in a form that is readily intelligible to him or her; (c) be given reasons if a request made under subparagraphs (a) and (b) is denied, and to be able to challenge such denial; and (d) challenge data relating to him or her and, if the challenge is successful to have the data erased, rectified, completed or amended.41 ● Accountability Principle—A data controller should be accountable for complying with measures which give effect to the principles stated above.42

35

See supra note 33 at 14 (explaining in para. 7 the Guidelines governing the protection of privacy and transborder flows of personal data). 36 Id. 37 Id. 38 Id. 39 Id. at 15. 40 Id. 41 Id. 42 Id.

“Free data?”: open science in the age of personal data protection 401 The OECD Guidelines apply to personal data, which means “any information relating to an identified or identifiable individual (data subject)”43 that, “because of the manner in which they are processed, or because of their nature or the context in which they are used, pose a risk to privacy and individual liberties.”44 Then they follow a risk-based approach. This is a core difference with respect to the EU approach to data protection regulation, which applies to personal data processing even when it does not pose a clear risk for individual liberties.45 Moving to the EU legal system, the reference framework, at the level of general declarations, is that of Article 8 of the Charter of Fundamental Rights of the EU (recognizing the protection of personal data as an autonomous fundamental right) and Article 16 of the TFEU (Treaty on the Functioning of the European Union). The operating rules are now provided by the new General Data Protection Regulation (“GDPR”),46 approved in 27 April 2016 and finally enforced on 25 May 2018. GDPR updates the European legal framework on personal data protection, which for more than 20 years was based on Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995. It is of fundamental importance not only because it reviews the protection tools aimed at protecting privacy in the face of new technological challenges, but also because it has direct effect in all the Member States of the Union. The Regulation was, in fact, designed to unify the European legal framework in this matter, overcoming the contradictions or inconsistencies that emerged between individual national implementations. The EU model is now a sort of global benchmark, since European lawmakers believe that this regulatory intervention may have a global “domino effect,” because it represents the most thorough approach to guaranteeing the protection of personal data. GDPR incorporates the principles expressed at the level of OECD Guidelines, integrating them with the particular guaranteed European approach. The principle of “accountability” assumes a pivotal role and provides that, considering the nature, the scope, the context and the aims of the processing, as well as the different levels of risk to the rights and freedoms of individuals, the data controller shall put in place appropriate technical and organizational measures to ensure and be able to demonstrate that the data are processed in accordance with the Regulation (GDPR, art. 24(1)). The accountability finds its first operational expression in so-called “privacy by design”: art. 25 GDPR, indeed, requires incorporating the principles and rules on the protection of personal data starting from the design of the process, and especially at the level of IT solutions. The principle is taken up and explained in Recital 78, which states that appropriate technical and organizational measures must be taken to ensure that the requirements of the Regulation are met; the controller should

43 Id. at 13. The OECD personal data definition is similar to the definition in EU data protection law (art. 4, lett. a, GDPR). 44 Id. at 2. 45 See EU Charter of Fundamental Rights, art. 8 (stating “1. Everyone has the right to the protection of personal data concerning him or her. 2. Such data must be processed fairly for specified purposes and on the basis of the consent of the person concerned or some other legitimate basis laid down by law”). For a critical view on this position, see Raphaël Gellert, We Have Always Managed Risks in Data Protection Law: Understanding the Similarities and Differences Between the Rights-Based and the Risk-Based Approaches to Data Protection, 2 european data proteCtIon law revIew 481, 492 (2016). 46 Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (“General Data Protection Regulation”).

402 Research handbook on intellectual property and technology transfer then adopt internal policies and implement measures which meet in particular the principles of data protection by design and data protection by default, for example by minimizing the processing of personal data, pseudonymizing personal data as soon as possible, offering transparency with regard to the functions and processing of personal data, and so on.47 It is important to stress how GDPR requires that personal data be processed in a manner consistent with the purpose identified and notified to the person concerned with the processing (data subject) (principle of purpose limitation): the purpose has to be determined, explicit and legitimate; data cannot be further processed in ways that are incompatible with this purpose. An applicative scenario definitely interested in the phenomenon of OD, both as a creator and as a user of these, is that of scientific research. GDPR reiterates the preference of European lawmakers for processing for research purposes, whether they are “secondary” or primary, confirming the approach of the previous directive. It is in fact daily practice in scientific research that personal data be used for purposes other than those of the initial collection. Anyone who deals with personal data for research purposes can therefore benefit from forecasts that favor further processing, even in the particular case of sensitive data. The Regulation creates, first of all, an exemption from the principle of limitation for research purposes. Article 5(1)(b) states: further processing of personal data for archiving purposes in the public interest, scientific or historical research or for statistical purposes is not, in accordance with Article 89(1), considered incompatible with the initial purposes (“limitation of aim”).

Article 89 establishes, therefore, the safeguard measures that the data controllers must implement in order to be able to carry out further processing activities for research purposes. Another general rule is then confirmed to be in line with the European approach to the protection of personal data: consent is still, if not expressly excepted, the main tool for authorizing the controller to process personal information. This consent must be “freely given” and specific to the processing operation that is intended to be implemented. This requirement may represent a challenge for the research activity, as often “it is not possible to fully identify the purpose of the processing of personal data for the purposes of scientific research at the time of data collection (Recital 33).” If the processing for a purpose other than that for which the personal data was collected is not based on the consent of the data subject or on a specific statutory law of the Union or of the Member States, the data controller must perform a sort of compatibility test in order to check if this new processing is compatible with the purpose for which the personal data were initially

47

Privacy by design has to be implemented together with another general principle that has been codified in the Regulation: so-called “privacy by default,” which requires the adoption of appropriate technical and organizational measures to ensure that only the personal data necessary for each specific processing purpose are processed: GDPR, art. 23(2). On privacy by design, see Ann Cavoukian, Privacy by design: the definitive workshop—A foreword by Ann Cavoukian, 3 IdentIty Info. soC’y 247 (2010); Bert-Jaap Koops & Ronald E. Leenes, Privacy Regulation Cannot Be Hardcoded. A Critical Comment On The “Privacy By Design” Provision In Data Protection Law, 28 Int. rev. law CoMput. teCh. 1 (2013); Ann Cavoukian & Sylvia Kingsmill, Privacy by Design Setting a New Standard For Privacy Certification (2016).

“Free data?”: open science in the age of personal data protection 403 collected.48 This tool has some new features. Pursuant to art. 6(4), in fact, the data controller must take into account, among other things: (a) any link between the purposes for which the personal data were collected and the purposes of the further processing envisaged; (b) the context in which the personal data were collected, in particular with regard to the relationship between the data subject and the data controller; (c) the nature of personal data, especially if special categories of personal data are processed … or whether data relating to criminal convictions and offenses are processed …; (d) the possible consequences of further treatment envisaged for the interested parties; (e) the existence of adequate guarantees, which may include encryption or pseudonymisation.

If the outcome of the test proves that none of these elements has been modified in such a way as to make the subsequent processing unjustified or otherwise unlawful, the compatibility test will be deemed to be satisfied and the new processing activity lawful. Otherwise, further elaboration will have to rely on a different legal basis, such as the collection of a new consent by the data subject.49 B.

Open Data and Data Protection Regulation: Main Challenges

Managing the peculiarities of Open Data dissemination when the database includes personal data is not an easy task. Actually, little has been done to figure out a balanced solution and much has been left to willing interpreters who wish to face such a complex and insidious operational scenario.50 The major challenges are listed below.51 The first critical aspect is related to the “Purpose specification principle” (or “Purpose limitation principle”), which represents a cornerstone of many data protection regulations around the world. This is a very vague concept that is only partly clarified at the statutory level. It is also a very interesting legal issue concerning the possible clash between the different purposes of the processing on the one hand, and the circulation of content governed by an OD license on the other. While in the privacy context the focus on the “specific purpose” principle of the

48 On the compatibility of processing, see Art. 29 OF Working Group, Opinion 03/2013 on purpose limitation, adopted on 2 Apr. 2013 (WP 203). 49 See Recital 50 of GDPR. However, the Regulation also suggests that, at least in some circumstances, the research itself may provide a legitimate basis for the processing of personal data, even without the consent of the person concerned. For example, the data controller may lawfully process personal data, without such consent, when this is “necessary for the purposes of the legitimate interests pursued by the controller or by a third party, except where such interests are overridden by the interests or fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child.” See GDPR, Art. 6(1)(f). 50 Some work has been done in the EU in an attempt to reconcile privacy with open data by a thematic network founded by the European Commission to investigate on Legal Aspects of Public Information (LAPSI). See Heiko Richter, Open Science and Public Sector Information—Reconsidering The Exemption For Educational And Research Establishments Under The Directive On Re-Use Of Public Sector Information, 9 JIPITEC 51 (2018); Cristina Dos Santos, et al., On Privacy and Personal Data Protection, 6 Masaryk u. J.l. & teCh. 337 (2012); Mireille van Eechoud, et al., LAPSI Position Paper on Access to Data (2014), available at http://dare.uva.nl/?document/?2/?162858 (last visited Oct. 17, 2019). 51 See Borgesius et al, supra note 10, at 2073.

404 Research handbook on intellectual property and technology transfer processing forms the hub of the whole system of protection, OA expressly stresses the ability to reuse data “for any purpose.” This principle is very connected to another: the “Use limitation principle.” Personal data should only be used in accordance with the specified purpose, except “(a) with the consent of the data subject; or (b) by the authority of law.” GDPR refers to “purpose limitation principle” requiring collection of personal data only for “specified, explicit and legitimate purposes and not further processed in a manner that is incompatible with those purposes” (GDPR, art. 5(1)). Hence, personal data could be used for a new (incompatible) purpose only if the data subject has given specific consent to the new processing (this consent is considered necessary by some commentators before releasing data as open data).52 “Collection limitation principle” requires, instead, that there should be limits to the collection of personal data and any such data should be obtained by lawful and fair means and, where appropriate, with the knowledge or consent of the data subject.53 In the EU context this concept is named “Principle of data minimisation” (GDPR, art. 5(1)(c)), directly linked to so-called “Privacy by default.” This principle, in its various meanings and instances, impacts on OD processes in that they are characterized in an intrinsic way by the kind of freedom that should also feature in the management and reuse of data. Moreover, the “security and accountability principles” come into play. The data controller54 must adopt technical and organizational security measures to ensure that no one can access personal data unless authorized. She must also be able to show that everything in her power has been made to go unharmed by possible sanctions or claims for damages. It is evident that through Open Data dissemination both the controller and the data subject lose control of the information at the same time. That makes it impossible to properly address this privacy requirement. Furthermore, “data quality principles” requires appropriate accuracy, completeness and relevancy of personal data. This principle is particularly relevant in the context of OD, since releasing incorrect data could have a “detrimental effect” and also negatively impact the dissemination of data contained in the database and their quality. From this perspective privacy regulation and open data share the same tendency towards data accuracy. Finally, the “transparency principle” emphasizes the importance that the processing of personal data be completely clear and transparent, especially towards the data subject. This is to prevent possible misuse by the controller by virtue of the information asymmetry that is typical in this context. It represents probably the most relevant principle within the FIPs.55

52

See Bart van der Sloot, On the Fabrication of Sausages, or of Open Government and Private Data, 3 JedeM 1, 14 (2011). 53 See also Art. 29 Working Party: Opinion of the Article 29 Data Protection Working Party on Open Data and Public Sector Information (“PSI”) Reuse, at 20, 1021/00/EN WP 207 (June 5, 2013); Opinion on Purpose Limitation, 00569/13/EN WP 203 (Apr. 2, 2013). 54 According to the OECD Guidelines the party that “is competent to decide about the contents and use of personal data regardless of whether or not such data are collected, stored, processed or disseminated by that party or by an agent on its behalf.” OECD, Art. 1 (2013). Under EU data protection law the data controller is the party “which, alone or jointly with others, determines the purposes and means of the processing of personal data.” See GDPR, art. 4(7) (emphasis added). A party that processes personal data on behalf of the controller is the “data processor.” Id. at 8. 55 See Paul de Hert & Serge Gutwirth, Privacy, Data Protection and Law Enforcement: Opacity of the Individual and Transparency of Power, prIvaCy & CrIM. l. 61, 91 (2006).

“Free data?”: open science in the age of personal data protection 405 GDPR stresses the relevance of this principle, putting it as a point of reference for the processing together with that of accountability; in particular, Recital 39 states: “The principle of transparency requires that information and communication to the processing of those personal data be easily accessible and easy to understand and that clear and plain language be used.” It is evident that a complete lack of reuse restrictions, as provided by OD, clashes with the purpose specification principle: OD may, indeed, be used by anyone, for any purpose, without reuse restrictions. Then the main critical issue is represented by the (apparent) contrast between two principles that characterize the rules we are analysing: “specific purpose” (data protection) v. “any purpose” (OD). C.

A Paradigmatic Example as Interlude: The Horizon 2020 Research Pilot Policy with Reference to Open Data

In order to test the impact of approaches based on open logic in research contexts involving the use of personal data, I refer in this section to the EU Horizon 2020 research project funding as a paradigmatic example. Horizon 2020 is a funding program set up by the European Commission, the executive body of the European Union, to support and promote research in the European Research Area (“ERA”).56 It is the eighth of the framework programs for research and technological development, financial research, technological development and innovation (Framework Programs for Research and Technological Development), with an emphasis on innovation, accelerating economic growth and providing solutions to end users that are often government agencies. The name of the program has been changed to the Framework Program for Research and Innovation. Within this funding program an “Open Research Data Pilot” is being run, which demonstrates the EU’s support for open access to research results, in order to create greater efficiency, improve transparency and accelerate innovation. The Pilot aims to improve and maximize access to and reuse of research data generated by projects.57 The advantages of open access to research data are many: it builds on previous research results, improving quality of results; it fosters collaboration and avoids duplication of effort, achieving greater efficiency; it accelerates innovation, since faster to market means faster growth; it involves citizens and society, improving transparency of the scientific process. In the first stage (2014–2015) the Pilot comprised various selected areas of Horizon 2020: future and emerging technologies; research infrastructures (including e-infrastructures); leadership in enabling and industrial technologies; Social Challenge and so on. Over the years the number of areas has grown. For now, all projects covered by the Work Programme going forward will, by default, be part of the Open Research Data Pilot. The rationale stems from the assumption that knowledge produced by academic and scientific institutions has to be accessible to the academic community and society at large without

56 Horizon 2020, eur. CoMM’n, available at https://ec.europa.eu/programmes/horizon2020/ (last visited Mar. 28, 2019). 57 See Niels Dietrich & Andrea Wiebe, Open Data Protection—Study on Legal Barriers to Open Data Sharing—Data Protection and PSI 181–210 (2017).

406 Research handbook on intellectual property and technology transfer economic, legal or technological restrictions, enforcing the three main features of open access: free accessibility, further distribution, and proper archiving.58 The Open Research Data Pilot applies to two types of data:59 ● data, including associated metadata, needed to validate the results presented in scientific publications as soon as possible (i.e. statistics, results of experiments, measurements, observations resulting from fieldwork, survey results, interview recordings and images, and so on); ● other data, including associated metadata, as specified and within the deadlines laid down in a data management plan. Projects participating in the Pilot are required to: ● deposit the research data, preferably in a research data repository: online research data archives, which may be subject based/thematic, institutional or centralized;60 ● as far as possible, take measures to enable third parties to access, mine, exploit, reproduce and disseminate this research data (free of charge for any user): an effective way to perform that is to attach a Creative Common License (CC BY or CCO tool) to the data deposited. In order to optimize the potential future sharing and reuse of data, a “Data Management Plan” (“DMP”) can help participants to consider any issues or challenges that may be encountered and be useful for identifying solutions to overcome these. A DMP should be thought of as a “living” document outlining how the research data collected or generated will be handled during and after a research project: the DMP needs to be updated over the course of the project whenever significant changes arise, such as new data or changes in the consortium policies or composition. Thus, a DMP describes the data management life cycle for the data to be collected, processed and/or generated by a Horizon 2020 project. Projects participating in the Horizon 2020 Open Research Data Pilot are required to develop several versions of a DMP.61 The main information contained in a DMP are: how to handle research data during and after the end of the project; what data will be collected, processed and/or generated; which methodology and standards will be applied; whether data will be shared/made open access; how data will be curated and preserved (including after the end of the project). In this regard another important initiative deserves to be mentioned. A group of stakeholders from academia, industry, funding agencies and scientific publishers has worked on setting up a set of principles that can serve as guidelines for those looking to improve the reusability of the data they manage. These principles, defined in the FAIR acronym (Findable, Accessible,

58

See Open Access Publishing and Scholarly Societies—A Guide, open soC’y Inst., 2005, at 7. See Horizon 2020: Guidelines to the Rules on Open Access to Scientific Publications and Open Access to Research Data in Horizon 2020, eur. CoMM’n, available at http://ec.europa.eu/research/ participants/data/ref/h2020/grants_manual/hi/oa_pilot/h2020-hi-oa-pilot-guide_en.pdf (last visited Mar. 29, 2019). 60 See Zenodo, openaIre proJeCt, available at https://www.zenodo.org/ (last visited Mar. 29, 2019). 61 Guidelines on FAIR Data Management in Horizon 2020, eur. CoMM’n, July 2016, available at http://ec.europa.eu/research/participants/data/ref/h2020/grants_manual/hi/oa_pilot/h2020-hi-oa-data -mgt_en.pdf (last visited Oct. 17, 2019). 59

“Free data?”: open science in the age of personal data protection 407 Interoperable, Reusable),62 have been elaborated by the “Force11”63 group and recently adopted in the new European Commission guidelines on data management in projects funded by Horizon 2020.64 Since the Open Research Data Pilot provides that data should be made publicly available in a permanent way and without any kind of temporal restriction, it becomes immediately evident how this type of setting is clearly at odds with the principles and rights provided by the discipline on personal data protection (among all, the principles of purpose limitation and data minimization). This extensive use of personal data cannot be legitimized by the research exceptions described above with reference to the European regulatory framework (see Article 89 GDPR). The exception, indeed, guarantees that the strong data protection principles remain valid for the initial data collection. If the initial purpose of data collection is a scientific use, then the data subject shall have the possibility of consenting to the use of personal data, at least in certain areas of research or parts of research projects (Recital 33 of GDPR). Contrasts immediately spring to mind. The Pilot provides that the use of data is not, in any way, limited to specific scientific purposes and, even under the scientific research exception, the principle of data minimization is still applicable and therefore must be ensured by the application of appropriate safeguards. In light of these considerations, the deposit and making available of research data which include personal data in an open access repository cannot be legitimized by means of the research exemptions provided by GDPR. Moreover, the use of personal data cannot be legitimized by the consent of the data subject. The consent has to be specific and informed, with regards to the use of the data and the possible recipients. In an open data context, future uses as well as possible recipients are inherently not determinable; then it is simply impossible to properly inform the data subject in a declaration of consent and to address the requirements provided by the privacy regulation. Thus, consent cannot have legal effects. Even admitting the possibility of adopting the consent to legitimize the processing, it would then remain impossible to guarantee the exercise of the withdrawal of consent that determines the obligation to delete data and no longer process them. Within the Pilot, data are available in an open research data repository, without restrictions: there is no control on data. The only viable chance to guarantee compliance with the regulations on personal data protection while observing the requirements of the Open Research Data Pilot is to perform an effective anonymization of data to be opened up. Recital 26 provides a definition of anonymous data confirming that the Regulation is not applicable to this type of information:

62 Guiding Principles for Findable, Accessible, Interoperable and Re-usable Data Publishing, forCe 11, available at https://www.force11.org/fairprinciples (last visited Mar. 29, 2019). 63 Id. 64 See Horizon 2020 Program: Guidelines on FAIR Data Management in Horizon 2020 (version 3.0, July 26, 2016), available at http://ec.europa.eu/research/participants/data/ref/h2020/grants_manual/hi/ oa_pilot/h2020-hi-oa-data-mgt_en.pdf (last visited Oct. 17, 2019).

408 Research handbook on intellectual property and technology transfer The principles of data protection should therefore not apply to anonymous information, namely information which does not relate to an identified or identifiable natural person or to personal data rendered anonymous in such a manner that the data subject is not or no longer identifiable.65

It is not an easy task to determine in every situation when information can be defined anonymous. GDPR in Recital 26 adds standards to delimit a determination that can only remain on a case-by-case basis: To determine whether a person is identifiable, account should be taken of all the means that is reasonably likely to be used, such as singling out, either by the controller or by another person to identify the natural person directly or indirectly. To be ascertained whether it is reasonably likely to be used to identify the natural person, required for identification, taking into consideration the available technology at the time of processing and technological developments.

When considering the passage related to “all the means likely reasonably to be used,” one needs to take into account the following factors: the cost of performing the identification, the intended purpose, the way in which the processing is structured, the benefits expected by the data controller, the interests at stake for the individuals, the risks related to technical malfunction or organizational errors. Finally, it is a “dynamic concept:” the risks of re-identification may vary over time, due to technical progress.66 Therefore, it is advisable to proceed to open-access dissemination of only those data that are unidentifiable and non-personal: data that are, where needed, rendered anonymous or aggregated in such a way as not to leave (theoretically) any possibility of identifying the data subjects.

IV.

“SPECIFIC PURPOSE V. ANY PURPOSE”: HOW TO BALANCE THE OPEN APPROACH WITH DATA PROTECTION

When one deals with data, especially in the academic context, the activity carried out typically involves the collection, use, transfer, sharing, and storage of data. One is faced with problems that are essential for the correct implementation of a virtuous circle linked to the management and circulation of scientific data. The long-debated issue of the management of IPRs (patents for invention, copyright and sui generis law, trade secret and so on) and their correct circulation through assignments and licenses are added in an increasingly evident criticality related to the possible personal nature of data processed. On the one hand data, understood as mere information or as a mere fact, seem in principle removed from the exclusive control of IP, on the other even more extensive and stringent forms of exclusivity, such as the SGDR established

65

It is pivotal to not confuse the notion of anonymous information with that of pseudonymized data of which GDPR has provided a precise definition: “means the processing of personal data in such a manner that the personal data can no longer be attributed to a specific data subject without the use of additional information, provided that such additional information is kept separately and is subject to technical and organizational measures to ensure that the personal data are not attributed to an identified or identifiable natural person” (GDPR, art. 4(5)). Pseudonymized data remain personal data and therefore subject to compliance with the rules on the protection of personal data. 66 See Robert Gellman, The Deidentification Dilemma: A Legislative and Contractual Proposal, 21 fordhaM Intel. prop. MedIa & ent. l. J. 33 (2010).

“Free data?”: open science in the age of personal data protection 409 by European legislation, outline types of private control that heavily affect circulation and data sharing. Furthermore, the protection of personal data calls into question identified or identifiable people and poses problems of considerable importance in the field of data management. In this context, the “open” phenomenon takes place: the “terms” sometimes change according to contexts (Open Science, Open Data, Open Source, Open Access, and so on) but the goals are the same: that is to eliminate the (legal, technical, economic) barriers that stand between information and the public (private citizens, public administrations, businesses or other organizations) with reference to access and reuse of scientific research results. The logic of openness is gradually spreading in the scientific community and in the business sector. In a broader sense, this approach to the dissemination of knowledge expresses the free online availability of digital content and involves the whole idea of freely usable knowledge and creativity (in more general terms this is referred to as “knowledge transfer”). The scenario analysed in this Chapter represents a privileged context in which to study how to balance conflicting interests and rationales. To truly succeed in “freeing data,” when these refer to personal information, it is necessary to weigh the concerns involved, evaluate the objectives, guarantee the principles and the rights that the current discipline recognizes to individuals in order to protect, ultimately, their dignity. And the challenge is to do that, while fostering the realization of virtuous processes that, in some respects, only the “open” movements can nowadays guarantee. We must, first of all, consider that “data” must be correctly managed in all the phases of an activity that starts from a research phase and then continues in a phase of information dissemination (typical of knowledge transfer) and/or in commercial exploitation at the level of service offered to the public.67 To begin with, a risk-benefit analysis has to be conducted to inform the design and implementation of open data programs. The complexity of balancing utility and privacy in OD means that there is no “one size fits all” solution for any dataset: releasing data carries benefits for the public as well as potential risks to individual privacy, since the value that OD could yield outweighs the potential privacy risks of releasing those data. Secondly, controllers have to consider privacy at each stage of the data lifecycle. They have traditionally focused on privacy only when releasing open data, but effective privacy management requires taking it into account at all stages of a dataset’s lifecycle. Risks for personal data can emerge throughout the OD lifecycle of collection, maintenance, release and deletion. One should therefore carefully consider what data to collect and store, and not just what data to release. This requires adopting the principles and rules established in the data protection regulation from the collection phase, thus laying the “legitimate” basis for the activity concerning the processing of personal data that will eventually result in dissemination activities. Furthermore, it is pivotal to develop operational structures and processes that codify privacy management widely throughout the organization that is processing personal data. Organizations such as universities should develop clear and consistent data management processes to regularly evaluate the risks and benefits of releasing data. Internal awareness of and attention to privacy risks has to be increased.

67

See Ben Green, Ariel Ekblaw Cunningham, Paul Kominers, Andrew Linzer, & Susan Crawford, Open Data Privacy, berkMan kleIn Ctr for Internet & soC’y researCh publ’n, 2017, available at https://dash.harvard.edu/handle/1/30340010 (last visited Oct. 17, 2019).

410 Research handbook on intellectual property and technology transfer Finally, it is essential to invest some effort in including society in the processes related to OD in order to increase its involvement. This also means trying to incorporate public input into public data decisions and keep the public informed about new developments. Public engagement and public priorities have to be emphasized as essential aspects of a data management program. In conclusion, I turn to the previously described operational solution to the main issue. There is an evident conceptual and value contrast between the OD phenomenon, which stresses the importance of the widest openness, sharing and reuse, eliminating any type of (technical, juridical, economic) barrier, and the data protection regulation, which circumscribes and limits the processing of personal data only to the purposes identified, declared and then made known ex ante to the data subject. It is also clear that compliance with privacy principles cannot be guaranteed when a database is released following an OD approach, as the controller/holder of the database completely loses control over the information thus released and over possible further uses of data. This is precisely the mantra I have repeatedly cited of “any purpose vs. specific purpose.” The only truly trustworthy application solution to date is to use effective anonymization techniques that make the information released no longer connected to an identified or identifiable natural person. It is important to bear in mind that this is often a “relative” process, since absolute anonymization seems to have disappeared with the emergence of big data and technologies that exponentially increase the computational capacity of computers; and it is also intrinsically “dynamic,” as technologies evolve along with the risks of re-identification.68

68

See Sophie Stalla-Bourdillon & Alison Knight, Anonymous data v. personal data—a false debate: an EU perspective on anonymization, pseudonymization and personal data, 34 wIs. Int’l l. J. 284 (2017).

PART III GLOBAL PERSPECTIVES ON INTELLECUTAL PROPERTY AND TECHNOLOGY TRANSFER

19. A European perspective on intellectual property and technology transfer James A Cunningham, Marco Romano and Melita Nicotra

I.

INTRODUCTION

For the purposes of this Chapter we focused our attention mainly on technology transfer within the European Union (“EU”). The global recession in 2008 has economically and socially impacted EU Member States in different ways. For many firms, the focus has been on survival—i.e., ensuring that their firms do not fail—since this recession. Other firms have experienced weaker growth and lower levels of profitability. This has resulted in firms having less resources to focus on their own research and development. This has made them more risk adverse in developing new partnerships and collaborations with outside organizations, particularly universities and public research laboratories. It also has made them more cautious with investments with respect to intellectual property (“IP”). One of the key policy challenges among Member States has been on how best to recover from this recession ensuring a more stable and sustainable economic context. Consequently a key policy focus for the EU Commission and among Member States is focused on supporting and growing innovative activity across Europe in private for-profit firms, public organizations and services. In essence, the policy issue is how best to support the innovation capability within firms and one element of this is to encourage greater levels of collaboration between firms and knowledge-based organizations such as universities and public research laboratories. For firms this means that they have access to new knowledge and scientific discoveries that supports their efforts to sustain competitiveness and secure a sustainable competitive advantage in international markets. An element of encouraging such collaborations between industry and universities and public research laboratories is removing any barriers. One of these main barriers has centered around IP arrangements and parties’ rights that support different technology transfer mechanisms such as licences, spin-outs and material transfer agreements. This barrier can mean that firms are less inclined to enter such collaborations or to agree to the IP terms set by universities and public research organizations for technology transfer. Firms seek such collaborations that minimize the transaction costs and time and provide them with the maximum commercial scope to exploit the transferred technology. To overcome such barriers universities have standardized technology transfer and IP agreements for firms within different geographical terrorities. For example, in Ireland this has been in place for some years with the support and coordination of KT Ireland.1 With respect to technology transfer offices (“TTOs”) within the EU there has been a significant growth in their development. This in part reflects the evolving change of missions for 1 Knowledge Transfer Ireland, available at https://www.knowledgetransferireland.com/Model -Agreements/Catalogue-of-Model-Agreements/ (last visisted Mar. 29, 2019).

412

A European perspective on intellectual property and technology transfer 413 universities to incorporating a third mission centered on technology and knowledge transfer.2 In the UK for example, there has been a significant growth in enhancing and further strengthening TTOs within universities and this has resulted in different organizational arrangements.3 The growth in the EU of university-based TTOs also has led to an increase in patent portfolios owned by universities. Part of the remit of these TTOs is to be more proactive within their institutions in seeking out and exploiting IP that has been generated by faculty. Against this background our book Chapter is structured as follows: We begin the Chapter by providing an overview of European technology transfer. We then turn our attention to examining some of the empirical studies of European technology transfer and the value of university IP. We then focus on the impact of university IP on society and academic productivity and the cost of university IP. We conclude the Chapter by discussing some future avenues of research.

II.

EUROPEAN TECHNOLOGY TRANSFER

A.

IP Rights of Universities and Public Research Centers: Conflicting Views

IP is one of the channels through which knowledge flows from universities and research centers to industry, contributing to economic growth. Patents are the main form of IP rights used to produce new knowledge. The patent system is an important incentive for the production of knowledge but, on the otherhand, it limits the free circulation of knowledge throughout the economy. Actually, the patent is a legal monopoly for the use of a specific knowledge, which is assigned to the person or organization that first develops the knowledge and is willing to apply it to commercial end. All the advanced economies in the world have established a patent protection system of new knowledge, although there are still large differences in the exact rules under which inventions can be patented in various systems. Important elements of any patent system are that all patented knowledge must be made known through the patent document; that not all patent knowledge becomes the property of the patent holder; and that there is a maximum patent term, after which the patented knowledge becomes public property. The maximum duration of a patent in most patent systems is around 20 years. Therefore, the patent supports the further technological development of the area in which the patent is granted, and potentially benefits others different from the patent holder. Other researchers than those employed by the patent holder can use the knowledge described in the patent to develop new knowledge. However, not all knowledge is patentable. For example, in the European context, in the field of genetic research, a company cannot simply patent a genetic code as such, but must always write a specific application using the genetic code. Applications that are unknown at the time of patenting are not covered by the patent and are freely available for future research and commercial development. In contrast, in the United States patent system,

2

James Cunningham & Brian Harney, Strategic Management of Technology Transfer: the New Challenge on Campus (2006). 3 Alison F. Campbell, “How to set up a technology transfer office: experiences from Europe” in Intellectual Property Management in Health and Agricultural Innovation: A Handbook of Best Practices 559, 563–5. Anatole Krattiger, Richard T. Mahoney, Lita Nelsen, Jennifer A. Thomson, Alan B. Bennett, Kanikaram Satyanarayana, Stanley P. Kowalski, Gregory D. Graff, & Carlos Fernandez, “Intellectual Property Management” in Health and Agricultural Innovation: A Handbook of Best Practices (2007).

414 Research handbook on intellectual property and technology transfer patents on the genetic code are much broader and may include a wide range of applications not yet discovered. Such a wide patent obviously offers a higher incentive to the patent holder, but also limits the benefits for the whole society. Speaking of technology transfer, we cannot but refer to the patenting activity of universities and public research centers. Nelson and Arrow argue that it may not be convenient for private companies to invest in fundamental research and, as a result, this role is generally reserved to universities and other public research organizations.4 Consequently, firms can draw on the results of such research to carry out applied research or experimental development. However, the question that emerged for scholars over the years is whether it is right to offer an incentive to the creation and dissemination of knowledge, through patents, to institutions, such as universities and public research centers, which are financed by public funds just to create and disseminate knowledge.5 Despite the fact that academic patenting appears as a paradox to several researchers since 1980, with the approval of the Bayh-Dole Act in the US, several countries have promoted the institutionalization of university patenting.6 The Bayh-Dole law in the United States allows universities to patent inventions financed by federal funds and to maintain the copyrights that may be generated by the licenses of such patents. The main argument for university patents is that they facilitate technology transfer from universities to industry.7 There are several arguments in favor of patenting and commercializing research results of universities and public research centers. First, IP rights are needed for the creation of appropriate incentives for further investments in applied research by companies. If the university could not assign the exclusivity for the use of a patent to a company that will develop the applied knowledge, the company would not be encouraged to carry out such applied research if their competitors could enter the market using the university knowledge available for free. In the absence of the monopoly rent expected by the licensee, academic inventions would never find the way to the market.8 Second, supporting the commercialization of research results by universities and public research centers is the potential piracy of university discoveries. Without the patent protection of an innovation generated in the university, opportunistic companies could move to other countries to patent the knowledge originally developed by the university, limiting the applica-

4 See, e.g., Richard Nelson, The Simple Economics of Basic Scientific Research, 67 J. pol. eCon. 297 (1959); Kenneth J. Arrow, “Economic welfare and the Allocation of Resources for Invention” in The Rate and Direction of Inventive Activity: Economic and Social Factors (Richard Nelson ed., 1962). 5 Bart Verspagen, University Research, Intellectual Property Rights and European Innovation Systems, 20 J. of eCon. surv. 607 (2006). 6 David C. Mowery & Bhaven Sampat, University Patents and Patent Policy Debates in the USA, 10 Indus. & Corp. Change 781 (2001). 7 Edmund W. Kitch, The Nature and Function of the Patent System, 20 J. l. and eCon. 265 (1977); R.S. Eisenberg, Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research, 82 va. l. rev 1663 (1996); Martin Kenney & Donald Patton, Reconsidering the Bayh-Dole Act and the Current University Invention Ownership Model, 38 researCh polICy 1407 (2009); Albert N. Link, et al., An Empirical Analysis of the Propensity of Academics to Engage in Informal University Technology Transfer, 16 Indus. & Corp. Change 641 (2007). 8 Richard Jensen & Marie C. Thursby, Proofs And Prototypes For Sale: The Tale Of University Licensing, 91 aM. eCon. rev. 240 (2001); J. Colyvas, et al., How Do University Inventions Get into Practice?, 48 MgMt sCI. 61 (2002).

A European perspective on intellectual property and technology transfer 415 bility of a knowledge that was born as free.9 Thirdly, another argument put forward in favor of university patenting is the hypothesis of greater quality control on the application of university fundamental research. If a university holds a patent for a discovery, it can control which organization develops the discovery further and exclude applications with inferior quality or unethical applications. It is also considered that the patenting of universities overcomes the problems of lack of awareness of researchers-inventors on the marketability of their research results.10 TTOs can actively stimulate researchers to reveal their useful findings so that they can be patented. Finally, the IP of the universities can also generate financial advantages deriving from royalties.11 B.

Institutional Contexts in Europe and USA

As outlined previously, the first of a series of provisions on IP rights and university politics is the Bayh-Dole Act, approved by the US Congress in 1980, which grants to US universities IP rights that arise from research financed from federal funds.12 Since the introduction of the Bayh-Dole Act, the number of patents granted to US universities has grown.13 In a recent survey of the Association of University Technology Managers it is reported that the number of invention disclosures has been on the rise the previous five years, growing to 25,825 in 2016; 16,487 new US patent applications were filed, a gain of 3% over the prior year, and 7,021 US patents were issued, up 5% from the previous year; consumers and businesses benefited from 800 new products.14 The success of this law led many European governments to intervene on their own legislation on the IP of their universities.15 Among the first to emulate the US legislative innovations was the British government, which in 1985 abolished the exclusivity of BTG (British Technology Group, a government organization) on academic inventions and assigned to universities the right both to patent inventions of their scientists and to market them also through exclusive licenses.16 In the first years of the new millennium, other European countries modified the rules on patent ownership resulting from academic research in a direction consistent with the

9

Verspagen, supra note 5. Mowery & Sampat, supra note 6. 11 Verspagen, supra note 5. 12 David C. Mowery, et al., The Growth of Patenting and Licensing by US Universities: An Assessment of the Effects of the Bayh-Dole Act of 1980, 30 res. pol’y 99 (2001). 13 David C. Mowery, “America’s Industrial Resurgence: An Overview” in U.S. Industry in 2000: Studies in Competitive Performance 1 (Mowery ed., 1999); David C. Mowery, Richard R. Nelson, Bhaven Sampat, & Arvids Ziedonis, Ivory Tower and Industrial Innovation (2004); see Mowery & Sampat, supra note 6; David C. Mowery & Bharen N. Sampat, The Bayh-Dole Act of 1980 and University-Industry Technology Transfer: A Model for Other OECD Governments, 30 J. teCh. transfer 115 (2005); Richard R. Nelson, Observations on the Post-Bayh-Dole Rise of Patenting at American Universities, 26 J. of teCh. transfer 13 (2001); Bharen N. Sampat, et al., Changes in University Patent Quality After the Bayh-Dole Act: A Re-Examination, Int’l J. of Indus. org. 21, 1371 (2003). 14 Licensing Activity Survey: FY 2016, the assoCIatIon of teChnology Managers, InC. (2016). 15 Francesco Lissoni & Fabio Montobbio, Brevetti Universitari Ed Economia Della Ricerca in Italia, Europa E Stati Uniti. Una Rassegna Dell’evidenza Recente, 2 polItICa eConoMICa 259 (2006). 16 Maxine Clarke, British Technology Group—UK Technology Transfer Grows, nature 385 (1985); Henry Gee, Privatization Ahead for BTG, nature 272 (1991). 10

416 Research handbook on intellectual property and technology transfer spirit of the Bayh-Dole Act.17 As pointed out by Verspagen,18 these measures were in line with the indications of the European Commission which stated that “it is vital that knowledge flows from universities into business and society. The two main mechanisms through which the knowledge and expertise possessed and developed by universities can flow directly to industry are the licensing of university IP, and spin-off and start-up companies.19 Denmark, Germany, Austria, Norway, Finland, Slovenia, Slovakia, Hungary and the Czech Republic abolished the so-called “academic privilege” (which is still present in Sweden). It is a rule according to which university professors are entitled to IP rights on their discoveries.20 The main reason for the abolition of such privilege refers to the expensive administrative and transaction costs that the single researcher could face when protecting and commercializing the patent. Such costs can be better borne by the university’s TTO, which processes a broader patent portfolio and has its own fixed costs already covered by other activities. Moreover, TTOs typically have access or dedicated commercialization expertise that can support more effectively the commercialization of scientific discoveries. In the same years in which other European Countries provided for the abrogation of the academic privilege, the Italian legislator introduced this institute, together with new rules on patent taxes.21 In 2001, the Italian Financial Law attributed the ownership of academic inventions to the researchers themselves, and not to universities.22 The rationale of the law was to shift the economic incentive to commercialization of research results from university offices burdened by bureaucratic and organizational inefficiencies to the individual researcher, who is presumably more flexible and sensitive to the incentive.23 The academic privilege is currently present also in Sweden.24 In all other European countries included in a study by Geuna and Rossi, universities have the right to own patents on research discoveries.25 As argued by the OECD (2003), the situation in which individual researchers have the IP right on their inventions may not be optimal, because individuals are generally less able to exploit these rights than large institutions such as universities.26 When individual researchers own patents, the bargaining power of universities in public-private research joint ventures is very low and the risk of market failure following a “wrong” transfer of property rights

17

Ted Agres, Euros for Discoveries?, 16 sCIentIst 42 (2002); Turning Science into Business: Patenting and Licensing at Public Research Organizations, OECD (2003). 18 Versagen, supra note 5: Original Versagen (2006) 19 An Assessment of the Implications for Basic Genetic Engineering Research of Failure to Publish, or Late Publication of, Papers on Subjects Which Could be Patentable as Required Under Article 16(B) of Directive 98/44/EC on The Legal Protection of Biotechnological Inventions, european CoMMIssIon (2002), at 7. 20 Aldo Geuna & Federica Rossi, Changes To University IPR Regulations in Europe and the Impact on Academic Patenting, 40 res. pol’y 1068 (2011); Francesco Lissoni, et al., Academic Patenting in Europe: New Evidence from the KEINS Database, 17 res. evaluatIon (2008). 21 Nicola Baldini, et al., Institutional Changes and the Commercialization of Academic Knowledge: A Study of Italian Universities Patenting Activities Between 1965 and 2002, 35 res. pol’y 518 (2006). 22 Della Malva, et al., L’attività Brevettuale Dei Docenti Universitari: L’Italia In Un Confronto Internazionale, 2 eConoMIa e polItICa IndustrIale 43 (2007). 23 Margherita Balconi, et al., Ma Sui Brevetti Funziona Bene L’intesa Tra I Docenti Universitari E L’industria, Il Sole 24 Ore, 19 Gennaio (2002). 24 Geuna & Rossi, supra note 20. 25 Id. 26 OECD, supra note 17.

A European perspective on intellectual property and technology transfer 417 Table 19.1

Ownership of IPRs at universities in selected European countries Country

Institution

Austria

♦ (2002)

Belgium

♦ (1997/98)

Czech Republic

♦ (1990)

Denmark

♦ (2000)

Finland

♦ (2007/2010)

France

♦ (1982)

Germany

♦ (2002)

Greece

♦ (1995)

Hungary

♦ (2006)

Inventor

◊ ◊ ◊ ♦ (2001/2005)

Italy The Netherlands

♦ (1995)

Norway

♦ (2002)

Poland

♦ (2000)

Slovak Republic

♦ (2000)

Slovenia

♦ (2006)

Spain

♦ (1986)

◊

♦ (1949)

Sweden Switzerland

♦ (1911)

UK

♦ (1977/1985)

Notes: Inventor ownership is assigned on certain types of inventions. In brackets: years in which last change regulation took place.

becomes wider.27 From this perspective, the legal situation regarding the ownership of university patents in Italy and Sweden appears as inefficient. Finally, one of the major changes with European IP law is the creation and introduction of the EU Unified Patent Court. Currently in a European Union context each Member State has its own legal system and for example in the case of patent laws these are considered in the first instance in the Member State’s own judicial system, with recourse to the European Court of Justice on issues relevant to EU wide directives. With the introduction of the EU Unified Patent Court (“UPC”) there will be a significant change in the adjunction of patent law and cases in a European context. The focus of the UPC:28 will be a court common to Contracting Members States and thus part of their judicial system. It will have exclusive competence in respect of European patents and European patents with unitary effect … the UPC’s rulings will have effect in the territory of those Contracting Member States having ratified the UPC Agreement at the given time. The UP will not have competence with regard to national patents.

To date, all Members States of the EU except for Poland and Spain have signed the UPC Agreement and it is current in set up phase.

27

Verspagen, supra note 5. See Unified Patent Court, available at https://www.unified-patent-court.org (last visited Mar. 29, 2019). 28

418 Research handbook on intellectual property and technology transfer

III.

SOME EMPIRICAL EVIDENCE BASED ON STUDIES OF EUROPEAN TECHNOLOGY TRANSFER

There has been a growth in scholars conducting research on different aspects of technology transfer using European data. In the main these studies have focused on specific country contexts with some studies using cross-country data based on primary or secondary data sources. With the various regional differences within Europe, conducting research on technology transfer can be challenging but also can yield further valuable insights into differences aspects of technology transfer.29 A.

Technology Transfer Offices

In the US, over the past 20 years there has been a substantial increase in patents and licenses and in the number of TTOs. The US has registered a continuous growth of university patents, licenses and disclosure of inventions.30 At the same time, numerous researches in the US have analyzed the nature and the determinants of this greater licensing activity, as well as the nature, growth, profitability of TTO activities and the determinants of their efficiency.31 In Europe, the role of TTOs has been much less studied from an empirical point of view, first, because there is a very high heterogeneity among countries and also because there is no systematic and extensive collection of data as in the United States with the AUTM investigations. An exception is provided by the survey developed by the Association of Science and Technology Professionals (“ASTP”) on technology transfer activities in Europe. In Italy for example, Baldini et al. in 2006 showed how universities started to adopt patent policies and regulations only from the mid-1990s.32 Bach et al. and Conti and Gaule at the same time showed that in Belgium, Denmark and France, TTOs were small, with a very limited number of staff members, and that licensing activity was not so developed as in the United States.33 Conti and Gaule (2011) demonstrated that the distribution of licenses was distorted

29 See Simon Mosey, et al., Technology Entrepreneurship Research Opportunities: Insights From Across Europe, 42 J. of teCh. transfer. 1 (2017). 30 Licensing Activity Survey: FY 2016, supra note 13. 31 Saul Lach &. Mark Schankerman, Incentives and Innovation in Universities, 39 rand J. of eCon. 403 (2008); Dante Di Gregorio & Scott Shane, Why Do Some Universities Generate More Start-Ups than Others?, 32 res. pol’y 209 (2003); Jensen & Thursby supra note 8; Jerry G. Thursby & Marie C. Thursby, Are Faculties Critical? Their Role in University-Industry Licensing, 22 ConteMp. eCon. pol’y 162 (2004); Jerry G. Thursby & Sukanya Kemp, Growth and Productive Efficiency of University Intellectual Property Licensing, 31 res. pol’y 109 (2002); Toby E. Stuart & Waverly W. Ding, When Do Scientists Become Entrepreneurs? The Social Structural Antecedents of Commercial Activity in the Academic Life Sciences, 112 aM. J. of soCIo. 97 (2006); Ajay Agrawal, Engaging the Inventor: Exploring Licensing Strategies for University Inventions and the Role of Latent Knowledge, 27 strategIC MgMt J. 63 (2006). 32 Nicola Baldini, et al., Institutional Changes and the Commercialization of Academic Knowledge: A Study of Italian Universities Patenting Activities Between 1965 And 2002, 35 res. pol’y 518 (2006). 33 Annamarie Conti & Patrick Gaule, Is the US outperforming Europe in university technology licensing? A new perspective on the European Paradox, 40 res. pol’y 123 (2011); Laurent Bach, Nicolas Carayol, & Patrick Llerena, Assessing the Performance of French University Transfer Offices: Preliminary Results, International JA Schumpeter Society Conference: “Innovation, Competition and Growth: Schumpeterian perspectives,” Nice–Sophia Antipolis, 21–24 June 2006.

A European perspective on intellectual property and technology transfer 419 and many institutions had very few licenses.34 Moreover, the OECD in 2003 emphasized that most of the public research centers assigned a very limited number of licenses each year and that, in Italy, in the Netherlands and in Switzerland, an important part of the licensing agreements was concluded for patents pending or unpatented inventions as well as for materials protected by copyright.35 Chapple et al. (2005) noted that TTOs were more developed in UK; however, in absolute terms, they were characterized by low levels of efficiency.36 The authors stressed the need to improve the skills of TTO managers and licensors in the UK. Again with reference to the efficiency of TTOs, Conti and Gaule (2011) found that the skills of TTO members play an important role in determining productivity.37 In particular, they claimed that the use of PhDs seems to reduce the costs of coordination resulting from interactions between TTOs and academic researchers. The most recent data relating to the technology transfer activities of universities in Europe derive from the ASTP-Proton Annual Knowledge Transfer Survey. In their latest report for the year 2015, they collected and analyzed a large set of data, demonstrating a great diversity in the volume of knowledge transfer activities in Europe. This is done by the fact that TTOs have been active for some time in some countries, while they are relatively new in others. ASTP-Proton have identified an aggregate number of 1,924 professionals involved in technology transfer, for a total of 410 institutions. The management of IP is a key activity. On average, a European TTO received 29 reports of invention, submitted 12 priority patent applications and obtained 9 patents in 2015. However, almost 24% of the KTOs interviewed reported that no patents were granted in 2015. One strategy that seems to be quite common among TTOs is to abandon the patenting process by the end of the year or after completion of the patent procedure if no commercial partner has been identified. This is due in part to resource constraints within TTOs and the prioritization of resources against demands and key performance targets/ metrics. The vast majority (92%) of TTOs have at least some patents or patent applications in their portfolio and on average 21% of patents are licensed or optioned. These numbers show that European TTOs actively manage their IP and suggest that they have been reasonably successful in attracting commercial interests in the application of their IP. With regard to contracts and license income, most of the organizations interviewed have entered into contract research agreements, consulting agreements, and license agreements. Counsulting seems to be the most common type of interaction with industry, especially in the UK. Some 8% of the analyzed institutions seems to be very active in this field (>500 consultancy contracts in the reference year). In terms of income, the contract research stands at over €1.1 billion. Interviewed organizations reported 18,400 licenses executed, including 9,248 software licenses. Finally, spin-off training is a constant activity for the European TTOs, with over half of the organizations having founded at least one spin-off in the year of reference. However, the reported amounts of cashed-in equity were negligible for the vast majority (88%) of respondents.

34

Id. OECD, supra note 17. 36 Wendy Chapple, Andy Lockett, Donald Siegel, & Mike Wright, Assessing the Relative Performance of UK University Technology Transfer Offices: Parametric and Non-Parametric Evidence, 34 res. pol’y 369 (2005). 37 Conti & Gaule, supra note 33. 35

420 Research handbook on intellectual property and technology transfer B.

Patents

The Bayh-Dole Act in the US has been the focus of much empirical attention and this has seen a resultant increase in the numbers of US universities patenting. The situation in Europe is different. No single European third level institution can match the patent portfolios of large institutions such as MIT or Stanford universities.38 This difference is best captured by Lissoni et al. (2008) who note: This contrast between the USA and Europe has often been interpreted in the light of the general view of the existence of a “European Paradox,” according to the which European countries have a strong science base, but also many problems in translating scientific advances into commercially viable new technologies … Thus, the scarcity of university patents is seen both a signal of technology transfer deficit and a problem to be addressed through legislation.39

In a European context legal ownership—“professors’ privilege”—the balance between public research organizations (PROs) and universities and the lack of autonomy and IP skills and capabilities have been identified as factors that have prevented such a growth in European university patenting.40 For example, CNRT in Italy and VTT in Finland hold patents for academic inventors and the “professor privilege” means that individual academics are less likely to patent given the costs and lack of incentives to puruse this technology transfer mechanism. Increasingly, there has been a change to the professor privilege principle as outlined earlier in the Chapter, where institutions register, exploit and hold academic inventor patents. In essence the institution holds the property rights to intellectual capital, not the individual academic. Typically, the patenting of invention is formally organized through TTOs and there is a revenue share model where the academic inventor financially benefits from patent exploitation depending on the deal negotiated with the TTO and a third party. For example, in Germany the German Employees’ Inventions Act in 2002 changed the professor privilege and meant that individual academics no longer hold the property rights to their own inventions. There has been limited research undertaken about this major reform. For very productive inventors it generated significant financial rewards as reported by Harhoff and Hoisl (2007) but Von Proff et al. (2012) reported no significant increase in patenting after this reform.41 With respect to European academic patenting there has been a slow transfer of IP rights from the individual faculty member to the institution (their employer) with changes in national policies etc. in different European countries. This has resulted in universities setting up TTO to support academic patenting and other technology transfer mechanisms, which are discussed later on in this Chapter. Studies have focused on the impact these changes have in Europe in comparison to other countries. In a cross-country study of Europe, Japan and the US patent

38

OECD, supra note 17. Lissoni et al., supra note 20. 40 Chiara Franzoni & Francesco Lissoni, Academic Entrepreneurship, Patents and Spin-Offs: Critical Issues and Lessons for Europe, unIversItà CoMMerCIale luIgI boCConI 1 (2006). 41 Dietmar Harhoff & Karin Hoisl, Institutionalized Incentives For Ingenuity—Patent Value and the German Employees’ Inventions Act, 36 res. pol’y 1143 (2007); Sidonia Von Proff, et al., University Patenting in Germany Before and After 2002: What Role did the Professors’ Privilege Play?, 19 Indus, and InnovatIon 23 (2012). 39

A European perspective on intellectual property and technology transfer 421 citations over 20 years at the European Patent Office, Bacchiocchi and Montobbio (2009) conclude: university and PROs patents are more important than corporate patents … [T]here is not strong evidence in Europe that universities and PROs produce relatively more important inventions and if this occurs it is only in some specific fields. Moreover different countries in Europe display different pattern of diffusion and obsolescence of knowledge produced by universities and PROs.42

They also warn of the dangers of imitating US polices with respect to patenting in a European context particularly the Bayh-Dole Act given the national and institutional variations in a European context. While there has been an acknowledged increase in universities holding patents in Europe as a result of changes in property rights, this has not resulted in higher use and higher quality patents.43 Another large scale European study of patents and inventors found that there is limited job mobility among inventors, female participation is extremely low, personal and social rewards are important personal motivators for inventors and two third of patents reported in this study are team developed. The EU also has a larger share of unused patents than the US based on large scale inter-country studies by Torrisi et al.,44 and they suggest that a reason for this is the US market is more well developed than the European market. In assessing the knowledge spillovers using patent citation analysis Maurseth and Verspagen45 found that industry sectors and specialisms, geographical proximity and linguistics are considerations in overcoming barriers to knowledge transfer in the European content. With respect to patenting versus secrecy in a European context, Arundel (2001) using the European Community Innovation Survey concluded: The results of the analyses show that a higher percentage of R&D performing firms of all size classes find secrecy to be a more effective means of appropriation than patents … Furthermore, small firms, on average, do not rely more on patents than on secrecy in comparison with large firms.46

The differences between European and US patents citations are a result of institutional and cultural differences. In an extensive study of European patents and how firms decide to licence from patents, Gambardella et al. (2007) conclude that: The probability of licensing is higher in the case of greater protection and more codifed or general knowledge. It is also higher when there are more potential technology suppliers, and when the patent is of greater economic value.

42 Emanuele Bacchiocchi & Fabio Montobbio, Knowledge Diffusion From University and Public Research. A Comparison Between US, Japan and Europe Using Patent Citations, 1 J. teCh. transfer 169 (2009). 43 Geuna & Rossi, supra note 20. 44 Salvatore Torrisi, et al., Blocking and Sleeping Patents: Empirical Evidence from a Large-Scale Inventor Survey, 45 res. pol’y 1374 (2016). 45 Per Botolf Maurseth & Bart Verspagen, Knowledge Spillovers in Europe: A Patent Citations Analysis, 104 sCandInavIan J. of eCon. 531 (2002). 46 Anthony Arundel, The Relative Effectiveness Of Patents And Secrecy For Appropriation, 30 res. pol’y 611 (2001) (emphasis added).

422 Research handbook on intellectual property and technology transfer There has been very little research in Europe on patent assertion entities (“PAEs”). With changes forthcoming with UPC this is a significant change in the European patent market place.47 PAEs are more established in the US given the conducive environment. However, this activity is growing in the European context, particularly driven by US-based entities. The growth of PAEs more generally is influenced by the complexities of technologies and their evolving business model. Typically, European patent costs are lower as well as infringement damages and the relatively small size of the European market are some of the differences with the US system. In summing up the future of the PAE in a European context, Thumm (2018) notes: In Europe, the legal fragmentation of patent protection with national patents and under the European Patent Convention dis-incentivises PAEs from carrying out assertion activity on a pan-European scale. The introduction of the Unitary Patent (“UP”) and Unified Patent Court (“UPC”) is likely to modify the incentive structure for PAEs in Europe. Both the UP and the UPC have been described as game-changing events that could increase the amount of patent assertion activity in Europe. The main goals of the UP and the UPC to reduce the cost of maintaining and enforcing patent rights could play directly into the hands of PAEs.48

C.

Patent Activity in Europe

Starting from such antecedents, in Europe there has been an increase in university patent activity. It is difficult to find specific European statistics on this topic.49 An effort in this direction was made several years ago, in parallel with the new legislative scenarios in European countries. Reference is made to Bacchiocchi and Montobbio (2009), Lissoni et al. (2008); Breschi et al., (2007).50 These studies generally showed a positive trend in the growth of the university’s patent activity in several European countries, but at the same time they underlined that such numbers were still very low in absolute value, both with respect to the patent activity of European public research centers and to the patent activity of US universities. More recently, other studies have attempted to reconstruct the statistical data on patenting activity in European universities, considering both one country and more countries.51 Acosta et al. (2018) surveyed European Patent Office patents and assigned to European universities patents that appeared in the list of applicants.52 When there was more than one university applicant, each was assigned a patent. Their research resulted in 3,330 European university-owned patents related to 391 universities located in Europe for the period from 2001 to 2004. Of the 391 universities, 31 have no university-owned patents; 192 universities have between one and five such patents; 89 have between six and 10; 37 have 11–20; 18 have 21–30; and 24

47

Brian J. Love, et al., “Patent Assertion Entities in Europe” in Patent Assertion Entities and Competition Policy (Daniel Sokol ed., 2017). 48 Nikolaus Thumm, The Good, The Bad and the Ugly—The Future of Patent Assertion Entities in Europe, teCh. analysIs & strategIC MgMt (Feb. 2018). 49 Geuna & Rossi, supra note 20. 50 Bacchiocchi & Montobbio, supra note 42; Lissoni, et al., supra note 20; Stefano Breschi, The Scientific Productivity of Academic Inventors: New Evidence from Italian Data, 16 eCon. InnovatIon new teCh. 101 (2007). 51 Geuna & Rossi, supra note 20; Manuel Acosta, et al., Does Technological Diversification Spur University Patenting?, 43 J. teCh. transfer 96 (2018). 52 See Acosta, supra note 51.

A European perspective on intellectual property and technology transfer 423 universities own more than 30 patents. On average, there are 8.5 patents per university for the whole period. The recent statistics of “ASTP-Proton Knowledge Transfer Europe” can be found in the Annual Knowledge Transfer Survey. From the last available report, referring to the year 2015, realized on the interviews of 410 institutions in Europe, it emerges that the majority of universities (92%) have at least some patents or patent applications in their portfolio, and on average 21% of patents are licensed or optioned. Interviewed organizations reported 18,400 licenses executed, including 9,248 software licenses. ASTP-Proton also notes a great diversity in the volume of technology transfer activities in Europe. This is not surprising, given that TTOs have been active for some time in some countries and they are relatively new in other countries. In general, the reason for the low number of patents compared to the US is assigned to the lower awareness of European researchers of the applicability and commercialization of fundamental research.53 Mowery and Sampat argue that in the US, the major interest of universities in the field of patents preceded the introduction of the Bayh-Dole Act, corresponding to a greater awareness of the opportunities for applying fundamental research to commercial end.54 Unfortunately, in Europe the empirical basis for comparing this awareness among university researchers in different countries is very weak. However, Verspagen55 refers to Porter (2001) who, in his “Innovation Lecture” held at the Dutch Ministry of Economic Affairs, stated: “universities in the Netherlands have traditionally been much less commercially oriented than in the US, where commercial activities are more highly regarded. In the Netherlands, universities have little contact with companies, and filing patents and seeking to license technology to the private sector is not part of the culture.” Porter also presented empirical evidence on this.56 It is also true that the university patenting activity in Europe is likely to be distorted by the fact that often, in many European countries, private companies own IP rights over the output of university research. Consequently, the technological activities of European universities are underestimates.57 In this sense, Lissoni et al. (2008) searched university professors among inventors of patents published by private companies and showed that the share of patents with university inventors on the total number of patents in the considered countries was between 3% and 6%.58 Authors concluded that European universities’ contribution to patenting is not

53

Mowery & Sampat, supra note 6. See id. 55 See Versagen, supra note 5; Michael Porter, Innovation and competitiveness: findings on the Netherlands. Den Haag: Ministry of Economic Affairs 37 (2001). 56 Porter, supra note 55. 57 See Lissoni, supra note 20; Francesco Lissoni, et al., Ownership And Impact Of European University Patents, Paper, presented at the thIrd dIMe-brICk ConferenCe, torIno Apr. 23–24, 2010; Eric J. Iversen, et al., A Baseline For The Impact Of Academic Patenting Legislation in Norway, 70 sCIentoMetrICs 393 (2007); Paola Giuri, et al., Inventors and Invention Processes in Europe: Results from the Patval-EU Survey, 36 res. pol’y 1107 (2007); Margherita Balconi, et al., Networks Of Inventors And The Role Of Academia: An Exploration Of Italian Data, 33 res. pol’y 127 (2004); Manfred Schmiemann & Jean-noël Durvy, New Approaches to Technology Transfer from Publicly Funded Research, 28 J. teCh. transfer 9 (2003); Thomas Gering & Ulrich Schmoch, “Management of Intellectual Assets by German Public Research Organizations” in Turning Science into Business: Patenting and Licensing at Public Research Organizations (2003). 58 Lissoni, supra note 20. 54

424 Research handbook on intellectual property and technology transfer much inferior, in terms of percentages over national activity, to that of their US counterparts. In the United States, commercial companies own only 24% of US academic patents, while in Europe this percentage reaches 60% in France, 72% in Italy and 81% in Sweden. Geuna and Nesta (2006) cite evidence from Italy, Finland and Belgium that the number of university patents is seriously underestimated when only patents with universities as applicants are counted.59 They refer to the studies of Balconi et al. (2004) for Italy, to Saragossi and Van Pottelsberghe (2003) for Belgium, and to Meyer (2003) for Finland.60 D.

Material Transfer Agreements

Empirical studies of material transfer agreement in a European context are limited in comparison to other forms of technology transfer mechanisms in US research.61 In scientific fields such as biotechnology they are more dependent on the nature environment and this has been the focus of MTA research and EU policies. For example, an EU (2004) report identified some challenges with respect to MTA for publicly funded research and they were publications rights, commercialization exploitation (particularly future) and ownership of research results. In reflecting on the US experience and how the EU could learn, Rodriguez (2008) recommends that MTAs should be part of research policies and notes the potential benefits for the EU as: “The fact that the EU is a supranational state would not make a significant difference because the vested interests of users and owners of research materials must be served at a national level as well. The main focus of such regulation should be right enforcement, because it is more favourable to society to have clear rules in science and technology.”62 For researchers a key question is whether being involved in MTA agreements enhances research visibility and ultimately career prospects. Using Belgian data, Rodriguez et al. (2008) found that: “researchers that availed themselves of these contracts were more visible compared to those who did not use them, controlling for seniority and co-authorship.”63

IV.

THE VALUE OF UNIVERSITY INTELLECTUAL PROPERTIES: SOME EMPIRICAL INSIGHTS

There are limited European studies focused on the characteristics of university patents and their relative value. At the US level, Henderson et al. (1998) compared the universe of univer-

59 Aldo Geuna & Lionel Nesta, University Patenting And Its Effects On Academic Research: The Emerging European Evidence, res. pol’y 790 (2006). 60 Balconi, et al., supra note 57; Sarina Saragossi & Van Pottelsberghe De La Potterie, What Patent Data Reveal About Universities: The Case of Belgium, 18 J. teCh. transfer 47 (2003); Martin Meyer, Academic Patents as an Indicator of Useful University Research?, 12 res. evaluatIon 17 (2003). 61 See Alvin L. Kwiram, et al., University-Industry Consortium Agreements Center For Process Analytical Chemistry: A Case Study, 20 J. teCh. transfer 45 (1995); David C. Mowery & Arvids A. Ziedonis, Academic Patents and Materials Transfer Agreements: Substitutes or Complements?, 32 J. teCh. transfer 157 (2007). 62 Victor Rodriguez, Governance of Material Transfer Agreements, 30 teCh. In soC’y 122 (2008). 63 Victor Rodriguez, et al., On Material Transfer Agreements and Visibility of Researchers in Biotechnology, 2 J. of InforMetrICs 89 (2008).

A European perspective on intellectual property and technology transfer 425 sity patents between 1965 and 1992 with a random control sample of corporate patents.64 They measured the importance and generality of university patents. High generality indicates that the patent has a widespread impact and has influenced the subsequent innovations in different fields of technology. The generality was measured using the number of technological classes to which the citing patents belong and the importance was measured by counting the citations received from each patent. They demonstrated that university patents were more general and important throughout the period. The authors concluded that after the Bayh-Dole Act, there was an increase in the propensity to patent, but fewer general and important inventions were produced. Other authors have also studied extensively and empirically the characteristics of American university patents.65 Their researches were mainly focused on the effects of the Bayh-Dole Act on the quality of the university patents. At the European level, Sapsalis et al. (2006) and Sapsalis and Van Pottelsberghe (2007) examined the academic patents held by six Belgian universities and found that their citations did not differ substantially from that of a sample of corporate patent control.66 Bacchiocchi and Montobbio (2009) compared academic patents belonging to US, Japanese and European universities and found evidence of a citation premium compared to corporate ones.67 However, the result was valid only for US universities, while patents owned by European and Japanese universities did not seem to be more cited than average. Considering academic patents as having at least one university professor among the inventors, other studies on European academic patents have explored if the type of ownership (public or private) of academic patents is related to their quality. At a theoretical level, Aghion and Tirole (1994) argue that in the case of joint research projects between universities and private companies, the assignment of patents to the company rather than to the university can lead to market failure, so to a lower value of innovation. Czarnitzki et al. (2011) empirically analyzed the German case, confirming the hypothesis of the “citation premium” of the academic patents on the company patents.68 However, the citation premium appeared to be higher for academic patents owned by universities and professionals, compared to those owned by companies, and decreased over time. Conversely, a study set in the United Kingdom finds a quality premium for academic patents owned by private companies in the short and medium term (up to six years after the pri-

64

Rebecca Henderson, et al., Universities As A Source Of Commercial Technology: A Detailed Analysis Of University Patenting 1965–1988, 80 revIew of eConoMICs and statIstICs 119 (1998). 65 Sampat, supra note 13; David C. Mowery & Arrids A. Ziedonis, Academic Patent Quality and Quantity Before and After the Bayh-Dole Act in The United States, 31 res. pol’y 399 (2002); Jerry G. Thursby, et al., US Faculty Patenting: Inside and Outside the University, 38 res. pol’y 14 (2009); Bharen N. Sampat, Patenting and US Academic Research in the 20th Century: The World Before and After Bayh-Dole, 35 res. pol’y 772 (2006); Carlos Rosell & Ajay Agrawal, Have University Knowledge Flows Narrowed?: Evidence From Patent Data, 38 res. pol’y 1 (2009). 66 Elefthérios Sapsalis, et al., Academic Versus Industry Patenting: An In-Depth Analysis of What Determines Patent Value, 35 res. pol’y 1631 (2006); Elefthérios Sapsalis & Bruna Van Pottelsberghe De La Potterie, The Institutional Sources of Knowledge and the Value of Academic Patents, 16 eCon. of InnovatIon and new teCh. 139 (2007). 67 Bacchiocchi & Montobbio, supra note 42. 68 Dirk Czarnitzki, et al., Commercializing Academic Research: The Quality of Faculty Patenting, 20 Indus. and Corp. Change 1403 (2011)

426 Research handbook on intellectual property and technology transfer ority year of patents) compared to the academic patents owned by universities.69 No difference appeared in citations over six years after the priority year of the patent. Similarly, Crespi et al. (2010), analyzing a sample of academic patents from the PatVal database, provided information on the financial value and the nature (applied or fundamental) of inventions.70 They concluded that university-owned academic patents do not differ significantly, for both the two dimensions, from the company-owned academic patents. Finally, Lissoni and Montobbio (2015) compared the value and impact of academic patents in five European countries with different institutional frameworks: Denmark, France, Italy, the Netherlands and Sweden.71 Authors found most academic patents were assigned to business companies, followed by universities, public research organizations, and individual inventors. In general, company-owned academic patents tend to be cited as non-academic, while university-owned patents tend to be less cited. However, the results change from one country to another based on the different autonomy enjoyed by universities in the considered country.

V.

IMPACT OF UNIVERSITY IP ON SOCIETY AND ACADEMIC PRODUCTIVITY

The positive effects of university patents on society derive from their direct contribution to the industry through the granting of licenses,72 from the flow of knowledge to the territory that derives from conversations or informal meetings with inventors,73 and from the creation of spin-off companies based on patented innovations that increase the competitiveness of territories.74 Patents also have positive consequences for the universities themselves. They have great potential as a source of income from licenses75 and for participating in spin-offs that exploit 69 Valerio Sterzi, Patent Quality and Ownership: An Analysis of UK Faculty Patenting, 42 res. pol’y 564 (2013). 70 Gustavo Crespi, et al., University IPRs and Knowledge Transfer. Is University Ownership More Efficient?, 19 eCon. of InnovatIon and new teCh. 627 (2010). 71 Francesco Lissoni & Fabio Montobbio, The Ownership of Academic Patents and their Impact. Evidence from Five European Countries, 66 revue eConoMIQue 1143 (2005). 72 Janet Bercovitz & Maryann P. Feldmann, Entrepreneurial Universities and Technology Transfer: A Conceptual Framework for Understanding Knowledge-based Economic Development, 31 J. teCh. transfer 175 (2006); Bo Carlsson, Zolton J. Acs, David B. Audretsch, & Pontus Braunerhjel, Knowledge Creation, Entrepreneurship, and Economic Growth: A Historical Review, 18 IndustrIal and Corporate Change 1193 (2009). 73 See, e.g., Dante Di Gregorio & Scott Shane, Why Do Some Universities Generate More Start-Ups than Others?, 32 res. pol’y 209 (2003); Douglas Woodward, Octario Figueiredo, & Paulo Guimaraes, Beyond the Silicon Valley: University R&D and High-technology Location, 60 J. of urban eCon. 15 (2006); David B. Audretsch & Erik E. Lehmann, Does the Knowledge Spillover Theory of Entrepreneurship Hold for Regions?, 34 res. pol’y 1191 (2005); Andrea Bonaccorsi, Massimo G. Colombo, Massimiliano Guerini, & Cristina Rossi-Lamastra, The Impact of Local and External University Knowledge on the Creation Knowledge-Intensive Firms: Evidence from the Italian Case, 43 sMall bus. eCon. 261 (2014). 74 Vittorio Chiesa & Andrea Piccaluga, Exploitation and Diffusion of Public Research: The Case of Academic Spin-Off Companies In Italy, 30 r&d MgMt 329 (2000). 75 Link, et al., supra note 7; Donald S. Siegel, et al., Toward a Model of the Effective Transfer Of Scientific Knowledge from Academicians To Practitioners: Qualitative Evidence from the Commercialization of University Technologies, 21 J. of engIneerIng and teCh. MgMt 115 (2004).

A European perspective on intellectual property and technology transfer 427 the results of patents.76 Moreover, in some contexts, patenting researchers can tap into technological knowledge to develop new ideas for research in related scientific fields. Concerning the impact of patents on university and academic research, several studies have focused on the relationship between academic patents and publications, both from a quantitative and qualitative point of view. It is easy to hypothesize that the opportunity to benefit from the IP rights of universities could have the effect of redirecting researchers’ attention from fundamental research to technological applications problems. This could result in less ambitious publications with less impact on the scientific community. For example, it has been argued by Henderson et al. (1998) that applied research offers opportunities for rapid patents and that universities would have an incentive to abandon fundamental research in favor of applied research.77 However, numerous empirical studies have not found evidence of such concerns. In the US context, Agrawal and Henderson (2002) analyzed the patenting and publication behavior at the Department of Mechanical and Electrical Engineering at MIT (68 interviews) and demonstrated that there is no evidence of a trade-off between patents and publications.78 Azoulay et al. (2007) used a panel of 3,862 scientists in the life sciences and found no evidence of a negative effect of patents on the quantity and quality of publications.79 Markiewicz and DiMinin (2004) found complementarity between patenting and publication by comparing a panel of 150 randomly selected academic inventors at the USPTO and a control group of 150 scientists who were not inventors.80 Stephan et al. (2007) used about 10,000 scientists and found that patents are positively and significantly related to the number of publications.81 At the European level, this theme has been investigated, among others,82 by Breschi et al. (2007, 2008), Calderini et al. (2007); Carayol (2007); Van Looy et al. (2011); Guldbrandsen and Smeby (2005); and Ranga (2003).83 Overall, evidence from such studies indicates that high-quality university scientists are also active in patenting; patenting can be preceded by a higher than normal scientific productivity or can be followed by a flurry of publications. Breschi et al. (2007) investigate the scientific productivity of Italian academic inventors, using the database of the European Patent Office, 1978–1999.84 They match data of 299 academic 76 Sigrid Sterckx, Patenting and Licensing of University Research: Promoting Innovation or Undermining Academic Values?, 17 sCI. and engIneerIng ethICs 64 (2011). 77 Henderson, et al., supra note 64. 78 Ajay Agrawal & Rebecca Henderson, Putting Patents in Context: Exploring Knowledge Transfer at MIT, 48 MgMt sCI. 44 (2002). 79 Pierre Azoulay, et al., The Determinants of Faculty Patenting Behavior: Demographics or Opportunities?, 63 J. of eCon. behav. & org. 599 (2007). 80 Kira Fabrizio & Alberto Di Minin, Commercializing the Laboratory: The Relationship Between Faculty Patenting and Publishing, haas sChool of busIness, mimeo (2004). 81 Paula E. Stephan, et al., Who’s Patenting in the University? Evidence from a Survey of Doctorate Recipients, 16 eCon. of InnovatIon and new teCh 71 (2007). 82 Marina Ranga, The Exploration-Exploitation Dichotomy and the Impact of Environment Dynamics on University-Industry Partnerships, dphIl thesIs spru, University of Sussex (2003). 83 Breschi, supra note 50; Stefano Breschi, et al., University Patenting and Scientific Productivity: A Quantitative Study of Italian Academic Inventors, 2 eu. MgMt rev. 91 (2008); Nicolas Carayol, Academic Incentives, Research Organization and Patenting at a Large French University, 16 eCon. of InnovatIon and new teCh. 119 (2007); Mario Calderini, et al., If Star Scientists do not Patent: the Effect of Productivity, Basicness and Impact on the Decision to Patent in the Academic World, 36 res. pol’y 303 (2007). 84 Antonio Della Malva, Stefano Breschi, Francesco Lissoni, & Fabio Montobbio, L’attività Brevettuale Dei Docenti Universitari: L’Italia In Un Confronto Internazionale, 34 eConoMIa e polItICa IndustrIale (2007).

428 Research handbook on intellectual property and technology transfer inventors, with an equal number of non-patenting researchers. They find no trade-off between publishing and patenting or fundamental and applied research and find instead a strong and positive relationship between patenting and publishing. They suggest that it is not patenting per se that boosts scientific productivity, but rather the advantage derived from solid links with industry, as the strongest correlation between publishing and patenting activity is found when patents are owned by business partners, rather than individual scientists or their universities. Besides, Breschi et al. (2008) matched a sample of 592 Italian academic inventors and controls. Also in this case they found academic inventors publish more and better quality papers (basing on citations) than their non-patenting colleagues, and increase their productivity after patenting.85 Calderini et al. (2007), analyzing a sample of Italian researchers in the field of materials science, find that scientists who deal with applied research find it easier to produce industrial applications than their colleagues engaged in fundamental research.86 The authors interpret their results suggesting that, for the former, more academic research leads to more exploitable results, hence greater possibilities for patenting; for the latter, more academic research makes it more unlikely to find time to produce industrial applications. Van Looy et al. (2011) reported a positive relationship between scientific productivity and patenting activity in a sample of European universities. Furthemore, Crespi et al. (2011) indicate that in the United Kingdom academic patenting is complementary to publishing at least up to a certain level of patenting output, after which they found some evidence of a substitution effect.87 Referring to the trade-off between basic research and applied research, Gulbrandsen and Smeby (2005), in a survey among members of the university departments in Norway, found less involvement in fundamental research by the members of departments partly financed with external industrial funds. Firstly, researchers financed by industry perform a substantially lower basic research compared to other researchers without external funding or with external funding of a different type from the industrial ones; however, researchers with industrial funding achieve less experimental development than other researchers. Conversely, Ranga (2003) provides statistical evidence that at Katholik Universiteit van Leuven (“KUL”) in Belgium there was no significant shift to applied research publications in the period 1985–2000, when the university was receiving increasing funding from industry and was increasing the number of patents. However, both the study of Gulbrandsen and Smeby (2005) and of Ranga (2003) have produced evidence that researchers who have received industrial funding publish more articles than other researchers, but the quality of the journals was not investigated, and both studies have found a positive statistical association between industrial funding of university research and patenting activities With respect to the impact of patenting on research, it has been studied the hypothesis that IP rights prompt researchers to not divulge the results of their research to prevent the possibility of a future patent application being compromised. Therefore, there may be delays in publication to satisfy the requirement of novelty of patent laws. Academic researchers seeking to obtain a patent, in their own name or in the name of their universities or business partners,

85

Breschi et al., supra note 83. Calderini et al., supra note 83. 87 Gustavo Crespi, et al., The Impact of Academic Patenting On University Research And Its Transfer, 40 res. pol’y 55 (2011). 86

A European perspective on intellectual property and technology transfer 429 should keep their inventions secret until the patent application is filed.88 This aspect will be explored in the following section. Finally, it is theoretically argued that the university’s IP rights also affect teaching.89 Teaching is not associated with a heavy weight in the evaluation of the performance of university professors; therefore, teaching has a low impact on their careers. When the result of the patent is used in the academic evaluation process, this creates incentives for researchers to reduce their time/effort in some of their activities and, given the current weighting scheme, teaching can be the activity suffering the greatest reduction of time. However, at the European level, to the best of our knoweldge, in this regard no empirical studies exist.

VI.

THE COST OF UNIVERSITY IP

From a practice perspective one of the ongoing strategic and operational concerns for university management teams is the cost of university IP. In essence universities need to have financial and human capital resources to manage the university IP process and to manage their IP portfolio. In attempting to better understand the challenges surrounding the cost of university IP some authors have reduced the effects that are considered positive and have listed the negative aspects of university patenting.90 In this regard, Geuna and Nesta underline that literature and policy makers have often had an optimistic attitude towards institutionalization of university patenting.91 In one example, they note how the National Audit Office of the United Kingdom (“NAO 2002”) is dedicated exclusively to the description of the benefits and the means to improve the commercialization of the IP generated by the university, ignoring the focus on the costs of greater university involvement in the marketing of research.92 In the National Audit Office report, the authors do not deal with the fact that in UK universities, most of the university TTOs do not generate positive net income.93 Moreover, Geuna and Nesta point out that political literature generally overlooks the reality of university research spin-offs based on patents owned by the university.94 Often, returns on equity investments in young companies based on IPR are distorted; the typical success rate is rather low. Moreover, Geuna and Nesta (2006) identify five main possible negative effects of a greater institutionalization of IP rights: the trade-off between publication and patenting; the increase of applied research and the reduction of the fundamental research; reduction of teaching commitment; potential block that patents could create on future research, especially in areas

88

See, e.g., Andrew Webster & Kathryn Packer, “When Worlds Collide: Patents in Public Sector Research” in Universities and the Global Knowledge Economy (Henry Etzkowitz & Loet Lleydesdorff eds, 1997); Kenneth G. Huang & Fiona E. Murray, Does Patent Strategy Shape The Long-Run Supply of Public Knowledge? Evidence From Human Genetics, 52 aCad. of MgMt J. 1193 (2009); Fiona Murray & Scott Stern, Do Formal Intellectual Property Rights Hinder the Free Flow of Scientific Knowledge? An Empirical Test of the Anti-Commons Hypothesis, 63 J. of eCon. behav. and org. 648 (2007). 89 Geuna & Nesta, supra note 59. 90 See Mowery & Sampat, supra note 6; Geuna & Nesta, supra note 59. 91 Geuna & Nesta, supra note 59. 92 Comptroller & Auditor General, Delivering the Commercialisation of Public Sector Science, CoMptroller and audItor general (Feb. 2002). 93 Id. 94 Geuna & Nesta, supra note 59.

430 Research handbook on intellectual property and technology transfer where progress is largely cumulative; a negative impact on the culture of open science, in the form of greater secrecy, delays in publication, higher costs of access to research material or tools.95 Most of these aspects have already been discussed in the previous section and for some of them, we have noted how empirical research has often confirmed the unfounded nature of these fears. Here, we focus on what is considered the most controversial point of the institutionalization of university patents, to which reference has already been made in the introduction to the Chapter: patents have an impact on the culture of open science and on scientific progress. Scientific research usually works in an atmosphere of openness and sharing of knowledge, data and research results. This open nature of the scientific process is responsible for much of its success: the development of new knowledge thrives because researchers are based on mutual results, mutually fertilizing the perspectives of others through the discussion of results, and the sharing of data sources.96 Patents can turn this open culture into a more closed one. Many authors argue that university patents can restrict access to public knowledge and, in the long run, change the rules of open science. This can be achieved by decreasing informal interaction, incentives to increase secrecy in research and teaching, late publications, limited access to patented search tools, costly negotiations and opposition procedures. According to Nelson (2004), the increasing privatization of the scientific commons induces a risk to see an important share of future scientific knowledge becoming private property and hence falling outside the public domain. Moreover, the increasing use of the patent system has reinforced the fears that universities may be tempted to diffuse knowledge exclusively via license agreements, hence limiting the free flow of scientific knowledge and impeding further academic research. As analyzed by Perkmann et al. (2012), there is some evidence that academic researchers involved in patent activities practice higher degrees of secrecy than their non-inventors colleagues. Empirical research suggests that increased academic patenting may slow the diffusion of academic knowledge. Most research on this issue focuses on patents in the life sciences because, in this field, IP is of high potential value. In the US context, Blumenthal et al. (1997) collected data between October 1994 and April 1995 from 2,052 faculty members in the life sciences at the 50 US universities.97 They found that faculty members who have research relationships with industry are more likely to restrict their communication with colleagues, and high levels of industrial support may be associated with less academic activity without evidence of proportional increases in commercial productivity. Campbell et al. (2002), in a survey of 1,897 geneticists, demonstrated that almost half of them had received requests or additional information, data or materials related to their published research and that 10% of all requests for further information were rejected.98 Walsh et al. conducted 70 interviews with lawyers, managers and scientists and showed that exclusive licenses are pervasive and have potential negative effects on technological progress.99 Murray 95

Id. Partha Dasgupta & Paul A. David, Towards A New Economics Of Science, 23 res. pol’y 487 (1994). 97 David Blumenthal, et al., Withholding Research Results in Academic Life Science: Evidence From A National Survey of Faculty, 227 J. of the aM. Med. ass’n 1224 (1997). 98 Eric G. Campbell, et al., Data Withholding in Academic Genetics: Evidence from a National Survey, 287 J. of the aM. Med. ass’n 473 (2002). 99 John P. Walsh, et al., “Effects of Research Tool Patents and Licensing on Biomedical Innovation”, in Patents in the Knowledge-Based Economy (2003). 96

A European perspective on intellectual property and technology transfer 431 and Stern (2007) constructed a sample of 169 patents associated with publications published in Nature Biotechnology in the period 1977–99.100 The authors concluded that the initial knowledge disseminated through scientific publication and patents was granted with some delay. They also found that the citation rate decreases between 10 and 20% after a patent grant, and the decline is more pronounced for researchers with public sector affiliations. Therefore, they rejected the hypothesis that IP does not affect the dissemination of scientific research, and the existence of IP restrictions in subsequent research cannot be excluded. Huang and Murray (2009) show that the granting of gene patents decreases the long-term production of public genetic knowledge.101 At the European level, Webster and Packer developed a questionnaire involving UK universities and semi-structured interviews to TTO managers, patent agents and examiners in the UK in 1993.102 The survey addressed, among others, the issue of dissemination of research results. Although they did not report statistical results, the authors stated that “it is apparent from our survey that academic dissemination can be compromised.” They pointed out that a number of respondents have become much more careful in choosing which information to disclose in their publications to avoid compromising future patent applications. In addition, the European Commission report (2002) surveyed public and private researchers to investigate the delay in the dissemination of certain academic research. The report identifies the political concerns that “a public research policy that supports both rapid dissemination to foster scientific progress and patenting to support exploitation of the results of publicly funded research has to establish framework conditions that help researchers to avoid conflicts of interest, e.g. ensures rapid publication while giving protection to the results” (European Commission, 2002, p.10). Some researchers in the European context report considerable delays in the publication of research results of less experienced academic patentees or young TTOs.103

VII.

CONCLUSIONS AND FUTURE AVENUES OF RESEARCH

From the Chapter it is evident that there are institutional, cultural and technological differences between Europe and other countries and continents. Even within Europe there are variations in technology transfer activities and IP. National policies and legislative approaches adopted by Member States of the EU have provided a strong underpinning scientific base for technology transfer, IP protection and exploitation. Regional specialisms in particular technology areas in addition to local norms and practices add further complexity to European technology transfer. The more recent growth in European university TTO is a welcomed development and provides some tangible evidence as to the institutional prioritization of third mission activities and technology transfer. The challenge for Europe with respect to technology transfer is to remove the barriers to technology transfer between universities, PROs and industry. This still remains a significant constraint in ensuring that the European technology transfer system is as effective as possible so as to encourage greater knowledge flow within and outside of Europe.

100

Murray & Stern, supra note 88. Huang & Murray, supra note 88. 102 Webster & Packer, supra note 88. 103 Azèle Mathieu, et al., Turning Science Into Business: A Case Study of a Major European Research University, 35 sCI. and pub. pol’y 669 (2008). 101

432 Research handbook on intellectual property and technology transfer While there is no overarching legislation in Europe such as the Bayh-Dole Act and other legislative framework and policy interventions to support university industry technology transfer, it is clear that Europe has replicated aspects of the US legislative and institutional supports to further the growth of technology transfer in Europe. Moving the individual property rights or “professors’ privilege” from the individual faculty member to the institutional ownership has been a significant change for all stakeholders. It has resulted in the growth of TTOs within universities and has led to the growth of a new set of skills and expertise that universities need to possess in order to meet the growing expectations of them as engines of economic growth.104 While much progress has been made to date there is an ongoing need to further enhance the technology transfer and IP skills and competences within TTOs and more widely in universities and PROs. The creation and progress towards a functioning UPC is an ambitious and potentially game-changing development for Europe. It has many positive aspects in terms of providing further confidence and support for Europe being progressive with respect to IP. For inventors and firms in theory it should provide benefits in terms of reducing the transactions costs, resources and time with respect to IP litigation. It provides professional services firms with new commercial opportunities to innovate and provide clients with a more enhanced product offering that supports their innovation and R&D ambitions. Only time will tell how lawyers, technology transfer professionals and firms will view decisions by UPC and how they in turn will influence their individual behaviors and decision making. There are significant opportunities to progress research on IP and technology transfer in a European context. There are significant opportunities for further research at the micro level with a particular focus on individual actors such as lawyers, technology transfer practitioners, TTO directors, scientists in the principal investigator (PI) role105 and how they use IP to further their individual professional and individual ambitions and aspirations. So for example, how proficient are scientists in the PI role with respect to basic technology transfer and IP strategies? How reliant are technology transfer practitioners and TTO directors on in house or external legal advice for strategic and operational decisions? How innovative are European lawyers with respect to IP and technology transfer based product offerings? The creation of the UPC provides opportunities for scholars from different disciplines to focus on different aspects of its operations, impacts and influences in the IP legal-direction setting in a European context. For scholars focusing on technology transfer, the impact of UPC in lowering transactions costs and improving the efficiency of the Technology Transfer and IP process can provide relevant policy and practice insights. For legal scholars, the operations, the nature of the cases before the UPC, the dynamics of the litigation process in this new pan-European court and their judgments provide opportunities for further research. Also, from

104 Maribel Guerrero, James A. Cunningham, & David Urbano, Economic Impact of Entrepreneurial Universities’ Activities: An Exploratory Study of the United Kingdom, 44 res. pol’y 748 (2015). 105 James A. Cunningham & Paul O’Reilly, Macro, Meso and Micro Perspectives of Technology Transfer, 43 J. teCh. transfer 545 (2018); James A. Cunningham, Paul O’Reilly, Conor O’Kane, & Vincent Mangematin, “Publicly Funded Principal Investigators as Transformative Agents Of Public Sector Entrepreneurship” in Essays in Public Sector Entrepreneurship 67 (2016); Ciara Fitzgerald & James A. Cunningham, Inside the University Technology Transfer Office: Mission Statement Analysis, 41 J. teCh. transfer 1235 (2016); James A. Cunningham, Matthias Menter, & Conor O’Kane, Value Creation in the Quadruple Helix: A Micro Level Conceptual Model of Principal Investigators As Value Creators, 48 r&d MgMt 136 (2018).

A European perspective on intellectual property and technology transfer 433 a legal professional services perspective, research that explores the product offering response to the UPC and the associated business models is worth exploring. University based TTOs are well established and have evolved more in the US in terms of skills, expertise, resources and structure. Within the European context, university TTOs in comparison are overall still in the embryonic stages of evolution and development. PROs are more established and well developed in European countries such as Germany, France and the UK. For both university-based TTOs and PROs, there is a need for research that examines the drivers of IP strategies in these settings and how these entities implement IP policy and practices in their own institutional contexts. Moreover, there is a need to better understand the decision dynamics and criteria used within these settings to decide which technology transfer mechanisms to support and devote more resources to, and how these offices acquire the necessary skills and capabilities to maintain their boundary spanning organizational role. Research is also required in a European context that centers on the impact of EU directives that directly and indirectly influence technology transfer policies within Member States. Moreover, different Member States have adopted different policy and legal mixes to support effective technology transfer and to support IP protection. Large-scale studies are warranted to explore the influence and impacts of these variations and to move towards building large-scale datasets that capture the more nuanced as well as more generalizable impacts. Finally, there is a need for large scale pan European study of the cost of university IP to address a significant empirical deficit, and such research would also contribute to practice.

20. The current state of university technology transfer in China Zhang Chu and Shi Xiaoxue

I.

INTRODUCTION

Innovation has become a buzzword relevant to the economic development of all countries in the world. University research and academic discovery, especially the quantity, quality and speed of their transformation to the market via technology transfer, have become important criteria measuring a nation’s innovation capability and level of economic development. As a pivotal force within a nation’s innovation system, universities harbor favorable conditions for scientific research and a great wealth of academic and research resources. Thus, the role of universities cannot be ignored in the advancement of national innovation. A large number of research tasks, either at a state, local or enterprise level, can be undertaken by universities; these institutions also boast a wide array of research discoveries, as well as solid groundwork and multiple favorable conditions for the transformation of these discoveries. As China is vigorously strengthening the construction of its innovation ecosystem and implementing an innovation-driven development strategy, a growing state-level attention has been paid to university technology transfer, as manifested by continuously increasing state investment and ever more attention placed on the subject by universities, their research faculty and regional governments. This Chapter presents a panoramic view of the current status and problems related to technology transfer in Chinese universities in an attempt to provide valuable references for ensuing studies and practices involving this topic. It does so by presenting a comprehensive review of university technology transfer in China, including a review of policies and normative documents related to university technology transfer in China; the history and current status of university technology transfer in China; the success and experience of technology transfer in Chinese universities; and existing problems with university technology transfer in China.

II.

REVIEW OF POLICIES AND NORMATIVE DOCUMENTS RELATED TO UNIVERSITY TECHNOLOGY TRANSFER IN CHINA

At the state level, China has introduced an intensive array of policies and regulations related to university technology transfer. A “trilogy” of laws related to these activities have taken shape. They include: ● “Law of the People’s Republic of China on Promoting the Transformation of Scientific and Technological Achievements” (promulgated and implemented in 1996, amended in 2015

434

The current state of university technology transfer in China 435 with the “Law on the Transformation of S&T Achievements”), passed by the Standing Committee of the National People’s Congress; ● “Several Provisions on Implementing the Law of the People’s Republic of China on Promoting the Transformation of Scientific and Technological Achievements” (the “Provisions”) (issued in 2016), issued by the State Council of China; and ● “Action Plan for Promoting the Transfer and Transformation of Scientific and Technological Achievements” (issued in 2016) issued by the General Office of the State Council. In July 2016, the State Council published the “National Plan for Scientific and Technological Innovation During the Thirteenth Five-year Plan Period,” which dedicates the entire content of Chapter 21 to university technology transfer by focusing on the mechanisms of university technology transfer. At the local level, a succession of detailed provisions on university technology transfer have also been set forth by provincial and municipal governments based on the actuality of regional transformation, as an effort to follow the guidance of the CPC Central Committee and the State Council. For example, Beijing, Hubei and Nanjing have formulated and promulgated local policies on university technology transfer like the “Ten Provisions on University S&T Achievements Transformation,” “Ten Golden Provisions” and “Nine Provisions on Technological Innovation.” As of now, a relatively complete set of policy systems on university technology transfer encompassing top-level policies and regulations, mid-level research and innovation plans, and local and ministerial opinions has taken its initial shape. A number of characteristic policies and normative documents are outlined in Table 20.1. Compared to earlier policies and regulations, the legal documents outlined above reflect a revolutionary change in terms of the transformation of university technology transfer under the new circumstances in China. Such a change is mainly manifested by the following aspects: First, the management mechanism of university research with market potential has been improved and modified. According to the existing legal norms and policies, the management system of academic research in China’s universities has been systematically and profoundly changed. Definitive norms relating to the management, pricing, evaluation and supervision of academic research in Chinese universities have been provided. Each university can now exercise autonomous management over its faculty’s discoveries and adopt market-oriented pricing approaches. Evaluation of discoveries has also shifted from the earlier unified approach to the present classified approach. As prescribed by the “Law on the Transformation of S&T Achievements,” classified evaluation approaches should be established for different forms of discoveries. Points of emphasis for supervisory management over university inventions have also migrated from pre-implementation approval and registration to mid- and post-implementation supervision. Second, the degree to which scientific research personnel are incentivized has been increased. Both the “Law on the Transformation of S&T Achievements” and the Provisions attempt to tackle the practical difficulties in profit distribution arising from the process of university technology transfer. Presently, there exist two major difficulties relating to profit distribution: first, how to drive the scientific research personnel’s motivation to translate their research discoveries to the market; and second, under the current institutional mechanism, scientific researchers who concurrently assume leadership roles may have scruples about obtaining profits generated from their discoveries, given their administrative positions. To address these knotty problems, the “Law on the Transformation of S&T Achievements” and

436 Research handbook on intellectual property and technology transfer Table 20.1

Major regulatory and policy documents pertaining to university technology transfer in China

Name

Content

“Opinions on Accelerating the

Deepen the reform of management over university research;

Time of Issuance or Latest Amendment June 2014

Transformation and Industrialization promote the reform of management over scientific research of Scientific and Technological

assets; deepen the reform of financial and funds management;

Achievements in Scientific Research strengthen incentive mechanisms for scientific research Institutions (Interim)”

personnel; enhance scientific research institutions’ application and promotion of new technologies and products.

“Law of the People’s Republic

As a guiding legal document aiming to promote university

of China on Promoting the

technology transfer, the law has prescribed the forms, organizing

Transformation of Scientific and

and implementation approaches, support measures and other

Technological Achievements”

details related to the transformation of China’s university

August 2015

technology transfer ecosystem. “Circular on the Action Plan

Sets forth the basic principles, main objectives, eight key tasks,

for Promoting the Transfer and

and organizing and implementational supports for promoting

Transformation of Scientific and

university technology transfer in China.

April 2016

Technological Achievements” “Guiding Opinions of the

Encourages research personnel to leave their research jobs

Chinese Academy of Sciences on

with their discoveries and engage in entrepreneurship causes;

Accelerating the Promotion of

delegates the power to use, dispose of inventions and manage

the Transfer and Transformation

income to affiliated units of the Chinese Academy of Sciences;

of Scientific and Technological

supports and guides affiliated units of the Academy to explore

Achievements in the New Era”

innovative approaches to university technology transfer.

“Circular of the Beijing Municipality Simplifies the budgetary planning and evaluation process for on Policy Measures to Further

August 2016

September 2016

scientific research projects funded by the municipal government;

Improve the Management of

entrusts the units and personnel undertaking scientific research

Scientific Research Projects and

projects with greater autonomy; innovates the investment and

Funds Supported by the Municipal

support approaches for government- backed funds; accelerates

Government”

the promotion of translatable research.

“Circular on Improving Income

Extends the time limits of tax levies for stock options, restricted

Tax Policies Relating to Equity

shares and option awards in listed companies; implements

Incentives and Technology Shares”

optional preferential tax policies for investments and shares

September 2016

made in the form of technological achievements; sets forth supportive management measures. “Several Opinions on Implementing Promotes the formation of a distribution mechanism that reflects November 2016 Distribution Policies Oriented to the the enhancement of knowledge-based values; increases the Enhanced Values of Knowledge”

autonomy of scientific research institutions and universities in income distribution; further enhances the incentive and guiding role of scientific research funds; strengthens the long-term incentive effect of IP pertaining to scientific researchers involved in technology transfer; permits lawful part-time employment and pay for scientific research personnel and professors.

the Provisions explicitly encourage scientific research institutions such as universities to carry out multiple forms of technology transfer and significantly increase the standard of awards granted to top-tier scientific researchers. For example, the “Law on the Transformation of S&T Achievements” provides that upon successful completion of technology transfer, a university

The current state of university technology transfer in China 437 should take out an amount no less than 50% of the net income generated by the transfer as an award to personnel who have made significant contributions to it. Such a proportion represents a dramatic increase over the earlier amount of 20%, which is intended to enhance its incentive effect. In the meantime, the law also explicitly sets forth the eligibility of scientific researchers who concurrently serve in leadership roles to receive monetary awards. The newly amended law prescribes that it is both necessary and legally and regulatorily compliant for scientific researchers concurrently serving in leadership roles to receive a certain amount of award and compensation from the income generated from successful technology transfer outcomes to which they have contributed. Moreover, the new law provides in detail that researchers serving in management positions shall not receive options or cash awards from enterprises; other researchers serving in leadership roles may receive certain amounts of cash awards, with an exception of option incentives. Scientific research personnel who do not serve in major leadership positions may obtain a certain amount of option incentives or cash awards. Third, the role of enterprises in the process of university technology transfer has been emphasized. By highlighting the principal role that enterprises play in R&D and industry-university-research cooperation, the “Law on the Transformation of S&T Achievements” expects to proactively enhance the important effect of enterprises in forming alliances with the subjects of innovation, including universities, and promoting the cooperation between industry, universities and academic researchers. The law supports enterprises and universities jointly establishing R&D platforms and transformation institutions in order to facilitate technology transfer. It also encourages the creation and transformation of university research taking place through multiple forms, such as innovation alliances and student bases for research and experiments. Generally speaking, these relatively new policies and normative documents pertaining to university technology transfer in China play a positive role in promoting the creation and transformation of academic research that can help society. However, it should not be overlooked that there is still some weakness in the enforcement of some norms, despite their strong propositions in terms of principles, and the initial legislative expectations and institutional effects have yet to be fully fulfilled. Many Chinese scholars and researchers have also raised questions and proposed recommendations on how to strengthen the enforcement of the above norms and ensure their implementation.

III.

HISTORY AND CURRENT STATUS OF TECHNOLOGY TRANSFER IN CHINESE UNIVERSITIES

Given the increasing attention to the issue, how should we steer technology transfer in universities toward a correct direction? A legislative review and policy study alone do not suffice. It is crucial to cast our eyes over history and trace the historic development trajectory of university technology transfer in China. A historic entry point not only leads to substantial experience and lessons learned, but also provides insights into implications for the present and future.

438 Research handbook on intellectual property and technology transfer A.

Development History

In the 1940s and 1950s, the third technological revolution started to sprout. The transformation of research in universities across the world was thriving driven by the socio-economic development of each country. For example, the Silicon Valley, established on the basis of the Stanford Research Park delineated by Stanford University, has been an extraordinary outcome arising from the transformation of university research. Despite the fact that technology transfer from universities had started and gained rapid development in other countries, technology transfer in Chinese universities was actually separated from the development of a market-oriented economy, as China at the time was just lifted out of the war and all aspects of national construction and economic development had yet to be started. It was not until the 1970s, when China started to implement the “reform and opening up” policy, that scientific research in universities started to gain full-scale development in order to align with the practical needs of economic development. Generally, a number of barriers in terms of development concepts and work mechanisms, as well as a lack of corresponding incentive mechanism, had hindered the progress of scientific research and technology transfer in Chinese universities prior to the implementation of the “reform and opening-up,” preventing these achievements from fully realizing their economic and social values. Thanks to the four decades worth of efforts committed by the Chinese government in reform and opening-up, scientific research and technology transfer in Chinese universities had ushered in a new era. That progress can roughly be divided into the following developmental stages.1 1. The initiation period centering on horizontal technology services Following the implementation of the “reform and opening-up” policy, Chinese enterprises felt a pressing need to acquire research that could be integrated to the market and highly expected that they could leverage the strength of universities in scientific research to improve their manufacturing process so as to produce high-tech products that could meet market demands and generate economic profits. Under such demand, those scientific research projects that were commissioned by enterprises to universities with a view to meet actual needs and address practical problems marked the beginning of Chinese universities orienting their scientific research towards the market. During this period, the major part of the transformation of research in Chinese universities had been completed in the form of enterprise commissions, by which universities provided research services for relevant projects. Scientific research projects commissioned by enterprises to Chinese universities could be provided in many forms, including technical consultation, training, commissioned technological development, and provision of technical services. The initiation period of university-run enterprises 2. In the 1980s, Chinese universities started an exploration of establishing university-run enterprises. Aiming to fully leverage their advantages in technological innovation, the high-tech enterprises established by Chinese universities, at the time, were mainly operated in the form of factories and farms. The orientation of these enterprises also gradually shifted from venues

1

See Gong Zhenyu, Zhang Yanmei, Cao Guiquan, Pan Yun, and Jia Qijun, Investigation on Conversion of Achievements in Science and Technology of Institutions of Higher Learning, 3(2) J. tIanJIn u. 242 (2000).

The current state of university technology transfer in China 439 for students’ experiments and internships to an important forefront to promote the transformation and industrialization of research in universities, lift the industrialization level of national high-tech and new technologies, and facilitate regional economic development. An array of university-run high-tech enterprises—as exemplified by Tsinghua Tongfang, Grand Horizon, and Founder—emerged and grew into world-class and prestigious enterprises today. The period of close industry-university-research cooperation 3. Starting from the mid- and late-1980s, Chinese universities ushered in a period featuring a close integration between production and research, which was also a rapidly developing period for the transformation of university research. Apart from the approach of establishing university-run enterprises, the industry-university-research programs could also be implemented through joint establishment of research institutions and funds between universities and enterprises. Long-term cooperative opportunities granted by these institutions and funds have further extended the depth of scientific research in universities. 4. The period of modern enterprise construction relying on intellectual property There was no substantial difference between the enterprises established by Chinese universities during this stage and their earlier university-run counterparts. However, the former boast greater diversity in organizational forms. A variety of investment entities started to seek cooperation with universities in a joint effort to establish modern enterprises. Such an approach enabled risk-sharing and provided greater efficiency in utilizing and transforming research compared to enterprises solely established by universities. In addition, it also facilitated marketization of product manufacturing in university-run enterprises through leveraging the market operating experience possessed by market entities. During this period, the organizational forms of university-run enterprises can roughly be divided into two categories: first, enterprises where universities acted as shareholders or investors through contributing capital; second, university-run enterprises where universities acted as shareholders or investors by contributing intellectual property rights. B.

Current Development Status

1. Emphasis on transformation of university research Against the backdrop of increasingly enhanced construction of innovation systems and the implementation of an innovation-driven development strategy, the transformation of research in Chinese universities has gained unprecedented attention. Universities, governments, and enterprises have boosted their investment in university-led scientific research programs to a heightened degree in order to encourage them to accelerate the output of transferrable research. As can be seen from a host of laws, regulations and policy documents pertaining to R&D, and the output and transformation of university research that has occurred over recent years, the transformation of university research is regarded as a key component for implementing the innovation-driven development strategy in China. Great expectations have also been placed by the government which has introduced a number of measures to encourage and support the output and transformation of university research. R&D funds granted to Chinese universities have continued to rise. As of 2016, the R&D funds of Chinese universities increased by 7.4% to 107.22 billion yuan, a growth amounting to 7.36 billion yuan compared

440 Research handbook on intellectual property and technology transfer to those of the previous year. The composition of these funds include 68.70 billion yuan funded by the government, 31.05 billion yuan of enterprise funds, and a total of 7.4 billion yuan of overseas and other funds, which respectively account for 63.8%, 30.2% and 6% of the R&D funds of universities. Since 2005, the funds granted by the government have constantly taken up the largest proportion in the R&D funds of universities, which have basically remained above 54%.2 2. Continuously increased strength in scientific research in Chinese universities The vigorous support from the government has resulted in remarkable outcomes in the construction of scientific research teams, both quantitively and qualitatively, in Chinese universities. According to the latest data published by the Ministry of Science and Technology, as of late 2016, the total number of research personnel in Chinese universities was 360,000, a 6% year-over-year increase and accounting for 9.3% of the total scientific research personnel in the nation. In 2015, China’s scientific research institutions totaled 13,062, an increase of 1,330 as compared to the previous year. In terms of the classifications of research, China’s higher education institutions hired 167,000 researchers in basic sciences, 173,000 researchers in applied studies, and 21,000 researchers in experimental development in 2016. Of all scientific researchers (including those engaged in basic and applied sciences) in China, those working at universities account for 47.6%, more than 18.0% higher than the percentage of research institutions.3 Technology transfer outcomes in China 3. Over four decades worth of development, Chinese universities have achieved considerable success in technology transfer. According to the latest data published by the Ministry of Science and Technology, the number of technological transaction contracts entered into by universities and scientific research institutions have seen a significant growth. As of the end of 2016, Chinese universities and scientific research institutions had executed 90,573 technological contracts in the forms of technological transfer, technology invested as ownership stakes, and industry-university-research cooperation, with a transaction amount totaling 106.52 billion yuan, a 21.8% increase. The year-on-year increases in transactional amounts for technological contracts entered by universities and scientific research institutions were respectively 14.6% and 25.8%.4 As of 2016, universities had entered into 60,000 technological contracts in the technological market as sellers, accounting for a 4.7% growth compared with that of the previous year and 18.7% of total contracts of the same kind in the nation. The transactional amounts of these contracts increased by 14.6% to 36 billion yuan compared to that of the previous year, accounting for 3.2% of total transactional amount of nationwide technological contracts.5

2

See Statistical Analysis of China Universities’ R&D Activities in 2016, MInIstry of sCI. teCh. of the people’s republIC of ChIna, available at http://www.most.gov.cn/kjtj/201803/ P020180326496065933396.pdf (last visited Oct. 6, 2018). 3 See id. 4 See id. 5 See id. and

The current state of university technology transfer in China 441 4. University professors are encouraged to found businesses The substantial support provided by current policies has also resulted in great enthusiasm in university researchers to engage in research undertakings. In March 2017, the Ministry of Human Resources and Social Security issued the “Guiding Opinion on Supporting and Encouraging Professional Technical Personnel of Public Institutions to Engage in Innovation and Entrepreneurship,” which aims to support and encourage technical professionals serving in state-owned and public institutions to take part-time jobs in enterprises, scientific research institutions, universities and social organizations with businesses close to their original employers, or to engage in entrepreneurship projects and establish enterprises related to their specialties while still retaining their current employment relationships, so as to tap into their innovative potentials and accelerate the transformation of research to the market. In the meantime, the Guiding Opinion also supports and encourages professional technicians serving in state-owned public institutions to leave their posts and start businesses while still retaining their employment relationship up to three years. In China, the majority of university professors and researchers fall into the spectrum of personnel hired by public institutions, and thus are entitled to the preferential conditions granted by the above policy to engage in entrepreneurship activities in a part-time or leaving-position basis. While conducive to arousing the vitality and enthusiasm of professional technical personnel serving in public institutions like universities and scientific research institutions in committing to innovative and entrepreneurship activities, the Guiding Opinion also helps create a policy and institutional environment highly favorable for innovation and entrepreneurship.

IV.

SUCCESS AND EXPERIENCE OF TECHNOLOGY TRANSFER IN CHINESE UNIVERSITIES

Since the implementation of the “reform and opening-up” policy in China, university technology transfer has been continuously advanced with considerable successes and distilled experience, outlined as follows: A.

Policy Support

Policy support constitutes an important factor driving the success of the transformation of university technology transfer in China. Substantial policy support, either in the aspect of the construction of human resources and fund investment for scientific research or incentive measures for technology transfer, has been provided by the Chinese government. As shown by the data stated above, the workforce of scientific research personnel in Chinese universities has continued to rise over more than 40 years, which collectively represents the largest research team across the nation. In terms of R&D funds, government granted funds represent the largest source of R&D funds for universities. In the meantime, the government also issued dedicated documents to provide favorable conditions for scientific researchers serving in universities who are willing to engage in entrepreneurship activities. Undoubtedly, the high degree of policy support and institutional and governmental incentive mechanisms have enabled technology transfer successes for Chinese universities. Under such an atmosphere, university technology transfer has been gaining increasing attention and is poised for an even more rapid pace of development.

442 Research handbook on intellectual property and technology transfer B.

Highly-Diversified Cooperation and Communications

In terms of output and transformation of research, Chinese universities have long been actively seeking cooperative and communicative opportunities with social research institutions and other industry and academic institutions. Joint development institutions established between Chinese universities, enterprises, local governments and administrative departments have been serving an important means for the industry-university-research cooperation. Carrying out such diversified forms of communication can effectively prevent scientific research in universities from being disjointed from actual market demands. The cooperation between universities and social institutions continues to be extended to greater depth, which often takes place in two forms: the government-university cooperation mode and the university-enterprise cooperation mode. In the government-university cooperation mode, the government plays a dominant role in operating incubators (e.g., university science and technology parks), national engineering centers, technology transfer centers and university-run provincial research institutes. Focusing on mutual participation and cooperation, the university-enterprise cooperation mode integrates the innovative strength of universities and the capacity of enterprises in applying research, which typically takes place in the form of joint research institutes, collaborative research centers, innovation research institutes and joint laboratories. Another cooperative mode is university-enterprise integration. In China, some successful entrepreneurs may establish private universities by donating funds, where enterprises act as major shareholders. The technological needs of these enterprises will naturally become the target of R&D projects conducted in such universities, and their discoveries will also be the major technological resource accessible to enterprises. Both parties can effectively share their resources in R&D teams, management talents, and employment practices.6 Over recent years, the technical intermediaries and financial investment institutions have also gradually become another source of subjects for university-enterprise cooperation, exerting mutual effect on the transformation of research while also providing convenience for the research, development transformation, and application of university research. It is expected that such a cooperative mode will continue to prosper for some time to come. C.

Leveraging the Innovative Strength of Universities in Scientific Research Through Establishing University-Run Enterprises and Science and Technology Parks

Establishing university-run enterprises and university science and technology parks also constitutes an important channel to fulfill and facilitate the transformation of university research. Among these entities, university science and technology parks serve as incubators and entrepreneurship service institutions for high-tech enterprises, thereby playing a crucial role in promoting the transformation of research. In China, university science and technology parks refer to those institutions supported by research-oriented universities or university clusters that are intended to provide services for technological innovation and technology transfer by integrating comprehensive intellectual strengths of universities in talents, technologies, information, laboratory equipment and library resources with other advantages derived from social 6 See Wang Jian, Review of the Transformation of Scientific and Technological Achievements in China’s High Institution, 8 ChInese u. teCh. transfer 73 (2018).

The current state of university technology transfer in China 443 resources. These institutions should become bases of technological innovation, incubators for high-tech and new technology enterprises, bases for the clustering and training of innovative and entrepreneurial talents, and demonstrative bases for industry-university-research integration.7 According to China’s policies, technology enterprises with their addresses registered within university science and technology parks are entitled to preferential policies in tax and other aspects. In China, a host of difficult requirements must be met and an approval jointly approved by the Ministry of Science and Technology and the Ministry of Education must be obtained in order to establish a national-level university science and technology park. The emergence of China’s university science and technology parks can be traced back to 1999, when the Ministry of Science and Technology and the Ministry of Education jointly issued the “Circular on Organizing and Implementing the Pilot Construction Program of University Science and Technology Parks.” Based on that circular, fifteen universities were selected as pilot units to establish these parks. Shortly after, the construction of university science and technology parks in China officially began. As noted by the Ministry of Science and Technology and the Ministry of Education, the establishment of university science and technology parks should focus on activating the research resources possessed by universities and transform their strengths, talents and intellectual prowess into productivity and economic advantages. In terms of practical operations, incentive mechanisms should be formulated so as to stimulate the initiative of research personnel toward the ultimate goal of technology transfer. For universities that have established science and technology parks after obtaining approval, competent authorities—including the Ministry of Education—adopt a variety of approaches, including administrative inspections, to examine the implementation of university technology transfer, identify whether the expected targets have been met, and urge relevant universities to fulfill their targets. For instance, the Ministry of Education issued the “Circular on Conducting Special Supervision and Inspection on the Implementation of Policies Relating to the Transformation of Scientific and Technological Achievements” on September 19, 2018, which states that the Ministry would conduct a nationwide special supervision and inspection on the implementation of policies relating to university technology transfer. The supervision and inspection mainly focused on five areas: first, the working mechanisms of university technology transfer; second, the personnel management systems related to university technology transfer; third, the accountability mechanisms related to university technology transfer; fourth, how to simplify administrative procedures and delegate powers to encourage university technology transfer; and fifth, the implementation of policy incentives related to university technology transfer.8 In China, such administrative supervisory measures can effectively promote university technology transfer.

7 Interim Measures for the Administration of National University Science and Technology Parks, MInIstry of sCI. and teCh. of the people’s republIC of ChIna, Nov. 2010. 8 See Circular on Conducting Special Supervision and Inspection on the Implementation of Policies Relating to the Transformation of Scientific and Technological Achievements, MInIstry of eduC. of the people’s republIC of ChIna, available at http://www.moe.gov.cn/srcsite/A16/moe_784/201809/ t20180927_350204.html (last visited Oct. 9, 2018).

444 Research handbook on intellectual property and technology transfer

V.

EXISTING PROBLEMS WITH UNIVERSITY TECHNOLOGY TRANSFER IN CHINA

Despite the fact that considerable successes have been achieved by China’s universities in terms of technology transfer over four decades of development since the implementation of the “reform and opening-up” policy, there are still some shortcomings in this area that have hindered the effectiveness of university technology transfer. Many discussions have focused on this topic, in both practical and theoretical circles, and several research studies have contributed to the dialogue. By drawing on openly available data, this Chapter outlines some of the identified problems. A.

Restrictions of Scientific Research Mechanisms

Presently, there are a few institutional and mechanism problems existing in Chinese universities that can hinder technology transfer. Among them, the problem that has received the most criticism is a lack of evaluation and incentive mechanisms for technology transfer in the majority of universities. The evaluation systems on academic research present in most Chinese universities take into account the number of scientific papers published and the quality of the journals in which they are published. Considerations such as the research topics or projects undertaken are also evaluated. In particular, when scientific research personnel such as university professors are evaluated for professional and technical titles or promotions (or when they are initially considered for hire), the aforementioned factors predominate consideration of their merits. This reality has led to the domination of “academic thinking” and “expert-like thinking” in applying for and evaluating scientific research projects, and a lack of a “market-oriented thinking” with respect to academic research.9 Such a dominant evaluation system in China’s universities has inevitably spawned an excessive inclination toward “academic papers” and “research subjects.” As a consequence of these reward systems, scientific researchers have to distribute much of their energy to publishing papers in certain journals and applying for research grants, thereby significantly affecting the amount of time they can devote to pursuing research that might lead to technology transfer. Indeed, undertaking scientific research projects is supposed to be conducive to producing translatable research, but the same academic evaluation system and criteria—which focus on such textual accomplishments as academic papers—have been adopted in assessing the output of technology transfer. Such an evaluation mechanism results in weaker “market-oriented thinking” in scientific research personnel working in universities, as they pay much of their attention to the theoretical aspect of technology transfer and whether these research outcomes can help the final publication of their papers. This incentive structure significantly restricts the market transformation of academic research and further leads to a difficulty in fulfilling the promise of university technology transfer in China.

9 See Shen Yinan, Zhang Chao, Zhu Guofeng, & Sun Dongbai, Problems, Causes and Measures for the Transformation of S&T Achievements in Universities, 3 ChInese u. teCh. transfer 9 (2016).

The current state of university technology transfer in China 445 B.

A Gap Between Research and Market

On the one hand, under the restrictions imposed by the evaluation criteria on scientific research competence and the existing mechanisms for evaluating technology transfer, university researchers in China tend to deliver their research in the form of academic papers, thereby exerting an obstacle for market-oriented transformation of these achievements. On the other hand, training on market-oriented and business thinking is also scarcely available for China’s university researchers, who tend to conduct more basic research projects than applied ones. Scientific researchers are not distinctly aware of the market’s needs. Neither do they give much consideration to the economic and market benefits of their research. As a result, notable research discoveries can easily be misaligned with market need, and usually end up being shelved instead of being directly utilized. C.

Inadequacy of Professional Teams Working on Technology Transfer

Although universities boast a great wealth of research resources and teams, for the time being, the majority of scientific researchers are undertaking the forefront technical development tasks, thus having little time or energy to carry out work falling outside of the spectrum of scientific research or technological development. In addition, despite the fact that these researchers usually are experts in specific fields, their expertise does not mean that they will be skillfully completing all aspects of their research, nor does it mean that they will collaborate well in teams. In fact, the transformation of university research in China is a systematic project. Apart from basic research, professional teams are expected to carry out follow-up tasks like promoting and marketing their invention, so as to integrate resources and complement different strengths. However, under current conditions, there is still a long way to go for systematic and professional teams committed to seeing their discoveries reach the market. It would be of great significance for improving the efficiency of university technology transfer in China to build unified, coordinated, efficient and orderly functioning research teams. D.

Unvaried Sources of Research Funds

For the time being, the government funds the majority of scientific research projects in China. On a yearly basis, the Chinese government still accounts for the largest sponsor for research projects with market potential in universities. The question is whether such an investment approach is sustainable. The fact that relatively scarce research funds for applied research in Chinese universities come from enterprises may also indicate low expectations held by these enterprises for the outcome of research. Another reason may lie in the fact that as market situated entities, enterprises tend to focus more on short-term profits, and the trade-offs between cost and effectiveness may mean enterprises are unwilling to make excessive investment in long-term scientific research. For the long run, extending the sources of research funds to include enterprises would be helpful to universities so as to bridge them with the market and construct a more sustainable, market-oriented R&D model. Of course, to fulfill such a goal, there is a pressing need to address each of the many shortcomings of university technology transfer in China. Only when university technology transfer outcomes are improved, and market potentials realized, can market entities gain enough confidence in research conducted by universities.

446 Research handbook on intellectual property and technology transfer

VI.

CONCLUSION

This Chapter provides a panoramic review of the development course and current status of university technology transfer in China. As can be seen, China’s universities have gained substantial development and distilled a great wealth of experience in technology transfer over the past four decades, including enhanced attention, policy support and diversified forms of attention and support. However, it should also be noted that there are still many obstacles in terms of institutional mechanisms, research orientations, team building and investment resources that hinder the development and success of university technology transfer in China, thus requiring continued attention by academics and practitioners alike.

21. Make and share: intellectual property, higher education, technology transfer, and 3D printing in a global context Matthew Rimmer

I.

INTRODUCTION

During his Presidency, the United States president, Barack Obama, championed the Maker Movement in the hope that it would help transform education, innovation and manufacturing. Hosting a White House Maker Faire in 2014, Obama emphasized that 3D printing could play a key role in education, as well as training, employment and innovation.1 He highlighted the role of 3D printing in democratizing American manufacturing, innovation and entrepreneurship. Obama vowed: We’re going to rebuild our economy and restore our middle class, and give opportunities for people whose potential is not yet tapped. There are kids out there, there are adults out there right now who have a great idea. And they don’t have access to the capital they need. They don’t have the tools they need to put together a prototype. They don’t know how to link up with folks who could help refine those ideas. And what the Maker movement does, what technology does, what the information revolution does is it allows all those folks to suddenly be a part of this creative process. And what better place to do that than here in the United States of America? This is a place where we know how to invent and we know how to dream and we know how to take risks.2

Obama said: “I hope every company, every college, every community, every citizen joins us as we lift up makers and builders and doers across the country.”3 The President was hopeful that his innovation policy would help promote education, manufacturing and innovation policy in the United States. In terms of its methodology, this Chapter builds upon a growing comparative body of literature in respect of intellectual property (“IP”) and education. In Australia, Professor Sam Ricketson from the University of Melbourne, and Professor Ann Monotti from Monash University have written extensively about IP and universities.4 Michael Spence, the Vice-Chancellor of the University of Sydney, has also considered the claim of academics to

1 President Barack Obama, Remarks by the President at the White House Maker Faire, whIte house, June 18, 2014, available at https://obamawhitehouse.archives.gov/the-press-office/2014/06/18/ remarks-president-white-house-maker-faire (last visited Oct. 17, 2019). 2 Id. 3 Id. 4 See Ann Monotti & Sam Ricketson, Universities and Intellectual Property: Ownership and Exploitation (2003); Ann Monotti, Employment Law and Intellectual Property (Critical Concepts in Intellectual Property Law (2018).

447

448 Research handbook on intellectual property and technology transfer ownership of copyright and patents.5 In the United States, Corynne McSherry—now legal director at the Electronic Frontier Foundation—undertook groundbreaking work on IP and education.6 More recently, Professor Jacob Rooksby has sought to capture the increasing trends of commercialization of IP in higher education.7 There has been a fraught philosophical and ideological debate over IP and education. Professor Shubha Ghosh has considered the question of whether universities are special under IP law.8 He highlighted that universities occupy various rules—as owners, users, and enforcers of patent rights. James Bessen has considered the relationship between IP, innovation policy and education.9 Professor Joseph Stiglitz—the Nobel Laureate in Economics—has considered the relationship between IP and learning.10 This has been particularly evident in respect of allocation of patent rights within public sector research. Some individual researchers have maintained that they should be the proper owner of patented inventions in higher education. Universities and public research institutions have laid claim to patent ownership of scientific inventions. Industry partners have also emphasized their position as supports of researchers. Governments have sometimes exercised march-in rights in respect of publicly funded research. Advocates of open innovation have maintained that publicly funded research should be dedicated to the public domain, or shared under open licensing arrangements. There has been a scholarly debate about the extent to which IP law, policy and practice will be disrupted by the advent of 3D printing. There have been a number of scholarly monographs and collections on the topic of 3D printing regulation.11 On the topic of patent law, there has been a variety of different views as to the impact of 3D printing. Professor Mark Lemley of Stanford Law School has envisaged a world without artificial scarcity in light of new disruptive technologies—such as 3D printing.12 Timothy Holbrook is somewhat more apocalyptic—contending that 3D printing will undermine the capacity for patent enforcement.13 Nicole Syzdek has argued that there will be a staggered set of stages before 3D printing is accommodated within the patent system.14 By contrast, Geertrui Van Overwalle and

5 See Michael Spence, Workers in the “Groves of Academe”: The Claim of Academics to Copyright and Patents, oxford handbook Intell. prop. l. 829 (Rochelle Cooper Dreyfuss & Justine Pila eds., 2018). 6 See Corynne McSherry, Who Owns Academic Work?: Battling for Control of Intellectual Property (2003). 7 See Jacob H. Rooksby, The Branding of the American Mind: How Universities Capture, Manage, and Monetize Intellectual Property and Why It Matters (2016). 8 See Shubha Ghosh, Are Universities Special?, 49 akron l. rev. 671 (2015). 9 See James Bessen, Learning By Doing: The Real Connection Between Innovation, Wages, and Wealth (2015). 10 Joseph Stiglitz & Bruce Greenwald, Creating a Learning Society: A New Approach to Growth, Development, and Social Progress (2014); Dean Baker, Arjun Jayadev, & Joseph Stiglitz, Innovation, Intellectual Property, and Development: A Better Set of Approaches for the 21st Century, aCCessIbsa (2017). 11 Angela Daly, Socio-Legal Aspects of the 3D Printing Revolution (2016); Rosa Maria Ballardini, Marcus Norrgard & Jouni Partanen, 3D Printing, Intellectual Property and Innovation: Insights from Law and Technology (2016); Dinusha Mendis, Mark Lemley, & Matthew Rimmer, 3D Printing and Beyond: Intellectual Property and Regulation (2019). 12 Mark Lemley, IP in a World Without Scarcity, 90 N.Y.U. L. Rev. 460 (2015). 13 Timothy Holbrook, How 3D Printing Threatens Our Patent System, ConversatIon, Jan. 6, 2016. 14 Nicole Syzdek, Five Stages of Patent Grief to Achieve 3D Printing Acceptance, 49 u.s.f.l. rev. 335 (2015).

Intellectual property, higher education, technology transfer, and 3D printing

449

Reinout Leys have maintained that 3D printing is unexceptional, and will not disrupt the patent system.15 In contrast, the advocates of open innovation have championed 3D printing as a part of a commons-based movement, which challenges proprietary models of IP.16 Rather than survey all the various fields of IP,17 the Chapter focuses upon the debate over patent law and 3D printing in higher education. In particular, it compares and contrasts approaches taken to the topic of patent law and 3D printing in the United States, Canada, the European Union and Australia. Part II considers the theoretical discussion about 3D printing, the Maker Movement and education. Drawing upon the recent work of the World Intellectual Property Organization (“WIPO”) on breakthrough technologies, Part III explores the patent landscapes in respect of 3D printing and higher education. It highlights possible areas of specialization—including the creative arts and design; advanced manufacturing; health-care and bioprinting; and sustainable development. Part IV considers the vexed debate over patent ownership and licensing in higher education—looking at the context of 3D printing and additive manufacturing. This Part builds upon past work on IP and public sector research. It contends that the Maker Movement can learn previous conflicts over IP and education in the fields of information technology, biotechnology, access to medicines, clean technologies and Indigenous IP. Part V focuses upon the prospects of patent infringement in respect of 3D printing, and patent defenses and exceptions—such as the defense of experimental use, and possibly a new defense for the right of repair. Advocates for a Makers’ Bill of Rights have often called for the freedom to tinker, an expansion of the defense of experimental use, and the creation of a right of repair. In the conclusion, it is argued that universities should invest in open access institutions: makerspaces, fabspaces, 3D printing hubs, innovation networks and commons.

II.

3D PRINTING, THE MAKER MOVEMENT AND EDUCATION

The proponents of 3D printing and the Maker Movement have repeatedly emphasized the public benefits of the technology for education. It is worthwhile exploring the pedagogical philosophies behind the Maker Movement. It is also useful to develop a typology of different

15 Geertrui Van Overwalle & Reinout Leys, 3D Printing and Patent Law: A Disruptive Technology Disrupting Patent Law?, 48 Int’l rev. Intell. prop. & CoMpetItIon l. 504 (2017). 16 Jarkko Moilanen, Angela Daly, Ramon Lobato, & Darcy Allen, Cultures of Sharing in 3D Printing: What Can We Learn from the Licence Choices of Thingiverse Users?, 6 J. peer produCtIon dIsruptIon & l. (2015); Creative Commons, State of the Commons Report, 2016, available at https:// stateof.creativecommons.org/ (last visited Mar. 29, 2019); Paul Stacey & Sarah Hinchcliff Pearson, Made with Creative Commons (2017); David Bollier, To Find Alternatives to Capitalism, Think Small: Why Co-Ops, Regional Currencies, and Hackerspaces are pointing the way toward a new Economic Vision, natIon, Aug. 9, 2017. 17 For a discussion of copyright law and 3D printing, see Matthew Rimmer, The Maker Movement: Copyright Law, Remix Culture, and 3D Printing, 41 u. w. austl. l. rev. 51 (2017). For an analysis of designs law and 3D printing, see Mitchell Adams, The “Third Industrial Revolution”: 3D Printing Technology and Australian Designs Law, 24 J. l., Info. & sCI. 56 (2015–2016). For an evaluation of trade mark law and 3D printing, see Amanda Scardamaglia, Flashpoints in 3D Printing and Trade Mark Law, 23 J. l., Info. & sCI. 30 (2015). For a brief overview of trade secrets and 3D printing, consider Bryan Vogel, The Other IP: Trade Secret Law and 3D printing, robIns/ kaplan, Dec. 3, 2014.

450 Research handbook on intellectual property and technology transfer forms of innovation spaces—such as makerspaces, Fab Labs, TechShops and other innovation hubs. In his 2012 book, Makers, Chris Anderson contended that the rise of 3D printing was also leading to a revolution in education and pedagogy.18 He stressed: “As desktop fabrication tools go mainstream, it’s time to return ‘making things’ to the high school curriculum, not as the shop class of old, but in the form of teaching design.”19 Chris Anderson envisaged a reconstruction of the classroom: [I]magine if each design classroom had a few 3-D printers or a laser cutter. All those desktop design tools have a “Make” menu item. Kids could actually fabricate what they have drawn onscreen. Just consider what it would mean to them to hold something they dreamed up. This is how a generation of Makers will be created. This is how the next wave of manufacturing entrepreneurs will be born.20

Anderson considered the rise of makerspaces in university education—such as the evolution of Fab Labs from Neil Gershenfeld’s Center for Bits and Atoms at MIT.21 Anderson also considered the development of the Manchester Fab Lab in the United Kingdom. He contended that such an educational revolution could spark the next industrial revolution. In his influential 2005 book, Fab, Neil Gershenfeld described pedagogy at MIT, teaching “How to Make (Almost) Anything.”22 He observed that the students in the course “were doing much more than taking a class; they were inventing a new physical notion of literacy.”23 The innovative teaching at MIT then led to the development of the Fab Lab movement. In the 2017 book, Designing Reality, Neil Gershenfeld, Alan Gershenfeld and Joel Cutcher-Gershenfeld consider the evolution of the Fab Lab model.24 The writers observed: Fab labs are laboratories for fabrication (which we also think are fabulous labs). They began as an outreach project from MIT’s Center for Bits and Atoms (CBA) … CBA was founded to study the boundary between computer science and physical science.25

The writers observed: Each cycle would propagate best practices throughout the fab lab network, building a core collaborating community of local mentors and providing a cohort of trained students that then became available to help with new labs and programs.26

The writers observed the connections between communication, computation, fabrication and learning:

18

Chris Anderson, Makers: The New Industrial Revolution (2012). Id. at 55. 20 Id. at 20–1. 21 Id. at 46. 22 Neil Gershenfeld, Fab: The Coming Revolution of Your Desktop—From Personal Computers to Personal Fabrication 5 (2005). 23 Id. at 7. 24 Neil Gershenfeld, Alan Gershenfeld, & Joel Cutcher-Gershenfeld, Designing Reality: How to Survive and Thrive in the Third Digital Revolution (2017). 25 Id. at 18. 26 Id. at 27. 19

Intellectual property, higher education, technology transfer, and 3D printing

451

Whereas digital communication let us interact globally and digital computation lets us share knowledge, the addition of digital fabrication lets us exchange things as well as ideas.27

The leaders of the Fab Lab movement contend that “digital fabrication is part of a new teaching and learning ecosystem.”28 Fab Labs engage in a model of project-based, student-centered learning.29 In his 2014 book—The Maker Movement Manifesto—Mark Hatch, then the CEO of TechShop, contended that 3D printing would help transform the educational system in the United States.30 He observed: The entire educational system in the United States is outdated—built for a world that no longer exists, in a world that is continuing to change very rapidly. We have an incredible opportunity— and responsibility—to explore what education means in a fully networked, Internet-enabled, and makerspace-fueled world. Creating innovators and technology entrepreneurs should be one of education’s top priorities.31

In his 2017 book, The Maker Revolution, Mark Hatch dedicates a chapter to the topic of education.32 He contended that the “Maker Movement is helping to create a revolution in US education institutions … [with] the move to ‘constructivism’, ‘project-based education,’ ‘the flipped classroom,’ and ‘maker education’.”33 He emphasized that the Maker Movement was making education more experiential. Hatch considered proposals for classroom makerspare laboratories. He was hopeful that the Maker Movement would not only transform education, training and employment, but also manufacturing, industry and innovation. However, the model of TechShop itself proved to be a failure, with the company going into bankruptcy in 2017.34 In the 2016 text, Free to Make, the creator of MAKE Magazine and the Maker Faire, Dale Dougherty, has written about the rise of the Maker Movement with Ariane Conrad.35 He stressed the importance of the Maker movement to a reconceptualization of education: As more parents, teachers, principals, schools and school districts recognize the role for making in education, we have seen a renaissance in exploring alternative ideas about learning that come not so much from current research but from old sources.36

Dougherty commented:

27

Id. at 27. Id. at 152. 29 Julia Walter-Herrmann & Corinne Buching, FabLab: Of Machines, Makers and Inventors (2013). 30 Mark Hatch, The Maker Movement Manifesto: Rules for Innovation in the New World of Crafters, Hackers, and Tinkerers (2013). 31 Id. at 202. 32 Mark Hatch, The Maker Revolution: Building a Future on Creativity and Innovation in an Exponential World (2018). 33 Id. at 109. 34 Dan Woods, TechShop Closes Doors, Files Bankruptcy, Make, Nov. 15, 2017. 35 Dale Dougherty & Ariane Conrad, Free to Make: How the Maker Movement is Changing our Schools, Our Jobs, and Minds (2016). 36 Id. at 181. 28

452 Research handbook on intellectual property and technology transfer Maker-centered learning is “pull”. Students decide what they want to do. Making is being pulled into schools by educators because they see it as a very effective way to engage students in learning. They do it freely because of what it means to them and their students. This kind of change is being driven by parents, teachers and administrators—and students—who see themselves empowered to make change throughout the educational system. Research is not leading the way; nor is theory. Actions and practice are.37

With Peter Hirshberg and Marcia Kadanoff, Dale Dougherty has also considered the role of 3D printing and the Maker Movement in reinventing American cities.38 This book considers the importance of education in this study: “Instead of focusing on imparting content to students, educators are focused on enabling students to learn more collaboratively and to focus on solving real world problems of interest to them.”39 In her empirical work, Sarah R. Davies considered 3D printing both as a force of emancipation and a subject of commodification.40 She commented that “hacking and making were seen as relating to wider educational goals.”41 Davies commented: “Participation in a hackerspace might help to boost interest in science and technology, give you skills that related to your formal education, or recruit young people into STEM professions.”42 In his book The Zero Marginal Cost Society, Jeremy Rifkin considered the rise of 3D printing, the Internet of Things and collaborative capitalism.43 Chapter 7 of his book explores the transformation of education in the age of the Maker Movement: Preparing students for an era in which capitalist markets play a secondary role to the Collaborative Commons is beginning to force a rethinking of the educational process itself. The pedagogy of learning is undergoing a radical overhaul. So too is the way education is financed and delivered.44

Rifkin contends that “[t]he transition from the capitalist era to the Collaborative Age is altering the pedagogy of the classroom.”45 In his view, “[t]he authoritarian, top-down model of instruction is beginning to give way to a more collaborative learning experience.”46 Rifkin imagined: “In the Collaborative Age, students will come to think of knowledge as a shared experience among a community of peers.”47 He noted that a number of traditional universities and colleges were anxious to press forward: Like their counterparts in so many other sectors where new technologies are making possible a near zero marginal cost society and nearly free goods and services, they realize that the logic of optimiz-

37

Id. at 205. Peter Hirshberg, Dale Dougherty, & Marcia Kadanoff, Maker City: A Practical Guide for Reinventing American Cities (2016). 39 Id. at 80–1. 40 Sarah R. Davies, Making the Maker Movement (2017). 41 Id. at 215. 42 Id. at 125. 43 Jeremy Rifkin, The Zero Marginal Cost Society: The Internet of Things, the Collaborative Commons, and the Eclipse of Capitalism (2014). 44 Id. at 109. 45 Id. 46 Id. at 110. 47 Id. 38

Intellectual property, higher education, technology transfer, and 3D printing

453

ing the welfare of the human race in collaborative, networked Commons is so compelling that it is impossible to shut away.48

As part of their work on 3D printing and the new industrial revolution, social scientists Thomas Birtchnell and John Urry have considered how 3D printing has reconfigured the higher education sector.49 The pair claimed that “universities—as producers, diffusers and preservers of new knowledge—are inching ever closer to the material economy due to the emergence in recent years of 3D printing within research centers, design schools, laboratories and even academic libraries.”50 Birtchnell and Urry suggested that “university libraries are the forerunners in the convergence between material and informational intellectual capital.”51 They suggested that “this innovation is in response to the demand from engineering and design students for rapid prototyping tools.”52 Universities and public research institutions are responding to such demand “by purchasing and making available 3D printers to all students, staff and researchers.”53 Some pathbreaking institutions are establishing 3D model file collections.54 However, Birtchnell and Urry have noted that such institutions will also face risks and issues—including ones related to IP. They commented: In reorganizing themselves as triple helix collaborators and aligning the production of knowledge assets (inventions, patents and designs) with industrial capacities of materialization, universities become hubs, partners, or incubators in the convergent knowledge and material economy of the twenty-first century.55

Increasingly, there has been work on the Maker Movement in the discipline of education.56 Kylie Pepper, Erica Rosenfeld Halverson and Yasmin Kafai have commented that “the Maker Movement presents unique and timely opportunities toward reimagining the future of learning in ways that will resonate with the time-honored traditions of a democratic education, yet pushes us forward to embrace the necessary changes inherent to the 21st century.”57 They stress that there is a debate over open education in respect of the Maker Movement: “If we want

48

Id. at 119. Thomas Birtchnell & John Urry, A New Industrial Future? 3D Printing and the Reconfiguring of Production, Distribution, and Consumption (2016). 50 Id. at 51. 51 Id. 52 Id. 53 Id. 54 See for instance, Browniverse: 3D Models and Printing Digital Library, brown u. lIbr., available at https://repository.library.brown.edu/studio/collections/id_756/ (last visited Mar. 29, 2019); Free 3D Models, or. st. u., available at https://guides.library.oregonstate.edu/c.php?g=286201&p=1905549 (last visited Mar. 29, 2019); and 3D Printing in the Biomedical Library: Free 3D Models, u. pa., available at https://guides.library.upenn.edu/c.php?g=476257&p=3256090 (last visited Mar. 29, 2019). 55 Birtchnell & Urry, supra note 49, at 52. 56 Kylie Pepper, Erica Rosenfeld Halverson, & Yasmin Kafai, Makeology: Makerspaces as Learning Environment (2016); Kylie Pepper, Erica Rosenfeld Halverson, & Yasmin Kafai (ed.), Makeology: Makers as Learners (2017). 57 Kylie Pepper, Erica Rosenfeld Halverson, & Yasmin Kafai, “Introduction” in Makeology: Makers as Learners 4 (Kylie Pepper, Erica Rosenfeld Halverson, & Yasmin Kafai eds., 2017). 49

454 Research handbook on intellectual property and technology transfer making to be a global activity, we need to make sure that who is making and what is being made is open and accessible to everyone.”58 The role of 3D printing is expanding in the area of IP and education. In his 2016 book on The Branding of the American Mind, Jacob Rooksby considers the position of 3D Printing in light of wider debates over IP, commercialization and higher education.59 He comments that “3D-printer technology—which enables users to envision and then create products of all kinds, from paper clips to car engines—has emerged as a source of creative confusion for some institutions.”60 Rooksby notes that educational institutions have been adopting 3D printers to enhance the experience of students—both undergraduate and postgraduate. He worried: “As 3D technology and the MakerSpace movement develop, more student entrepreneurship projects risk being put on hold or shelved entirely … due to IP policies that lack clarity or inappropriately purport to broadly apply to student activities.”61 Thus, it is important to consider a range of IP considerations surrounding the introduction of 3D printing into universities, educational institutions and public research organizations.

III.

PATENT LANDSCAPES OF 3D PRINTING AND EDUCATION

The WIPO undertook a study of IP and breakthrough technologies in 2015.62 In particular, the report looks at the disruptive technologies of 3D printing, robotics and nanotechnology. The report highlights the role played by the public sector in the development of such breakthrough technologies: The greater prominence of universities and [Public Research Organizations] PROs in patent landscapes may partly reflect policy efforts to better harness the results of scientific research for commercial development. However, those policy efforts arguably recognize the critical role that upstream research plays in downstream technological progress. While academic patenting has become more prominent across most of the major patenting origins, there are also notable differences. In the case of Japan, universities and PROs never account for more than 10 percent of total first filings. By contrast, China generally shows the highest shares of academic patenting, exceeding 70 percent for nanotechnology and 50 percent for robotics. On the one hand, this may suggest more limited R&D capacity in Chinese firms in the relevant technology fields, which may imply a lower rate of technology commercialization.63

Public sector research institutions play a critically important role in the development of emerging technologies—such as 3D printing, robotics and nanotechnology. The report devotes a significant amount of consideration to 3D printing and additive manufacturing. The report highlights the significant role played by universities as a catalyst in respect of 3D printing technologies:

58

Id. at 6. See Rooksby, supra note 7. 60 Id. at 180. 61 Id. at 181. 62 World IP Report: Breakthrough Innovation and Economic Growth, geneva: wIpo, 2015, available at https://www.wipo.int/edocs/pubdocs/en/wipo_pub_944_2015.pdf (last visited Oct. 17, 2019). 63 Id. at 14. 59

Intellectual property, higher education, technology transfer, and 3D printing Table 21.1

455

Top ten university and public research organization patent applications since 1995

University Name

Country

Number of First Patent Filings

Fraunhofer Society

Germany

89

Chinese Academy of Sciences

China

79

Huazhong University of Science & Technology

China

46

MIT

US

37

Xi’an Jiaotong University

China

34

University of Southern California

US

31

South China University of Technology

China

27

Harbin Institute of Technology

China

24

TNO

Netherlands

24

Beijing University of Technology

China

17

Note: WIPO, available at https://www.wipo.int (last visited Mar. 29, 2019).

Universities are increasingly participating in this field—albeit at a much lower share than firms. In fact, a couple of the more important 3D printing processes originated from MIT and the University of Texas System, particularly the University of Texas, Austin. To this day, these two universities own considerable patent portfolios in the field. However, these university patents are usually licensed out to private firms for commercialization. For example, the inkjet 3D printing technology developed by MIT was licensed to several firms for their own application and commercialization.64

Nonetheless, the report also recognizes that there are major commercial investments in respect of 3D printing. Conglomerates like General Electric and Siemens, manufacturing companies like Mitsubishi, aerospace companies like Boeing, and specialist 3D printing companies such as Stratasys and 3D Systems, dominate the rankings of patent filings. The report relies upon Patstat data to identify top university and public research organizations engaged in 3D printing (see Table 21.1). The report usefully analyses the proportion of universities and public sector organizations filing patent applications in China, the United States, Germany and Japan.65 The report highlights the high proportion of academic patenting of 3D printing in China: In China the government has made large strategic investments in 3D printing technologies; these are more important in advancing innovation than company driven R&D. The heavy investment in 3D printing by the Chinese government is reflected in the number of patent applications filed by Chinese universities; in some cases these filings exceed those of US and European universities.66

This WIPO report reflects also the pattern of investment and acquisition by China in respect of emerging technologies. There has been a high level of patent applications by universities and public sector organizations in respect of 3D printing in China. There has been increasing research in respect of IP and 3D printing in China.67 As part of China’s innovation policy, Made in China 2025, there has been significant investment in

64

Id. at 100. Id. at 101. 66 Id. 67 Liu Xin & Yu Xiang, Potential Challenges of 3D Printing Technology on Patent Enforcement and Considerations for Countermeasures in China, 20 J. Intell. prop. rts 155 (2015); Angela Daly, et al., 65

456 Research handbook on intellectual property and technology transfer emerging technologies—such as advanced manufacturing, robotics and 3D printing. This has involved, amongst other things, a significant generation of IP assets in such areas.68 There has been much consternation amongst the Trump Administration about China’s innovation policy.69 This has in part been a factor behind the trade action by the Trump Administration against China in the World Trade Organization.70 There has been much academic debate over IP compliance in contemporary China.71 The report also highlights the importance of public-private partnerships in respect of 3D printing.72 The report observes the networked approach to innovation: Government initiatives also serve a second role—to provide linkages between the different actors in the ecosystem. Many of these initiatives bring together researchers in academia and the private sector along with manufacturers with the intention of diffusing the innovation throughout the economy. The US, for example, has poured USD 50 million into a public-private partnership to bring 3D printing technologies into mainstream manufacturing. This partnership brings together 50 companies, 28 universities and research labs and 16 other organizations. A similar initiative was recently announced by the Australian government that would bring together 14 manufacturing firms, 16 local universities, 4 industry agencies, the Australian federal agency for scientific research and the Fraunhofer Institute for Laser Technology. One of the manufacturing firms involved in the initiative is SLM Solutions GmbH, a German 3D printing manufacturer.73

Such collaborations have been of importance in bringing together industry, government and universities. There has been a discussion as to whether public-private partnerships are the best means to promote the governance of IP and sustainable development.74 A 2017 study of patents and 3D printing by John Hornick highlighted the growth and diversification of 3D printing applications: The field of 3D printing has been growing rapidly for years. It has applications in many areas of life and the economy, such as healthcare, aerospace, and parts replacement. 3D printing also reshapes supply chains and democratizes manufacturing. Fueled by this growth, 3D printing-related patent filings are trending upward.75

3D Printing and Intellectual Property Futures, Mar. 20, 2014; Hing Jai Chan, Hui Leng Choo, Onyeka Osuji, & James Griffin, Intellectual Property Rights and Emerging Technology: 3D Printing in China (2018). 68 Javier Hernandez, Seeking Greater Global Power, China Looks to Robots and Microchips, n.y tIMes, Aug. 14, 2017. 69 Matt Sheehan, Trump’s Trade War Isn’t About Trade, It’s About Technology, MarCo polo, Apr. 3, 2018. 70 United States Trade Representative, Section 301 Report into China’s Acts, Policies, and Practices Related to Technology Transfer, Intellectual Property, and Innovation, Mar. 2018; United States Trade Representative, Following President Trump’s Section 301 Decisions, USTR Launches New WTO Challenge Against China, Mar. 2018. 71 Kristie Thomas, Assessing Intellectual Property Compliance in Contemporary China: The World Trade Organization TRIPS Agreement (2017). 72 WIPO, supra note 62, at 102. 73 Id. 74 Margaret Chon, Pedro Roffe, & Ahmed Abdel-Latif, The Cambridge Handbook of Public-Private Partnerships, Intellectual Property Governance, and Sustainable Development (2018). 75 John Hornick, 3D Printing Patent Landscape, 3DPrInt (July 17, 2017), available at https:// 3dprint.com/181207/3d-printing-patent-landscape/ (last visited Oct. 17, 2019).

Intellectual property, higher education, technology transfer, and 3D printing

457

He observed: “Companies are obtaining far more patents than individuals, non-profits and universities combined.”76 The private sector is increasingly outstripping the public sector when it comes to 3D printing patents and applications. A new 2018 study by IFI Claims Patent Services provides further insight into recent trends in respect of patent law and 3D printing.77 This study showed that additive manufacturing—as defined by the Patent Classification B33Y—had one of the highest growth rates over the past years (only e-cigarettes was higher). The study highlighted that commercial companies were dominating the list of patent applicants in 2017. Key patent filing entities included General Electric, Xerox, Boeing, Desktop Metal, Hewlett-Packard Development, Ricoh Co and Stratasys. This study raises the question whether universities, educational institutions and public sector research organizations will be able to compete in patent races with commercial players—comprising both mainstream manufacturing companies and specialist 3D printing companies.

IV.

PATENT OWNERSHIP, COLLABORATION AND 3D PRINTING

It is worthwhile to compare how various jurisdictions have dealt with the allocation of patent rights in the settings of universities, educational institutions and public research organizations. The United States has sought to foster collaboration and co-operation between universities, industry and government through new institutions—such as America Makes. Canada has sought to develop a new innovation policy—with superclusters—together with a new IP policy. The European Parliament has sought to develop a comprehensive approach to IP, civil liability and 3D printing. Australia has grappled with problems with IP ownership and management. Australia has sought to overcome past innovation gaps with a new approach to innovation policy. A.

United States

The adoption of the University and Small Business Patent Procedures Act 1980 (US) (the “Bayh-Dole Act”) has certainly had a transformative impact upon IP and commercialization in higher education in the United States.78 The regime, though, has been somewhat inflexible in terms of dealing with new emerging technologies. Historically, the field of IP and biotechnology has seen conflicts between the public sector and the private sector. Sally Smith Hughes has documented the origins of the biotechnology industry, and the issues arising in respect of patents, politics and commercialization.79 There were conflicts between the University of California and Genentech in respect of patents and 76

Id. Michael Petch, Interview: New Study of 3D Printing Patents Reveals Second Fastest Growing Technology of 2017, 3d prIntIng Industry (Jan.10, 2018), available at https://3dprintingindustry.com/ news/interview-new-study-shows-3d-printing-second-fastest-growing-technology-2017-127179/ (last visited Oct. 17, 2019). 78 Wendy Schacht, The Bayh-Dole Act: Selected Issues in Patent Policy and the Commercialization of Technology, CRS Report for Congress, Apr. 3, 2008. 79 Sally Smith Hughes, Genentech: The Beginnings of Biotech (2011). 77

458 Research handbook on intellectual property and technology transfer biotechnology—relating to human growth hormone.80 Public researchers were involved in patent races with private biotechnology firms—like Myriad Genetics—in respect of genetic testing.81 There have been significant clashes in respect of the ownership patents and stem cell research.82 More recently, there have been patent races between competing universities in respect of CRISPR—gene-editing technologies.83 There has also been controversies over access to essential medicines, which had been developed by public institutions.84 The student-based group Universities Allied for Essential Medicines (UAEM) has pushed for flexible licensing in respect of publicly-funded humanitarian research.85 Federal courts have sometimes had to resolve these disputes. Notably, in the case of Stanford University v. Roche Molecular Systems, Inc., the Supreme Court of the United States held that title to a patented invention vests in the inventor, even if the inventor is a researcher at a federally funded laboratory under the Bayh-Dole Act.86 There has also been tensions over public sector research in respect of IP, technology transfer and clean technologies—and whether such work should be flexibly licensed.87 Universities and research institutions have had to engage with benefit-sharing agreements in respect of access to genetic resources.88 There has also been legal and ethical conundrums arising in respect of university research and Indigenous IP.89 The argument of this Chapter is that the Maker Movement needs to learn from some of the past precedents relating to IP and higher education, and avoid some of the past conflicts and disputes between the various stakeholders. President Barack Obama sought to promote a co-operative approach between industry, government and universities in respect of building capacity for 3D printing and advanced manufacturing. In the 2011 Advanced Manufacturing Report, the White House has called for new commitment by the administration to manufacturing.90 In response, Obama established

80 Matthew Rimmer, Genentech And The Stolen Gene: Patent Law and Pioneer Inventions, 5 bIo-sCI. l. rev. 198 (2002/2003). 81 Matthew Rimmer, “An Exorbitant Monopoly: The High Court of Australia, Myriad Genetics, and Gene Patents” in Research Handbook on Intellectual Property and the Life Sciences 56 (Duncan Matthews & Herbert Zech, eds, 2017). 82 Matthew Rimmer, The Attack Of The Clones: Patent Law And Stem Cell Research, 10 J. l. & Med. 488 (2003); Matthew Rimmer, The Last Taboo: Patenting Human Beings, 14 expert opInIon on therapeutIC patents 1061 (2004). 83 Jacob Sherkow, The CRISPR Patent Landscape: Past, Present, and Future, CrIspr J. (2018). 84 Thomas Pogge, Matthew Rimmer, & Kim Rubenstein, Incentives for Global Public Health: Patent Law and Access to Medicines (2010). 85 Universities Allied for Essential Medicines, available at https://uaem.org/ (last visited Mar. 29, 2019). 86 Stanford University v. Roche Molecular Systems, Inc., 563 U.S. 776 (2011). 87 Matthew Rimmer, Intellectual Property and Climate Change: Inventing Clean Technologies (2011); Matthew Rimmer, Intellectual Property and Clean Energy: The Paris Agreement and Climate Justice (2018). 88 Matthew Rimmer, The Sorcerer II Expedition: Intellectual Property and Biodiscovery, 6 MaCQuarIe J. Int’l & CoMparatIve envtl l. 147 (2009). 89 Matthew Rimmer, The Genographic Project: Traditional Knowledge and Population Genetics, 11 australIan IndIgenous l. rev. 33 (2007); Matthew Rimmer, Indigenous Intellectual Property: A Handbook of Contemporary Research (2015). 90 President’s Council of Advisors on Science and Technology, Report to the President on Ensuring American Leadership in Advanced Manufacturing, whIte house, June 2011, available at

Intellectual property, higher education, technology transfer, and 3D printing

459

a number of new manufacturing institutes. One of these new institutions is America Makes, the National Additive Manufacturing Innovation Institute.91 The organization notes: “Structured as a public-private partnership, we innovate and accelerate AM/3DP to increase our nation’s global manufacturing competitiveness.”92 In a new study of advanced manufacturing, William Bonvillian and Peter Singer have provided an analysis of the Advanced Manufacturing Institute model developed under Obama.93 Bonvillian and Singer evaluated the consortium approach taken in the development of the America Makes program.94 They commented that the Institute has worked to “create an infrastructure for the sharing of additive manufacturing ideas and research on development and evaluation of additive manufacturing technologies, on linking small and midsize firms with resources to enable them to use additive manufacturing.”95 Bonvillian and Singer acknowledge that the IP issues have been complicated: While information and IP sharing among the highly competitive larger aerospace firms proved complex, which has affected technology development, the institute has played a significant role in convening the new 3D printing community, helping participants learn which researchers and firms are working on such advances, thereby promoting connections and contracting.96

One of the key beneficiaries of America Makes are universities, educational institutions and public research organizations. In addition to raising specialist questions about patent law, bioprinting has posed particular regulatory issues.97 B.

Canada

In response to America Makes, Prime Minister Justin Trudeau’s Government has introduced the Innovation Superclusters Initiative.98 The Minister of Small Business and Tourism Bardish Chagger commented on the initiative: These five superclusters will help keep Canada as a world leader in innovation. And the technologies and ideas they will produce will transform our world and improve our quality of life. This is not just a Canadian endeavor; these superclusters will also have global reach. In the years to follow, this initiative will create well-paying jobs, give rise to new and groundbreaking products and solutions,

https://obamawhitehouse.archives.gov/sites/default/files/microsites/ostp/pcast-advanced-manufacturing -june2011.pdf (last visited Oct. 17, 2019). 91 America Makes, available at https://www.americamakes.us/ (last visited Mar. 29, 2019). 92 Id. 93 William Bonvillian & Peter Singer, Advanced Manufacturing: The New American Innovation Policies (2018). 94 Id. at 140–2. 95 Id. at 143. 96 Id. at 143. 97 Jamil Ammar, The “Medical Mile” Gearing Toward 3D-Bespoke Healthcare—A Comparison of United States and European Union Patent Regimes, 52 gonz. l. rev. 279 (2017). 98 Innovation Superclusters Initiative, gov’t Ca., 2018, available at https://www.canada.ca/ en/innovation-science-economic-development/programs/small-business-financing-growth/innovation -superclusters.html (last visited Oct. 17, 2019).

460 Research handbook on intellectual property and technology transfer and put Canada front and center in the global innovation race. That’s great news—for our economy, for Canada and especially for Canadian small business owners.99

There will be an Advanced Manufacturing Supercluster based in Ontario, Canada.100 Such an endeavor will no doubt include 3D printing and advanced manufacturing. There will be complicated issues involved in managing IP associated with such superclusters.101 Richard Gold, a McGill University law professor, commented: “They’ll have to move from fairly high-level statements about what they’re going to do to actually implementing it.”102 IP lawyer Jim Hinton commented: “Now, it isn’t clear to me that this framework ensures economic growth for the country.”103 IP lawyer Peter Cowan worried that foreign technology companies would benefit most from the superclusters: “They’ve contributed money and they want to make sure they have access to IP that will benefit them … Canadian ventures might not be able to compete.”104 The Canadian Government has also sought to improve its IP strategy. In 2018, the Government of Canada announced that it was investing $CA 85.3 million over five years “to help Canadian businesses, creators, entrepreneurs and innovators understand, protect and access IP through a comprehensive IP Strategy.”105 The first part of the strategy was to improve IP awareness, education and advice.106 This would involve building expertise through learning and education; IP awareness; IP legal clinics; and a new team of IP advisors. Professor Myra Tawfik from Windsor Law School emphasized that this strategy would address a significant gap in Canada’s innovation policy.107 She commented: Increasingly, the Canadian economy is dependent on Canadians generating new ideas and knowledge and retaining their value in this country. Intellectual property rights secure our creative and innovative efforts so that we can extract commercial value from them. In this way, intellectual property . . . is increasingly becoming the most important global currency. Canadian law and policy need to ensure that Canadian creators and innovators have all the necessary tools to successfully leverage their IP to competitive advantage.108

Tawfik commented:

99 The Hon. Bardish Chagger, Superclusters Announcement Closing Remarks, gov’t Ca., Feb. 15, 2018, available at https://www.canada.ca/en/innovation-science-economic-development/news/2018/02/ superclusters-announcement-closing-remarks.html (last visited Oct. 17, 2019). 100 Canada’s New Superclusters, gov’t Ca., available at https://www.ic.gc.ca/eic/site/093.nsf/eng/ 00008.html (last visited Mar. 29, 2019). 101 Sean Silcoff, Is Ottawa’s “Supercluster” Funding Initiative a Superboondoggle in the Making, globe & MaIl, Mar. 16, 2018. 102 Id. 103 Id. 104 Id. 105 Intellectual Property Strategy, gov’t Ca., 2018, available at https://www.ic.gc.ca/eic/site/108 .nsf/eng/home (last visited Oct. 17, 2019). 106 Id. 107 Myra Tawfik, Why It’s So Important for Canadians to be able to Leverage their Intellectual Property, globe & MaIl, Apr. 27, 2018. 108 Id.

Intellectual property, higher education, technology transfer, and 3D printing

461

The IP Strategy recognizes that if Canadian innovators possess only a rudimentary understanding of IP they will not succeed against global competitors who possess greater IP strategic skills and enjoy better access to a network of experienced expert advisors.109

Second, the IP Strategy considered strategic IP tools for growth. There was a proposal for expedited IP dispute resolution; an IP and standards-setting strategy; and an IP marketplace. Most striking of all has been the proposal for a pilot project on a patent collective: This collective, bringing together firms through a membership model, is a pilot initiative to support small to medium-sized firms in coming together to facilitate better IP outcomes for collective members. A third party, chosen through a competitive request for proposal process, will work with firms to promote IP best practices, provide patent intelligence and support, and obtain access to patents to help remove barriers to firms’ growth.110

Jorge Contreras has considered the possible structure and goals of such a collective, and examined various measures to support Canadian innovators.111 He contended that: the proposed Canadian patent collective avoid the acquisition and aggregation of patents and instead focus its limited resources on the following three supportive functions for Canadian industry: assist Canadian firms, through subsidies or other resource commitments, to participate in existing international defensive patent networks; encourage Canadian universities and research institutions to focus on commercially relevant “translational” research, and possibly to shift Canadian research funding priorities in this direction; and assess the potential benefits of facilitating patent sharing or pooling arrangements in select Canadian industries, and offering administrative and infrastructural support for such efforts.112

Contreras argued that such a focused model would be of most benefit: “As such, the Canadian patent collective could focus on adding value where it is truly needed to achieve collective benefits for Canadian industry and technology.”113 There was a need to ensure that the regime for a Canadian patent collective was consistent with competition policy and international trade law. Third, the IP Strategy promised law reform in respect of Canadian IP laws. A number of proposals were mooted—including minimum requirements for patent demand letters; a patent research exemption (which will be discussed in the next part on patent infringement and the defense of experimental use); and rules on standard essential patents.114 Moreover, there were reforms mooted for IP licenses in bankruptcy proceedings; and better IP agent governance. However, as a result of the United States-Mexico-Canada-Agreement 2018, Canada’s policy flexibilities in respect of IP will be significantly limited and reduced.115

109

Id. Intellectual Property Strategy, gov’t Ca., 2018, available at https://www.ic.gc.ca/eic/site/108 .nsf/eng/home (last visited Oct. 17, 2019). 111 Jorge Contreras, Considerations Regarding a Canadian Patent Collective, Centre for International Governance Innovation, CIgIonlIne, May 2, 2018, available at https://www.cigionline.org/publications/ considerations-regarding-canadian-patent-collective (last visited Oct. 17, 2019). 112 Id. at 9. 113 Id. 114 Intellectual Property Strategy, gov’t Ca., 2018, available at https://www.ic.gc.ca/eic/site/108 .nsf/eng/home (last visited Mar. 29, 2019). 115 United States-Mexico-Canada-Agreement, Off. of the U.S. Trade Representative, 2018 https:// ustr.gov/trade-agreements/free-trade-agreements/united-states-mexico-canada-agreement/united-states 110

462 Research handbook on intellectual property and technology transfer C.

European Union

The European Union has been a research leader in 3D printing, advanced manufacturing and additive manufacturing.116 There are a number of instances of innovative projects involving 3D printing in the European Union. The Eindhoven University of Technology in the Netherlands has been involved in innovative 3D printing of houses and architecture.117 There is a major bioprinting consortium based at AMBER in Ireland—linked with Trinity College Dublin.118 German research institutions have significant capacity in respect of advanced manufacturing—including in respect of 3D printing.119 The European Space Agency has investigated 3D printing for more efficient lunar base construction.120 There has been a 3D printing food project to help people with chewing difficulties.121 In 2017, the European Parliament published a Working Document titled Three-Dimensional Printing, a Challenge in the fields of IP Rights and Civil Liability.122 The publication stressed that the European Commission “has made 3D printing one of the priority areas of technology.”123 In June 2018, the European Parliament put forward a report on 3D printing—a challenge in the fields of IP and 3D printing.124 The report provides a historical overview of 3D printing: On an experimental level, three-dimensional printing (“3D printing”) dates back to the 1960s. Initially developed in the United States, 3D-printing technology started to break through into industry in the early 1980s. Not long after the technology had been developed, 3D printers began to hit the market, with companies offering both digital models and 3D printing services. 3D printing is, in fact, a general

-mexico; James McLeod, Canada “Caved on Intellectual Property Issues: Critics of Trade Deal, fIn. post, Oct. 1, 2018; Michael Geist, From Copyright Term to Super Bowl Commercials: Breaking Down the Digital NAFTA Deal, MIChael geIst, Oct. 1, 2018; Jeremy de Beer, Intellectual Property Chapter of USMCA Proves Canada’s Pragmatism, toronto star, Oct. 4, 2018. 116 Report on 3D Printing: Current and Future Application Areas, Existing Industrial Value Chains and Missing Competences in the EU, eu.CoMM’n, Sept. 20 2016, available at https://ec.europa.eu/ growth/content/report-3d-printing-current-and-future-application-areas-existing-industrial-value-chains -0_en. 117 Daniel Boffey, Netherlands to Build World’s First Habitable 3D printed Houses, guardIan, June 6, 2018. 118 AMBER, available at http://ambercentre.ie/ (last visited Mar. 29, 2019). 119 WIPO, supra note 62. 120 Lunar 3D printing, eu. spaCe agenCy, available at http://www.esa.int/Highlights/Lunar_3D _printing (last visited Mar. 29, 2019) 121 3D Food Printing Technology to Help People with Chewing Difficulties, horIzon 2020, Apr. 9, 2014, available at https://ec.europa.eu/programmes/horizon2020/en/news/3d-food-printing-technology -help-people-chewing-difficulties (last visited Oct. 17, 2019). 122 Committee on Legal Affairs, Working Document: Three-Dimensional Printing, a Challenge in the fields of Intellectual Property Rights and Civil Liability, eur. parl., Nov. 23, 2017, available at http:// www.europarl.europa.eu/sides/getDoc.do?type=COMPARL&reference=PE-612.302&format=PDF& language=EN&secondRef=01 (last visited Oct. 17, 2019). 123 Id.. 124 Report on Three-Dimensional Printing, a Challenge in the Fields of Intellectual Property Rights and Civil Liability, eur. parl., June 26, 2018, available at http://www.europarl.europa.eu/sides/getDoc .do?pubRef=-%2F%2FEP%2F%2FTEXT%2BREPORT%2BA8-2018-0223%2B0%2BDOC%2BXML %2BV0%2F%2FEN&language=EN (last visited Oct. 17, 2019).

Intellectual property, higher education, technology transfer, and 3D printing

463

term covering several types of technology for manufacturing physical objects in a range of materials based on a digital file and using a 3D printer. They were initially designed to make prototypes and this purpose still accounts for the largest share of the 3D technology market. The technology became accessible to the general public with the introduction of 3D printers for individuals, but that market is still marginal and is expected to remain so in the medium term, given the limited materials available to consumers: today 99% of items are printed with the same plastic, resin and metal materials. One of the main challenges for the 3D sector will be to combine several materials.125

The report noted that “the development of remote printing services, sometimes coupled with a platform for sharing 3D files online, means that anyone can print an object in 3D at a far higher quality than could be achieved with a low-end machine.”126 The report stressed the educational value of 3D printing: “The use of 3D printers in educational institutions and collaborative work spaces (‘fablabs’) also promotes universal access to the technology.”127 The report commented: “Most of today’s high-tech industries use this technology because it has a positive impact on innovation and the environment.”128 The report considered the diverse applications of 3D printing in industry: Expectations are already high in the medical sector, where this technology could have applications in the manufacture of prosthetics, dental implants, human skin and even organs (“bioprinting”): kidneys, for example. The same goes for the aerospace sector, where lighter components will help to reduce fuel consumption and preserve the environment and will enable savings to be made: Airbus currently has an aeroplane at the experimentation stage which has no fewer than a thousand 3D-printed components. The development of this technology is also of great interest to industries producing automotive spares, toys and household electrical appliances. Lastly, 3D printers and 3D scanners are increasingly being used in museums to restore historical objects and for research, particularly in archaeology.129

The report observed that 3D printing could bring about industrial transformation: By making on-demand production possible, 3D printing could offer many advantages to businesses: it could ease the strain on their logistical chains, reduce storage and transport operations, lessen environmental impact and cut spending on goods insurance and make it possible for them to reshore jobs, should they so wish.130

The report briefly considers questions of IP regarding 3D printing: In conclusion, legal experts consider that 3D printing has not had a dramatic impact on copyright. A 3D file would be considered a work and protected as such. However, it is fair to expect copyright problems to arise when 3D printing becomes widespread in industry. A future revision of Directive 2004/48/EC on the enforcement of intellectual property rights, which the Commission has announced for the current term, will be particularly important in this respect, all the more so if it is accompanied by soft-law action to provide information on the subject. However, it would be wise to distinguish between home printing for private use and printing for commercial use, and between B2B services and B2C services.131

125 126 127 128 129 130 131

Id. Id. Id. Id. Id. Id. Id.

464 Research handbook on intellectual property and technology transfer The report also broaches matters of civil liability in respect of 3D printing. The report stressed: “Innovation needs to be accompanied by law, without the law acting as a brake or a constraint.”132 The report “stresses the importance of creating a coherent legal framework to provide a smooth transition and legal certainty for consumers and businesses in order to promote innovation in the EU.”133 D.

Australia

In Australia, there has been much litigation over patent law and public sector research, particularly as a result of disagreements between key stakeholders.134 In the past, there have been conflicts in Australia between researchers, universities and industry over ownership of IP in respect of software.135 In the dispute over cancer research in University of Western Australia v. Gray [No.20], French J—who later became the Chief Justice of the High Court of Australia − held that university employees’ duty to research did not extend to a duty to invent: Absent express agreement to the contrary, rights in relation to inventions made by academic staff in the course of research and whether or not they are using university resources, will ordinarily belong to the academic staff as the inventors under the 1990 Act. The position is different if staff have a contractual duty to try to produce inventions. But a duty to research does not carry with it a duty to invent.136

His Honor suggested: UWA and other universities might well consider the alternative of deriving benefits from inventions produced by their staff by offering highly competent and experienced commercialization services in exchange for a negotiated interest in the relevant IP.137

In his opinion, “that alternative offers many benefits in terms of incentives, harmony and certainty that are not available through the enforcement of legal rights unlikely to be capable of precise definition.”138 This authoritative decision relied upon work by Professor Ann Monotti and Professor Sam Ricketson on public sector research.139 This decision was upheld on appeal by the Full Federal Court.140 The High Court of Australia refused as a special leave application for a further appeal.141 As such, Australian law recognizes that the duty to research does not involve a duty to invent.

132

Id. Id. 134 University of Western Australia v. Gray [No.20] (2008) FCA 498; University of Western Australia v. Gray [2009] FCAFC 116; University of Western Australia v. Gray [2010] HCATrans 11 (12 February 2010). 135 Victoria University of Technology v. Wilson (2004) 60 IPR 392. 136 University of Western Australia v. Gray [No.20] (2008) FCA 498. 137 Id. 138 Id. 139 Ann Monotti & Sam Ricketson, Universities and Intellectual Property: Ownership and Exploitation (2003). 140 University of Western Australia v. Gray [2009] FCAFC 116. 141 University of Western Australia v. Gray [2010] HCATrans 11 (Feb. 12, 2010). 133

Intellectual property, higher education, technology transfer, and 3D printing

465

The Vice Chancellor of the University of Sydney, Michael Spence, has questioned such assumptions.142 Taking a managerial view, he argued: The university claim is always seen as derivative of the primary claim of the individual academic. But we have seen that the claim of the individual academic is rather different to that of the ordinary author or inventor. That this is the case must surely be important in balancing her claim with that of both the university community of which she is a part, and the community that it serves more generally.143

No doubt, though, academic unions would be resistant to such a position on the grounds that it would diminish the rights of individual academic inventors within higher education. On occasion, there have been disputes over public research in information technology— such as CSIRO’s WiFi technologies.144 In the end, CSIRO and the Australian Government secured a significant windfall from the settlement of patent litigation with information technology companies.145 Malcolm Turnbull—Australia’s Prime Minister from 2015–2018—lamented that Australia lacked sufficient interaction between universities and industry.146 In conversation with the Vice-Chancellor of the University of Melbourne, Glyn Davis, Turnbull stressed: “There is no doubt that one of our major failures is that we are, as a nation, very poor at collaboration between primary research institutions, universities for the most part, and business.”147 Turnbull commented: “Everyone I talk to thinks that the problem is that academics have got - their incentives are very much associated with publish or perish.”148 He maintained that there was a need to foster greater interaction between universities, industry and government. As a result of such concerns, Turnbull instituted a new science and technology policy, during his term as the leader of the Conservative coalition of the Liberal Party and the National Party. He highlighted the role of 3D printing. For instance, Turnbull promoted the work of Stephen Brinks from 3D Brink at Western Sydney University.149 The Coalition Government has published a National Innovation and Science Agenda called the “Ideas Boom.”150 The

142

See Spence, supra note 5. Id. at 843–844. 144 CSIRO WLAN patent, Intell. prop. austl., available at https://www.ipaustralia.gov.au/tools -resources/case-studies/csiro-wlan-patent (last visited Mar. 29, 2019). 145 Misha Schubert, Australian Scientists Cash In On Wi-Fi Invention, sydney MornIng herald, Apr. 1, 2012. 146 Q&A—Melbourne Institute—The Australian Conference 2015 Economic and Social Outlook Conference, hon. MalColM turnbull Mp, Nov. 5, 2015, available at https://www.malcolmturnbull .com.au/media/qa-melbourne-institute-the-australian-conference-2015-economic-and-social-o (last visited Oct. 17, 2019). 147 Id. 148 Id. 149 Stephen Brinks, 3D print, and Western Sydney University, hon. MalColM turnbull, Oct. 16, 2015, available at https://www.facebook.com/malcolmturnbull/videos/10153763416501579/ (last visited Oct. 17, 2019). 150 Welcome to the Ideas Boom, nat’l InnovatIon & sCI. agenda, Dec. 7, 2015, available at http://www.innovation.gov.au/page/national-innovation-and-science-agenda-report (last visited Oct. 17, 2019). 143

466 Research handbook on intellectual property and technology transfer Australian Government has highlighted the example of Makers Empire—which received support and investor backing to achieve its vision of teaching kids to design for 3D printing.151 Coalition Government Minister Christopher Pyne has also stressed the importance of advanced manufacturing to Australia: Diffusing advanced manufacturing technologies across our traditional manufacturing sector could be game-changing in terms of our global competitiveness … Collaborating on innovation will be a critical enabler in the transformation of industry to become the producer of the goods and services of the future. We need to redouble our efforts in putting our incredible knowledge, capabilities and research to work and to better understanding the roadblocks to doing this, compared to other countries.152

There is a particular impetus to push for the transition of the economies of Victoria and Australia from old models of automotive car manufacturing to advanced manufacturing. Australia has successfully commercialized CSIRO’s research with 3D printing with the establishment of the publicly listed company, Titomic.153 The Australia Productivity Commission, though, was unconvinced that Australia’s innovation policy was coherent and well coordinated.154 The Productivity Commission agreed that there had been significant investments: “Australian governments have also invested in maker spaces, incubators and accelerators with the hope of attracting entrepreneurs and building critical mass to attract skills and investors.”155 Nonetheless, despite such initiatives, the Productivity Commission was concerned that Australia’s performance was still uneven: Australia is assessed as having good innovation infrastructure, public-sector organizations and human capital by international standards. Despite these strengths, Australia does not perform as well in terms of commercializing its ideas and innovations and in terms of diffusion as other countries.156

The Productivity Commission called for the Australian Government to reform Australia’s IP laws to improve its performance in terms of innovation policy.157 Journalists like Emma

151 The National Innovation and Science Agenda, Welcome to the Ideas Boom: Makers Empire, youtube (Apr. 17, 2016), available at https://www.youtube.com/watch?v=SWMI1RBvMHg. 152 Speech to Advanced Manufacturing Industry and Collaboration Forum, hon. MalColM turnbull, Sept. 30, 2015, available at https://www.pyneonline.com.au/media-centre/speeches/speech -to-advanced-manufacturing-industry-and-collaboration-forum (last visited Oct. 17, 2019). 153 Sarah Saunders, Titomic Announces New Board Member, Australian Patent for Kinetic Fusion 3D printing Technology, 3dprInt.CoM, Feb. 1, 2018, available at https://3dprint.com/202151/titomic -board-member-patent/ (last visited Oct. 17, 2019); Liam Mannix, World’s “Largest” 3D Metal Printer Launches in Melbourne, sydney MornIng herald, May 16, 2018. 154 Productivity Commission, An Overview of Innovation Policy, Shifting the Dial: 5 year Productivity Review, Supporting Paper No. 12, Canberra: produCtIvIty CoMM’n, 2017, available at http://www.pc .gov.au/inquiries/completed/productivity-review/report/productivity-review-supporting12.pdf. 155 Id. at 1. 156 Id. at 24. 157 Productivity Commission, Intellectual Property Arrangements, produCtIvIty CoMM’n, Report No. 78 (2016), available at http://www.pc.gov.au/inquiries/completed/intellectual-property/report (last visited Oct. 17, 2019); Karen Chester, What is Fair, produCtIvIty CoMM’n, Speech, Canberra, Feb. 24, 2017, available at http://www.pc.gov.au/news-media/speeches/fair (last visited Oct. 17, 2019).

Intellectual property, higher education, technology transfer, and 3D printing

467

Alberici have questioned whether Australia is fulfilling its aspirations of becoming an innovative economy.158

V.

PATENT INFRINGEMENT, THE DEFENSE OF EXPERIMENTAL USE AND 3D PRINTING

The hybrid technology of 3D printing has unsettled traditional conceptions of technology neutrality.159 There are some important comparative differences in terms of the treatment of patent infringement and exceptions jurisdiction-by-jurisdiction. 3D printing raises questions about patent liability and the scope and nature of patent exceptions and defenses. Questions about patent infringement and defenses are relevant to universities, educational institutions and public research entities—because they are not only homes to inventors, but they also are users of patents. The United States has had some early skirmishes over patent infringement and 3D printing. The limited range of patent exceptions and defenses may be of concern to the Maker Movement. Canada has sought to legislate for a new research exception in its new IP policy. The European Union has been considering the application of doctrines of patent infringement to 3D printing. There has been a discussion about the reformulation of patent defenses and exceptions in light of the emergence of 3D printing. Australia is also seeking to modernize its patent regime to better accommodate the emergence of new technologies—such as 3D printing. A.

United States

There have been a number of early scholarly articles on patent infringement and 3D printing in the United States. In the context of the United States, Timothy Holbrook has been worried that patent enforcement will be challenged by the emergence of 3D printing.160 He worries that the patent system will be confounded by additive manufacturing: It is unclear if courts or Congress will act to address these issues. What is inevitable, however, is that 3D printing will prove challenging to our patent system. There is a great irony here. One of the greatest innovations of our time may ultimately undermine a key engine of innovation, the patent system.161

Tabrez Ebrahim has considered the intersection of patent law, 3D printing and the digital environment.162

158 Emma Alberici, Are We Any Closer to Becoming An Innovative Economy?, abC news, May 7, 2018, available at http://www.abc.net.au/radionational/programs/breakfast/are-we-any-closer-to -becoming-an-innovative-economy/9734036 (last visited Oct. 17, 2019). 159 Nari Lee, “Revisiting the Principle of Technological Neutrality in Patent Protection in the Age of 3D Printing Technology and Cloud Computing” in TRIPS Plus 20, MPI Studies on Intellectual Property and Competition Law (Hanns Ullrich, et. al. eds., 2016). 160 Holbrook, supra note 13. 161 Id. 162 Tabrez Ebrahim, 3D Printing: Digital Infringement and Digital Regulation, 14 nw. J. teCh. & Intell. prop. 37 (2016).

468 Research handbook on intellectual property and technology transfer Thus far, there have been commercial skirmishes in respect of patent law and 3D printing. In the 2014 documentary Print the Legend, Luis Lopez and Clay Tweel highlight early IP disputes between 3D printing companies in the United States.163 The legal dispute in ClearCorrect Operating, LLC v. International Trade Commission over digital dental models raised important issues in respect of patent law, 3D printing and Internet jurisdiction.164 The majority of the United States Court of Appeals for the Federal Circuit (“CAFC”) was of the view that the International Trade Commission only had jurisdiction in respect of material things. 3D printing has also raised public policy questions about extraterritorial patent enforcement.165 As the commercial value of 3D printing in the field of manufacturing has grown in importance, there has been significant patent litigation in 2018 over the 3D printing of metal. In a jury trial, Desktop Metal Inc. was unable to establish its claims of patent infringement against its competitor Markforged Inc.166 Desktop Metal Inc.’s claims of trade secret violations against Markforged Inc. were pursued at trial. In the end, the parties agreed to settle their dispute on confidential terms.167 No doubt there will be future conflict of patents in respect of 3D printing—given the growing concentration of patent thickets in the field. As yet, there has not been any reported patent litigation over 3D printing involving universities, educational institutions and research organizations. Nonetheless, there have been concerns about the granting of poor quality patents in respect of 3D printing. The Electronic Frontier Foundation has highlighted the potential problem of patent thickets: While many core patents restricting 3D printing have expired or will soon expire, there is a risk that “creative” patent drafting will continue to lock up ideas beyond the 20-year terms of those initial patents or that patents will restrict further advances made by the open hardware community. The incremental nature of innovation in 3D printing makes it particularly unsuitable for patenting, as history has shown.168

The Electronic Frontier Foundation engaged in the crowd-sourcing of prior art in order to mount oppositions to the validity of a number of broad patents in respect of 3D printing.169 The civil society organization challenged patent applications related to Fabrication of Non-Homogeneous Articles Via Additive Manufacturing Using Three-Dimensional Voxel-Based Models; Build Materials and Applications Thereof; Method for Generating and Building Support Structures With Deposition-Based Digital Manufacturing Systems; Process for Producing Three-Dimensionally Shaped Object and Device for Producing Same; Additive

163

Print the Legend (Audax Films, 2014). ClearCorrect Operating, LLC v. ITC, 810 F.3d 1283 (Fed. Cir. 2015); and ClearCorrect Operating, LLC v. ITC, 819 F.3d 1334 (Fed. Cir. 2016). 165 Daniel Harris Brean, Patent Enforcement in Cyberterritories, 40 Cardozo l. rev. (forthcoming 2019). 166 Desktop Metal, Inc. v. Markforged, Inc., Case Number 1:18-CV-10524 (2018). 167 Desktop Metal Inc., Desktop Metal and Markforged Reach Amicable Resolution on Trade Secret Litigation, bus. wIre, Oct. 2, 2018. 168 Julie Samuels, Join EFF’s Efforts to Keep 3D Printing Open, eleCtronIC frontIer found, Oct. 24, 2012, available at https://www.eff.org/deeplinks/2012/10/join-effs-efforts-keep-3d-printing-open (last visited Oct. 17, 2019). 169 Julie Samuels, EFF and Partners Challenge Six 3D Printing Patent Applications, eleCtronIC frontIer found, Apr. 12, 2013, available at https://www.eff.org/deeplinks/2013/04/eff-partners -challenge (last visited Oct. 17, 2019). 164

Intellectual property, higher education, technology transfer, and 3D printing

469

Manufacturing System and Method for Printing Customized Chocolate Confections; and Ribbon Filament and Assembly for Use in Extrusion-based Digital Manufacturing Systems. Corynne McSherry—the legal director of the Electronic Frontier Foundation—has pushed for the recognition of the rights of makers at a policy level.170 However, patent law in the United States has been largely indifferent to the plight of makers, tinkerers and hackers. Historically, the United States provided strong recognition of a common law defense of experimental use. Justice Story, an associate justice of the Supreme Court of the United States, and the former Dane Professor of Harvard University, created both the defense of fair use under copyright law and the defense of experimental use in respect of patent law. In the 1813 appellate decision of Whittemore v. Cutter, Justice Story considered whether a party had infringed the patent assigned on a machine used to produce playing cards. He observed: “It could never have been the intention of the legislature to punish a man, who constructed such a machine merely for philosophical experiments, or for the purpose of ascertaining the sufficiency of the machine to produce its desired effects.”171 In the subsequent 1813 decision in Sawin v. Guild, Justice Story observed that “the making of patented machine to be an offence within the purview of it, must be the making with intent to use for profit, and not for the mere purpose of philosophical experiment, or to ascertain the verity and exactness of the specification.”172 The judge elaborated: “In other words, that the making must be with intent to infringe the patent right, and deprive the owner of the lawful rewards of his discovery.”173 In 1861, the decision in Peppenhausen v. Falke elaborated upon the nature of the common law defense of experimental use: “It has been held, and no doubt is now well settled, that an experiment with a patented article for the sole purpose of gratifying a philosophical taste, or curiosity, or for mere amusement, is not an infringement of the rights of the patentee.”174 While the defense of fair use has flourished under United States copyright law, the defense of experimental use has languished in United States patent law. There was significant controversy over the decision of the CAFC in Madey v. Duke University over patent law and experimental use.175 Judge Gajarsa upheld the appeal by Madey against Duke University.176 He observed on behalf of the court: Our precedent clearly does not immunize use that is in any way commercial in nature. Similarly, our precedent does not immunize any conduct that is in keeping with the alleged infringer’s legitimate business, regardless of commercial implications … In short, regardless of whether a particular institution or entity is engaged in an endeavor for commercial gain, so long as the act is in furtherance of the alleged infringer’s legitimate business and is not solely for amusement, to satisfy idle curiosity, or for strictly philosophical inquiry, the act does not qualify for the very narrow and strictly limited experimental use defense. Moreover, the profit or non-profit status of the user is not determinative.177

170

Corynne McSherry, EFF at Maker Faire Bay Area, eleCtronIC frontIer found., 2013, available at https://www.eff.org/event/eff-maker-faire-bay-area-2013 (last visited Oct. 17, 2019). 171 Whittemore v. Cutter, 29 F. Cas. 1120 (1813). 172 Sawin v. Guild, 21 F. Cas. 554 (1813). 173 Id. 174 Peppenhausen v. Falke, 19 F. Cas. 1048, 1049 (C.C.S.D.N.Y. 1861). 175 Madey v. Duke University, 307 F.3d 1351 (Fed. Cir. 2002). 176 Id. 177 Id. at 1362.

470 Research handbook on intellectual property and technology transfer The judge held that the district court attached too great a weight to the non-profit, educational status of Duke, “effectively suppressing the fact that Duke’s acts appeared to be in accordance with any reasonable interpretation of Duke’s legitimate business objectives.”178 He stressed that “Duke … like other major research institutions of higher learning is not shy in pursuing an aggressive patent licensing program from which it derives a not insubstantial revenue stream.”179 The judge directed that on remand the district court would have to revise and limit its conception of the experimental use defense: “The correct focus should not be on the non-profit status of Duke but on the legitimate business Duke is involved in and whether or not the use was solely for amusement, to satisfy idle curiosity, or for strictly philosophical inquiry.”180 The Supreme Court of the United States refused to hear an appeal of this particular case.181 There has been much disquiet and outcry that the United States has such a limited defense of experimental use, which is confined to actions “for amusement, to satisfy idle curiosity, or for strictly philosophical inquiry.”182 Nonetheless, the precedent in Madey v. Duke University has been applied in a number of technological and scientific contexts.183 There has been the occasional judicial warning, though, that studying or thinking about a patented invention is not an act of patent infringement.184 In a recent history of patent law and the research exemption, Nicholas Short contends: “The time is therefore ripe for Congress to revisit the research exemption.”185 In their provocative article, “Patents, Meet Napster”, Deven Desai and Gerard Magliocca have argued that there is a need for new patent exceptions to deal with 3D printing and the digitization of things.186 In a footnote, Desai and Magliocca lamented: “There is a solid argument that experimental use should be read more broadly to include at least some 3D printing in homes or businesses, but this is probably not a path that the Federal Circuit will explore.”187 Desai and Magliocca sought to overcome such deficiencies in current patent law:

178

Id. Id. at 1362–3. 180 Id. at 1363. 181 Duke University v. Madey, 539 U.S. 958 (2003). 182 fed. trade CoMMIssIon, To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy, Oct. 2003, available at http://www.ftc.gov/os/2003/10/innovationrpt.pdf (last visited Oct. 17, 2019); Rebecca Eisenberg, Patent Swords And Shields, 299 sCI. 1018 (2003); Janice Mueller, The Evanescent Experimental Use Exemption from U.S. Patent Infringement Liability: Implications for University/Nonprofit Research and Development, 56 baylor l. rev. 918 (2004); Donna Gitter, International Conflicts Over Patenting Human DNA Sequences In The United States And The European Union: An Argument For Compulsory Licensing And A Fair Use Exemption, 76 n.y.u. l. rev. 1623 (2001); Mark Janis, Experimental Use and the Shape of Patent Rights for Plant Innovation, Economics of Innovation and Science Policy, Department of Economics, Iowa st. u., 2003, available at https://www .card.iastate.edu/research/science-and-technology/papers/Janis-seminar-Fall-03.pdf (last visited Oct. 17, 2019); David Resnik, Patents And The Research Exemption, 299 sCI. 821 (2003). 183 Applera Corp. v. MJ Research Inc., 311 F. Supp. 2d 293 (D. Conn. 2004). 184 Classen Immunotherapies, Inc. v Biogen Idec, 659 F.3d 1057 (Fed. Cir. 2011). 185 Nicholas Short, The Political Economy of the Research Exemption in American Patent Law, fordhaM Intell. prop. MedIa & ent. l.J. 574 (2016). 186 Deven Desai & Gerard Magliocca, Patents, Meet Napster: 3D Printing and the Digitization of Things, 102 geo. l.J. 1691 (2014). 187 Id. at 1716. 179

Intellectual property, higher education, technology transfer, and 3D printing

471

Revising patent law to have a high minimum amount-in-controversy as a jurisdictional threshold would create a de facto fair use standard for home and experimental 3D printing activities. In addition, a patent DMCA would strike a balance between rights holders and intermediaries. As has happened in the copyright world, such a law has fostered new marketplaces and revenue models that allow for greater sharing, remixing, and selling of intellectual property. Without these changes, 3D printing could be mired in fights over protecting old business models. And mistaken regulation could fall into path-dependent solutions where creators are told to use a 3D printer only for certain purposes. These changes, therefore, balance interests and create the space 3D printing needs to become the foundation for the next wave of general-purpose computing and creation.188

There seems to be a desire here to create a broad defense of experimental use under United States patent law—together with a defense for home 3D printing activities. Attorney John Hornick has been critical of proposals to expand patent exceptions for 3D printing.189 Criticizing the proposal of Desai and Magliocca, Hornick argues: [T]he enactment of such a law could sound the death knell for patents, and for any company that makes products that can be made away from control. When consumers start making patented products instead of buying them, a personal exemption from patent infringement would excuse most infringing manufacturing. Although patent owners’ ability to enforce their patents would be subject to the Five Is, the potential to enforce them in appropriate situations would be better than having no right to do so because of a personal exemption from infringement. Moreover, such an exemption really isn’t necessary. If infringement away from control becomes common, it will be impractical or impossible to sue infringers.190

Hornick is thus resistant to an expansion of patent exceptions under United States patent law. His view is perhaps representative of the profession of IP attorneys. However, it is not necessarily a convincing argument. The United States is clearly out of step with other jurisdictions, which enjoy a much more generous defense of experimental use. Michael Weinberg—formerly of the civil society group Public Knowledge, and presently of the 3D printing company Shapeways—has observed that there are complications in terms of United States patent law in respect of repair and reproduction.191 He notes that “today the public is free to replicate unpatented elements of combination patents” and “they can repair and replace worn elements without securing an additional license or obtaining necessary replacement parts from the original manufacturer.”192 Weinberg fears: “When creating those replacement parts or unpatented elements becomes easier, manufacturers will likely begin to see it as piracy or theft.”193

188

Id. at 1720. John Hornick, 3D Printing Will Rock the World (2015). 190 Id. at 195. 191 Michael Weinberg, It Will Be Awesome If They Don’t Screw It Up: 3D Printing, Intellectual Property, and the Fight over the Next Great Disruptive Technology 13 (2010), available at https://www .publicknowledge.org/files/docs/3DPrintingPaperPublicKnowledge.pdf (last visited Oct. 17, 2019). 192 Id. 193 Id. 189

472 Research handbook on intellectual property and technology transfer B.

Canada

The TRIPS Agreement 1994 provides significant flexibility for member states in respect of exceptions to patent rights. Article 30 noted that: members may provide limited exceptions to the exclusive rights conferred by a patent, provided that such exceptions do not unreasonably conflict with a normal exploitation of the patent and do not unreasonably prejudice the legitimate interests of the patent owner, taking account of the legitimate interests of third parties.

The World Trade Organization Panel ruling in the Canada–Patent Protection case provides some guidance on the allowable extent of research exemptions under the TRIPS Agreement 1994.194 The Panel noted that “practically all Members of the WTO had such an exception albeit drafted in a great variety of ways.”195 The Panel commented that the exception was limited in character and narrowly defined: “It only applied to typically one out of five patent rights referred to in Article 28.1 of the TRIPS Agreement, since only use was permissible, while offering for sale, selling and importing were not permissible.”196 The Panel maintained that the patent holder’s legitimate interests do not include a monopoly on research: Given that the “basic patent deal” required the patentee to disclose his invention to the public and to accept that it served as the basis for further research, it could be reasonably argued that a “research monopoly” was not included in his legitimate interests and, therefore, the interests of third parties and their balancing with the patentee’s interests appeared to be redundant for the research exception.197

In Micro-Chemicals Ltd v. Smith Kline and French Inter-American Ltd,198 the Supreme Court of Canada held that a limited experiment to establish whether the experimenter could manufacture a quality product commercially in accordance with the specification of a patent was covered by the words “for experimental purposes relating to the subject-matter of the invention.”199 Professor David Vaver was quizzical about the attenuated development of the defense of experimental use in the United States.200 He doubted whether Canadian courts would follow the lead of the United States: “Canadian courts may, however, be less likely to read down statutory or common law exceptions in the way that US courts seem recently willing to do.”201 As part of its proposed IP Strategy, Prime Minister Justin Trudeau’s Canadian Government has promised to legislate for a patent research exemption.202 The IP Strategy noted: “Amendments aimed at affirming that there is no infringement when conducting experiments

194

Canada: Patent Protection of Pharmaceutical Products: Complaint by the European Communities and their Member States, Mar. 17, 2000, WT/DS114/R. 195 Id. at 55–6. 196 Id. at 56. 197 Id. 198 Micro-Chemicals Ltd v. Smith Kline and French Inter-American Ltd, 25 D.L.R. 79 (1971). 199 Id. 200 David Vaver, Canada’s Intellectual Property Framework: A Comparative Overview, 17 Intell. prop. J. 125 (2003–2004). 201 Id. 202 Intellectual Property Strategy, supra note 105.

Intellectual property, higher education, technology transfer, and 3D printing

473

that relate to the subject matter of a patent.”203 The IP Strategy affirmed: “However, resulting inventions still need to abide by existing patent laws before being sold or used for commercial benefit.”204 C.

European Union

In her ground-breaking book on 3D printing regulation, Angela Daly considers patent infringement and 3D printing.205 She commented that “intermediary liability for patent infringement has not yet developed to accommodate the changes that digitization has brought.”206 Daly suggests that “indirect patent infringement claims targeting intermediaries may also seem attractive to patent holders due to the difficulty and expense in tracking direct infringers.”207 She concludes: “With the absence of case law on these issues, both in the UK and other European national jurisdictions, it remains to be seen to what extent intermediaries will be liable for indirect patent infringement by disseminating 3D printing design files which contain a patented item.”208 Considering European law, Rosa Maria Ballardini, Marcus Norrgård and Timo Minssen explore the application of relevant laws and doctrines of patent infringement to 3D printing.209 The writers highlight legal uncertainty surrounding the interpretation of indirect patent infringement. Ballardini and her collaborators recommend: “Against this complex scenario we suggest that the sooner right holders begin strategizing on their own business and legal concerns, the less their business model will be disrupted by the advent of 3DP technology.”210 The European Union has taken a broader approach to the defense of experimental use than the United States, allowing for both non-commercial and commercial uses of patented inventions. The European Union has sought to encourage harmonization amongst its member states in respect of the research exemption under patent law. Article 27(b) of the Community Patent Convention (“CPC”) establishes the basis for an experimental use exception which exempts “acts done for experimental purposes relating to the subject matter of the patented invention.” There has been individual variation in terms of the implementation of the defense of experimental use in the European Union. There is also scope for the private and non-commercial use defense in the case of 3D printing technologies. Rosa Ballardini and Nari Lee have considered the issue at length.211 They commented:

203

Id. Id. 205 See Daly, supra note 11. 206 Id. 207 Id. 208 Id. 209 Rosa Maria Ballardini, Marcus Norrgård, & Timo Minssen, Enforcing patents in the era of 3D printing, 10 (11) J. Intell. prop. l. & praC. 850 (2015). 210 Id. at 865. 211 Rosa Maria Ballardini & Nari Lee, “The Private and Non-Commercial Use Defence Revisited: The Case of 3D Printing Technologies” in 3D Printing, Intellectual Property and Innovation: Insights from Law and Technology 169 (Ballardini et al. eds., 2016). 204

474 Research handbook on intellectual property and technology transfer The private use exception in European patent law has received little attention, partly because it is often deemed that the cost of manufacturing discourages a user from privately working a patented invention. With the advent of 3DP, however, this private working of an invention is expected to dramatically increase due to cost cutting developments in 3DP technology that enable home manufacturing. As such, home 3DP may disturb the balance that patent law aims to strike by contemplating the exception. As a corollary, more clarity may be required in applying the private use exception to this conduct.212

In their view: a flexible application of the private and non-commercial use exception for non-professional users and a cautious extension of liability to intermediaries (including suppliers of physical parts and machines) may be necessary to incentivize the diffusion and development of this important technology, while fairly protecting the interests of right holders.213

Particular sub-fields of 3D printing may pose pronounced issues both in terms of patent law and other forms of regulation. Bioprinting, in particular, could well raise special issues—given past precedents in respect of patent law and biotechnology.214 Likewise, medical devices and methods of human treatment could raise special issues.215 There could also be further issues arising in respect of essential patents and technical standards in additive manufacturing.216 D.

Australia

Tasmanian researchers Professor Dianne Nicol, Dr. Jane Nielsen and John Liddicoat have also been investigating 3D printing and patent infringement.217 In the jurisdiction of Australia, they have considered the question of patent infringement liability for the creation and distribution of CAD files.218 The authors contend that Australian law creates relatively clear liability in respect of patent infringement in respect of 3D printing. The researchers wonder whether there will be a need to revise Australia’s laws with respect to patent infringement: This technology is likely to become as central to Australian manufacturing as it will to manufacturing industries in other jurisdictions. Should authorization and supply infringement under Australian patent law become impractical mechanisms to enforce patent rights to the point where the economic

212

Id. at 187. Id. at 188. 214 Timo Minssen & Marc Mimler, “Patenting Bioprinting-Technologies in the US and Europe: The Fifth Element in the Third Dimension” in 3D Printing, Intellectual Property and Innovation: Insights from Law and Technology 117 (Ballardini et al. eds., 2016). 215 Dhanay Cadillo Chandler & Mika Salmi, “Law and Technology of 3D printing and Medical Devices” in 3D Printing, Intellectual Property and Innovation: Insights from Law and Technology 149 (Ballardini et al. eds., 2016). 216 Liguo Zhang, Inigo Flores Ituarte, & Rosa Maria Ballardini, “Essential Patents and Technical Standards in Additive Manufacturing” in 3D Printing, Intellectual Property and Innovation: Insights from Law and Technology 189 (Ballardini et al. eds., 2016). 217 Jane Nielsen & John Liddicoat, The Multiple Dimensions of Intellectual Property Infringement in the 3D Printing Era, 27 australIan Intell. prop. J. 184 (2017); John Liddicoat, Jane Nielsen, & Dianne Nicol, Three Dimensions of Patent Infringement: Liability for Creation and Distribution of CAD Files, 26 australIan Intell. prop. J. 165 (2016). 218 Liddicoat, Nielsen, & Nicol, supra note 217. 213

Intellectual property, higher education, technology transfer, and 3D printing

475

incentive provided by patents is eroded, then legislative action may be required to balance interests. Any contemplated change would also need to be consistent with Australia’s international and bilateral obligations, and should only be undertaken in response to evidence of necessity. Evidence of impact will invariably accumulate as the technology becomes mainstream.219

The researchers wonder whether Australia’s laws with respect to patent infringement are overbroad. Considering United Kingdom and Australian law, Dinusha Mendis, Jane Nielsen, Dianne Nicol and Phoebe Li examine the position of 3D printing.220 The authors contend: “As 3D printing continues to develop, it is very likely that patent and copyright laws will be strongly challenged but will continue to evolve and co-exist as they have done over the years in response to various technologies.”221 The Australian Labor Party’s Member for Gellibrand, the Hon. Tim Watts MP, has been concerned about the impact of patent infringement claims upon the Maker Movement in Australia: Patent laws are already becoming a similar hand-brake on digital innovation. As “Maker” communities and 3D printing grow in popularity, so too will disputes over patent infringement. In response, progressives should champion a new micro-economic reform agenda to re-evaluate intellectual property law from first principles—focusing on incentives and public benefits, not the mindless protection of statutory monopolies. This process should be led by economists and innovators, not lawyers and rent seekers. Without it, intellectual property will increasingly become an instrument of the protection of vested interests rather than the promotion of innovation. As progressives, we must stand up for the new online communities created by the digital revolution.222

Tim Watts insisted: “We are the Promethean party—the bearers of the fires of political change.”223 He recognized: “This task is a difficult one but I am confident in Labor’s future knowing that I share this mission with you all.”224 His colleague—the Hon. Jim Chalmers MP— has also highlighted the impact of 3D printing and additive manufacturing in Australia.225 3D printing also raises larger questions about the role and scope of patent exceptions and defenses in Australia. Australia has a long history of tinkering—dating back even before Federation. Early colonial patents in Victoria, for instance, feature the inventions of the Furphy’s—Shepparton engineers and blacksmiths.226 In her cultural history of tinkering, Katherine Wilson has written:

219

Id. Dinusha Mendis, Jane Nielsen, Dianne Nicol, & Phoebe Li, “The Co-existence of Copyright and Patent Laws to Protect Innovation: case study of 3D printing in UK and Australian Law” in The Oxford Handbook of Law, Regulation and Technology 451 (Brownsword et al., eds., 2017). 221 Id. at 469. 222 The Hon. Tim Watts, First Speech, House of Representatives, Australian Parliament (Feb. 2, 2013), available at http://www.aph.gov.au/Senators_and_Members/Members/FirstSpeeches/193430. 223 Id. 224 Matthew Rimmer, The Alchemy of Junk: Patent Law and Non-Coding DNA, 3(2) u. ottawa l. & teCh. J. 539 (2006). 225 The Hon. Jim Chalmers, 3D Printing: Not Yet a New Industrial Revolution, but Its Impact Will Be Huge, guardIan, Dec. 11, 2013. 226 History, furphy, available at http://www.furphys.com.au/component/content/article?id=37& itemid=60 (last visited Mar. 29, 2019). 220

476 Research handbook on intellectual property and technology transfer For a sector of Australians, tinkering is an impulsive habit of material problem solving with its own lexicon of values. It’s an iteration of freedom pride, dignity, ethics and artisanal joy—an unfettered way to live according to one’s own measures. It can be practical and utilitarian, a form of economic production, but also a form of scholarship, play, adventure, resourcefulness, and resilience. It can be a portal to social connection, community, spirituality, sanctuary, thrift, identity and political resistance.227

Despite this long history of tinkering and experimentation, there was legal uncertainty as to the status of experimental use under Australian patent law. In the wake of patent claims by biotechnology firm, Genetic Technologies Limited, there was anxiety amongst public policy makers as to whether researchers engaged in experimental use would infringe patents.228 While there was no statutory defense of experimental use, there was debate over whether or not Australia had inherited British common law regarding experimental use. As part of its inquiry into patent law and biotechnology, the Australian Law Reform Commission called for the adoption of a defense of experimental use.229 The Advisory Council on IP called for the adoption of a statutory defense of experimental use.230 Such public policy inquiries laid the ground for substantive patent law reform.231 Australia introduced a statutory defense of experimental use under patent law with the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 (Cth). Senator Kim Carr— the Minister for Innovation, Industry, Science and Research—explained the importance of research freedom in his second reading speech: We need to set our researchers free and ensure that the patent system encourages further innovation. The key reform in this category introduces an explicit provision permitting experimentation to be conducted without infringing patent rights. This amendment will give comfort to researchers in Australia, and is strongly supported by Australia’s research sector. The amendment strikes a balance between the rights of patentees and the rights of subsequent researchers. The exemption protects researchers and follow-on innovators as long as what they are doing is predominantly for research and experimental purposes.232

227

Katherine Wilson, Tinkering: Australians Reinvent DIY Culture 5 (2017). Rimmer, supra note 224. 229 Australian Law Reform Commission, Gene Patenting and Human Health: Issues Paper 27, sydney: austl CoMMonwealth, July 2003, available at https://www.alrc.gov.au/publication/gene -patenting-and-human-health-ip-27/ (last visited Oct. 17, 2019); Australian Law Reform Commission, Gene Patenting and Human Health: Discussion Paper 68, sydney: austl CoMMonwealth, Feb. 2004, available at https://www.alrc.gov.au/publication/gene-patenting-and-human-health-dp-68/ (last visited Oct. 17, 2019); Australian Law Reform Commission, Genes and Ingenuity: Gene Patenting and Human Health: Report 99, sydney: austl CoMMonwealth, June 2004, available at https://www.alrc.gov.au/ publication/genes-and-ingenuity-gene-patenting-and-human-health-alrc-report-99/ (last visited Oct. 17, 2019). 230 Advisory Council on Intellectual Property, Patents and Experimental Use: Issues Paper, Canberra: CoMMonwealth gov’t, Feb. 2004, available at http://www.acip.gov.au/library/ patentsexpuse.PDF; Advisory Council on Intellectual Property, Patents and Experimental Use: Final Report, Canberra: CoMMonwealth governMent, 2005, available at http://www.acip.gov.au/library/ ACIP%20Patents%20&%20Experimental%20Use%20final%20report%20FINAL.pdf. The Advisory Council on Intellectual Property has since been disbanded. Id. 231 Rimmer, supra note 224. 232 Senator Kim Carr, Second Reading Speech on the Intellectual Property Laws (Raising the Bar) Bill 2011 (Cth) Hansard, australIan senate, australIan parlIaMent, at 3485 (June 22, 2011). 228

Intellectual property, higher education, technology transfer, and 3D printing

477

In addition to the defense of experimental use, there was also an amendment to exempt research activities necessary for gaining pre-market or pre-manufacturing regulatory approval from infringement. Senator Kim Carr emphasized: “The bill will ensure that Australia maintains a world class IP system: an IP system that fosters Australian innovation in the modern global economy.”233 It is worthwhile considering the text of the defense of experimental use under Australian patent law. Section 119C of the Patents Act 1990 (Cth) deals with “Infringement exemptions: acts for experimental purposes.” The statutory provision provides: 1. A person may, without infringing a patent for an invention, do an act that would infringe the patent apart from this subsection, if the act is done for experimental purposes relating to the subject matter of the invention. 2. For the purposes of this section, experimental purposes relating to the subject matter of the invention include, but are not limited to, the following: (a) determining the properties of the invention; (b) determining the scope of a claim relating to the invention; (c) improving or modifying the invention; (d) determining the validity of the patent or of a claim relating to the invention; (e) determining whether the patent for the invention would be, or has been, infringed by the doing of an act.

This general defense would be of particular relevance to inventors, makers and designers involved in 3D printing and additive manufacturing in Australia. In the absence of any test cases, the exact contours of the defense of experimental use remain uncertain. Safe to say, the defense is broader than that in the United States, but narrower than the expansive approach in the European Union. The Australian Productivity Commission has considered whether Australia’s IP laws have been sufficiently adaptable and flexible to deal with new technologies.234 Considering the topic of 3D printing, the Productivity Commission was not yet convinced that 3D printing was posing systematic problems.235 The Productivity Commission recommended monitoring technological developments in the area and their implications for the IP system. In terms of the operation of the patent system, the Productivity Commission reflected that there was a need to reconcile the competing objectives of the patent system in a more finely calibrated fashion: Patents can advance human knowledge by encouraging socially valuable innovation that would not have otherwise occurred. However, if poorly calibrated, they also impose net costs on the community. By design, patent protection inhibits competitors from freely using an inventor’s technology, but over-protection can stifle competition more broadly, leading to reduced innovation and excessive prices. Moreover, by blocking subsequent innovators, patent protection can perversely inhibit the advancement of knowledge through “follow-on” innovation.236

The Productivity Commission has made some further recommendations as to how to improve Australia’s patent regime.237 The Australian Government has responded with a subset of

233 234 235 236 237

Id. Productivity Comm’n, supra note 157; Chester, supra note 157. Productivity Comm’n, supra note 158, at 345. Id. at 13. Productivity Comm’n, supra note 157; Chester, supra note 157.

478 Research handbook on intellectual property and technology transfer such recommendations, passing the Intellectual Property Laws Amendment (Productivity Commission Response Part 1 and Other Measures) Act 2018 (Cth). Further law reform is planned in the mid-term and the long-term in respect of IP. There still remains the issue of the treatment of repairs under patent law. As Mitchell Adams from Swinburne University has observed, the designs regime recognizes the right to repair (albeit in a narrow fashion)—but the patent regime does not have a specific exception for the right to repair.238

VI.

CONCLUSION

This Chapter has considered the relationship between IP, technology transfer, and higher education in the context of 3D printing, additive manufacturing and the Maker Movement. It has highlighted that the advent of makerspaces, Fab Labs, TechShops, innovation hubs and networks has represented an effort to improve the innovative culture and performance of universities, educational institutions and research organizations. There are certainly opportunities to refine patent policy and laws in respect of patent ownership, patent infringement and patent exceptions in light of the 3D printing revolution. IP law faces particular challenges in respect of education and public sector research in terms of patent ownership, patent infringement and patent exceptions. As Professor Shubha Ghosh has noted: While IP laws may be tailored to choices of business organizations, the case of universities shows how general rules can evolve to accommodate complex institutions with multiple goals that play a critical role in federal and state policies for invention, innovation and development.239

He notes, though: “An open question is whether existing special rules are adequate for protecting the interests of universities without balkanizing the federal policies regulating and promoting innovation.”240 The case of 3D printing has certainly highlighted the competing, multiple goals at work in respect of IP and innovation policy in higher education. Universities have created makerspaces, Fab Labs, TechShops and innovation hubs and networks for various reasons—including promoting patent disclosures and commercialization, the development of new educational pedagogies, and open innovation and sharing. In his book on The Branding of the American Mind, Jacob Rooksby has stressed the importance of the larger public interest in higher education: It is incumbent on all who are aware of higher education’s struggle with intellectual property to call on decision makers at colleges and universities, urging them to implement a public-serving path forward, erecting sensible and visible boundaries in the face of the alluring temptations that intellectual property protections often offer. Only by exposing the role of choice in this arena, and redefining,

238

See Adams, supra note 17; GM Global Technology Operations LLC v S.S.S. Auto Parts Pty Ltd (2019) FCA 97 (Federal Court of Australia) (explaining the opinion by Burley J on the right to repair under Australian designs law). 239 See Ghosh, supra note 8, at 694. 240 Id. at 394.

Intellectual property, higher education, technology transfer, and 3D printing

479

through legal and policy changes, institutions’ freedom to act, can we expect colleges and universities to make better decisions regarding intellectual property—decisions that benefit us all.241

In this context, there are important lessons to be learnt from 3D printing. The Maker Movement can help foster a culture of open access, open data and open innovation in universities, educational institutions and public resource organizations.

241

See Rooksby, supra note 7, at 289.

Index

AAAS Report (1934) 6–7 academic capitalism 49, 380 academic-industry partnerships 39 academic institutions, social values of 118 academic journal policies 162–4 academic patent market and patent policy 273–5 transfer of liability 272–3 academic research, corporate funding of 143 academic technologies commercialization of 227, 230, 232, 234 evolution of 167 transfer of 11, 30, 84, 131, 144, 168, 174, 185, 189, 198 accountability, principles of 401 action-oriented practical intelligence 221 Action Plan on Open Government (Canada) 395 additive manufacturing 449, 454, 457, 459, 462, 467, 474–5, 477–8 Advanced Technology Program (ATP) 136 affiliated non-profit organizations (ANPOs) 122–5 composition of trustees 123 conflicts of interest 123–4 creative use of 128 formation and utilization of 122 governance mechanisms of 124 rise and spread of 122 strategic use and coordination of 127–8 trustee-enabled exchanges 123 Alfred Mann Foundation 18 Allied Security Trust (AST) 258, 301 altruism 237 America Competes Act (2007) 139 America Invents Act (AIA, 2011) 49–50, 52, 138, 339 background of 340–43 patent law under 341 America Makes program 459 American Council on Education (ACE) 42 American economic competitiveness 377 Anderson, Chris 450 Annual Knowledge Transfer Survey 419, 423 anonymous data, definition of 407 anti-vaccine movement 129 Arizona State University (ASU) 18, 112, 125 School of Life Sciences 112 arms race, of innovation programs 202–6 impact of 209–13

Association of American Medical Colleges (AAMC) 42, 64 Association of American Universities (AAU) 42–4, 47–50, 55–6, 60, 65, 123, 124, 137, 202, 208, 273, 275 Association for Molecular Pathology v. Myriad Genetics 38 Association of Public and Land-grant Universities (APLU) 42, 55–6, 64–5, 137, 140, 167, 202, 208 Working Group 167 Association for Research and Innovation Advancement (ARIA) 140–41, 202 Association of Science and Technology Professionals (ASTP) 418–19, 423 Association of University Technology Managers, Inc. (AUTM) 42, 71, 84–6, 131, 132, 180, 202, 259, 265, 309, 415 advocacy evolution over time 134–41 Better World Project 140, 209 brand identity 141 early emergence (1974–2000) 134–6 contemporary policy advocacy 135–6 early advocacy by SUPA 134–5 evolution of 132 evolving identity of 141 individual-based membership of 137 institutionalization and identity refinement (2015–2018) 140–41 interest in ATP and SBIR programs 136 IP policy-related events 138 Licensing Activity Survey 65, 139 organizational development (2000–2012) 137–40 policy-related activities 136 pre-competitive technologies 136 role and influence of 134 strategies for professional development 138 ASTP-Proton Annual Knowledge Transfer Survey 419, 423 auctions 244, 250, 303–5 Bacon, Francis 375 bankruptcy 451, 461 auctioned patents in 304 Baselga incident 143, 165

480

Index Bayh-Dole Act (1980) 1, 6, 9–12, 17, 19, 27, 33, 35, 46, 57, 92, 131–2, 138–9, 166, 169, 193, 198–9, 208, 214, 236, 270, 292, 309, 319, 323, 333, 377–8, 421, 423, 425, 457 applications of 61, 69 enactment of 49, 69, 283 on European technology transfer 414–16 for federally-funded inventions 30 goals of 63, 252 for intellectual property management 69 legal treatment of patents under 69 legislative history of 60 march-in rights 61 passage of 49, 239, 242, 275 purpose of 252 rules pertaining to university patents 84 and university patents 240–41 university private march-in rights under 39 Big Data 285, 369, 410 Big Data to Knowledge (BD2K) initiative 386 biomaterials licenses 173–4 biomedical engineering innovations 176 biotechnology industry 111, 138, 225, 227–8, 457 Biotechnology Innovation Organization (BIO) 138 Bonvillian, William 459 Boston Consulting Group 120 Boston University litigation 316–17 Bozeman, Barry 109–10, 128, 225 brand identity 141 Branding of the American Mind, The (2016) 80, 454 Bremer, Howard 47, 131, 134–5, 139 British Technology Group (BTG) 415 Burk, Dan 216, 233 Business Process Reengineering and Benchmarking 126 Cahoy, Daniel R. 283–308 California Institute of Technology (CalTech) 5, 23, 244, 313, 315, 317–18 California Research Alliance 205 capital-intensive research programs 379 Carnegie Classification of Institutions of Higher Education 67 Carnegie Corporation 44 Carolina Express License Agreement 191 Carter-Johnson, Jennifer 4–40 Centers for Medicare and Medicaid Services (CMS) Open Payments database 160 Charter of Fundamental Rights of the EU 401 Chinese universities, technology transfer in existing problems with 444–5 gap between research and market 445 government-university cooperation in 442 history and current status of 437–41 current development status 439–41

481

development history 438–9 horizontal technology services 438 industry-university-research cooperation 439 initiation period of university-run enterprises 438–9 intellectual property 439 major regulatory and policy documents pertaining to 436 outcomes of 440 policies and normative documents related to 434–7 professional teams working on 445 “reform and opening up” policy 438, 444 scientific research and 440, 444 sources of research funds 445 success and experience of 441–3 highly-diversified cooperation and communications 442 policy support 441 science and technology parks 442–3 university-run enterprises 442–3 university professors and 441 Chu, Zhang 343–446 civil society organization 468 Code of Federal Regulations 35 “combination product” language 182 commercialization of technology 1, 13, 132, 275 commercialization theory, concept of 334–5 Communalism, idea of 30 concept centers and incubators 215, 229–30, 234 conflicts of interest (COI) 207, 431 academic journal policies 162–4 administration of 148 advisory committee members 151–2 affiliated non-profit organizations (ANPOs) 123–4 “best practice” recommendations for 152 Common Rule 147 concerns, critiques and further research on 164–5 defined 145 disclosure of 162 failure to disclose 163 financial 145–6, 148, 157, 161 governmental policies on 147–52 governmental reviewers 151–2 impact of 144 at individual level 145, 147–9 at institutional level 145, 149–51, 161 institutional oversight and guidance 153–5 management of 157–9 meaning of 145–7 monitoring and non-compliance 160–61 non-research 159–60 OHRP Draft Interim Guidance on 149

482 Research handbook on intellectual property and technology transfer policies, procedures and challenges 152–61 policies relating to 147 public disclosure of 148 reporting metrics and requirements 164 significant financial interests 155–6 evaluation of 157 under-reporting of 165 contemporary knowledge governance, in the university 376–80, 381 Contreras, Jorge 143–165, 461 copyright 156, 173, 337 copyright laws 23 versus patent law 379 Digital Millennium Copyright Act (DCMA) 81 IP policies for 71 management of 72 Michigan’s policy on 75–6 ownership of 72, 448 policy regarding 74 protection of 24, 396 publications of 72 Stanford’s policy on 73 to works of authorship 24 works of non-employees 74 cost of university IP 413, 429–31, 433 Costa, Michael 256–282 Cottrell, Frederick 7 Coulter Translational Partnership and Research Awards 176 Council of 24 October 1995 401 Council on Federal Relations (CFR) 47 Council on Governmental Relations (COGR) 42, 51 Court of Appeals for the Federal Circuit (CAFC), US 310, 320–21, 353, 362, 468 Creative Commons Public License (CCPL) 397 Cunningham, James A. 412–433 Dahl, Cynthia Laury 339–363 data and changing university 373–82 abbreviated history 374–5 changes and conflicts about the mission of the university 380–82 contemporary knowledge governance 376–80 knowledge in the modern university 376 knowledge through the centuries 375–6 conflicts with IP premises 371–3 defined 367 and the law 371–3 old and new 367–71 property rights in 372 data carpentry 369 data collection 121, 139, 158, 256, 366, 368, 371, 373, 400, 402, 407

data curation 369 data custodian 388 data-driven society 391 data governance of emerging university 382–9 as IP institution 383–5 normativity of 385–7 as sum of its stakeholders 387–9 policies for 364, 371, 378 university-based 384 data-intensive research 365–6, 385, 389 meanings and applications of 367–73 data-intensive university 367, 382, 389–90 normativity of 385–7 data lifecycle 369–70, 385, 409 data liquidity 369 data management 388, 409 Data Management Plan (DMP) 387–8, 406 life cycle 406 data minimisation, principle of 404 data mining 396 data portraiture 369 data privacy 385 by default 404 by design 401 guidelines for regulation of 400 security and accountability principles 404 data processing 368, 401 data protection principles of 402, 408 risk-benefit analysis 409 specific purpose versus any purpose 408–10 data security 366, 385 data sharing 365–6, 383, 385–8, 409 data steward 388 data visualization, practices of 369 datasets 378 concerning university 311–18 description of 312–15 Davis, Glyn 465 design, concept of 95, 96–9, 126 application of 102 decision-making 99–102 formulation of 127 in higher education 99 imaginative possibilities for 127–8 research universities and 99–102 in STEM arena 102, 129 digital communication 451 digital humanities 370 Digital Millennium Copyright Act (DCMA) 81 discretionary grants 251–2 discretionary patents 250–51

Index Donath, Judith 369 drug pricing, issue of 58–61 Ebrahim, Tabrez 467 economic recession 120 Eco-Patent Commons 302 Edison Trust 287 Electronic Frontier Foundation (EFF) 320–21, 448, 468–9 Emerging Pathogens Institute (EPI) 107 end user licenses (EULAs) 173 entrepreneurial universities 101, 231 entrepreneurship, university-based 200 Environmental Protection Agency 299 epidermal growth factor receptor (EGFR) 162 epistemic community 97 equity, in startup licensee dead equity 188 framework for software startups 192 participation rights 188–9 pro rata rights 188–9 purchase rights 188–9 receiving of 186–7 amount of 187–8 rights of first refusal 188–9 type of 184–6 Ethics in Government Act (1978) 151 ethos of science 6 EU Horizon 2020 research project 405–8 European Commission 405, 407, 416, 431, 462 European Convention on Human Rights 398 European Court of Justice 417 European Database Directive 396 European General Data Protection Regulation 393 European Parliament 397, 462 Directive 95/46/EC of 401 report on 3D printing 462 European Patent Office 175, 421, 422, 427 European Research Area (ERA) 405 European Space Agency 462 European technology transfer 412 academic privilege for 416 Association of Science and Technology Professionals (ASTP) 418 Bayh-Dole law on 414–16 commercialization of 414 data relating to 419 distribution of licenses 418 empirical evidence based on 418–24 growth of 432 institutional contexts in 415–17 institutional contexts in Europe and USA 415–17 IP rights of universities and public research centers 413–15

483

material transfer agreement 424 ownership of IPRs 417 patent activity in Europe 422–4 Technology Transfer Offices (TTOs) 418–19 European Union (EU) Continent-wide Directive on protection of databases 373 Member States 412, 433 projects involving 3D printing 462–4 technology transfer offices within 412 technology transfer within 412 Unified Patent Court (UPC) 417 European Union Charter of Fundamental Rights 398 Ewing, Tom 322 Fab Lab movement 450–51 fair information practice principles (FIPPs) 399 Fair Information Principles (FIPs) 393, 399 federally-funded research 92, 135, 196, 198, 213, 250, 255 federally-funded technologies, commercialization of 207, 211 Feldman, Robin 110, 324–338, 332, 336 financial incentive, from patent royalties 63, 129, 227 findable, accessible, interoperable, and reusable (FAIR) data 386 Firpo, Teo 309–323 first-inventor-to-file system 49–50 Food and Drug Administration (FDA) 146 Framework Program for Research and Innovation 405 freedom of information 82–3 freedom to operate (FTO) license 257, 283, 345, 347, 350–51 Free to Make (2016) 451 Frye, Brian L. 236–256 Furthering America’s Research Enterprise report (2014) 66 General Data Protection Regulation (GDPR) 401–2, 404–5, 407–8 Georgia Tech Research Institute (GTRI) 122, 125 German Employees’ Inventions Act (2002) 420 Gershenfeld, Neil 450 Ghosh, Shubha 69–91, 448, 478 global financial crisis (2008) 201 global health initiatives 115 Global IP Index 362 global knowledge economy 92 Google 171 Google Patents 311 government funded invention 309 government funded university research 238–9 commercialization of 252

484 Research handbook on intellectual property and technology transfer government patents 238–9 grassroots mobilization 133 GreenXChange 302 Guarda, Paolo 391–410 Hager, Mark 133 Haskins, Charles Homer 44 higher education 1, 42, 51, 53–5, 81, 102–3, 449–54 commercialization of IP in 448 design, concept of 99 patent ownership and licensing in 449 Hayter, Christopher S. 131–142 Hi-Tech Pharmaceuticals 315 high-tech industry 53 Hornick, John 456, 471 Hughes, Sally Smith 457 human capital resources 429 human interaction, in technology transfer 218, 307 Idea of a University, The (1852) 375 individual technologies, commercialization of 209 industrial revolution 286, 450, 453 industry–academic relationships, importance of 119 industry partnerships 39–40, 79, 118, 166–7, 202, 205 information sharing 392 information society 391, 397 information technology (IT) industries 154, 327, 336, 341, 369–70, 389, 449, 465 InfoSoc Directive 397 initial public offering 171, 336 innovation hubs 103, 197, 450, 478 innovation, markers of 329–30 institutional funding, levels of 106, 116 Institutional Patent Agreements (IPAs) 45, 131, 135 administrative challenges associated with 45 blocking of 46 institutional pluralism 379 Institutional Review Board (IRB) 154 institutionalization of university patents 414, 429–30 intellectual capital 420, 453 intellectual property (IP) 1, 94, 131, 148, 257, 340, 475 auction model in 303–5 Canadian IP laws 461 of Chinese universities 439 commercialization of 448 commoditization of 285 cost of 429–31 coverage of creations subject to IP policies 22–4 ownership by universities 24–5 people and obligations 21–2

dispute resolution procedures 28–9 education program 33 in higher education 448 impact on society and academic productivity 426–9 litigations 80–84 management of Bayh-Dole Act see Bayh-Dole Act (1980) emerging norms for 84 key provisions in 74–5 Office of Sponsored Programs (OSP) for 77 non-traditional transfer mechanisms human element 307 reliable valuation 306–7 ownership of 15, 25, 75, 118 policies in selected schools 71–80 policies at universities 19–29 policy goals 19–21 portfolio of 85 revenue sharing across the institution 26–7 amongst co-creators 27 rights of 4 structure of 19–29 technology transfer through sale and licensing of 283 valuation of see intellectual property valuation intellectual property (IP) revenues to commercialize research innovation 170 for funding TTO’s patent budgets 169–70 stakeholders directly benefiting from 168–9 to support additional research 169 timing or length of payments 183 intellectual property rights (IPRs) 391 application of 396 management of 408 open data and 395–8 ownership of 417 of universities 427 intellectual property valuation 166 extra-license mechanisms for 183 historical cost approach for 175–7 methodologies used by TTOs 174–83 for bridging a valuation gap 182–3 common methods 181–2 cost approach 175–8 example of valuation scenario 174–5 income approach 179 market approach 179–80 modern TTO and the role of 166–70 portfolio approach to 189–93 examples of TTOs adopting 190–92

Index motivations for adopting 189–90 situations still requiring 193 replacement cost approach for 178 royalty stacking 182 significance of 168–9 special cases of equity 184–9 royalty buyouts 183–4 stakeholders directly benefiting from 168–9 standardized startup terms for 191 for supporting additional research 169 Intellectual Ventures (IV) 244, 258, 300, 311, 322 inter partes review (IPR) 310, 339–40 effect on TTO policies and procedures 343–61 causing changes to budgeting 359–61 effects of 361 filing, licensing, enforcement, and budgeting 346–9 “freedom to operate” category 350 importance of 340–43 and patent enforcement 356–9 patent filing strategies and 346–9 risk to TTOs 343–6 sovereign immunity defense to 352–5 interdisciplinary schools 111–14, 121, 127–8 intermediary-assisted exchange 294 International Committee of Medical Journal Editors (ICMJE) 162 International Covenant on Civil and Political Rights 398 invention-related knowledge 215 inventions definition of 72 licensing of 169 patentable inventions 73 Stanford’s policy on 73 inventorship and allocating revenue 33–7 IP Exchange International (IPXI) 305 Italian Financial Law 416 job creation 54 joint inventorship 34–5 joint ventures 328–30, 416 Judson, Harry Pratt 44 Kappos, David 320 Katholik Universiteit van Leuven (KUL), Belgium 428 Kauffman Foundation 19, 203 Kerr, Clark 68, 380 know-how, transfer of 220, 224, 327 knowledge-based economy 209 knowledge-based enterprise academic departments 114–22

485

academic structure 102–3 affiliated non-profit organizations 122–5 bricks-and-mortar entity 107 centers and institutes 103–10 design, concept of 96–9 research universities and 99–102 dynamics of disciplines 115 emerging economic and sociopolitical agendas 108 interdisciplinary schools 111–14 loose coupling 100–101 modes of production 102–3 school-based design 113 school-level curriculum 112 stand-alone entity 107 STEM-oriented units (SOUs) 92 theoretical perspectives of 95–6 university as 92–130 knowledge-based framing, of the university 382 knowledge-based organizations 412 knowledge creation, functions of 201 knowledge governance 367, 371, 374, 381–3, 385 in the university 376–80 knowledge production 98–9, 115, 368, 374 and distribution functions of university 379 in Europe 375 in modern university 376 modes of 102–3 through the centuries 375–6 knowledge sharing, practices of 67, 374, 388 knowledge transfer 221–4, 226, 228, 230, 232–3, 409, 413, 419, 423 in European content 421 “land grant” universities 376 Leahy-Smith America Invents Act (2011) 49, 315, 321 Lee, Peter 214–235, 307 legal liability protection 296 Lemelson Foundation 295 Lemley, Mark A. 314, 319, 324–338, 331, 333–4, 448 Leslie, Stuart 94, 105 licensing bilateral model of 284 economic consideration in 171 freedom to operate (FTO) license 351 funding of patent budgets through 170 in higher education 449 for income 37–9 in-house 329 of inventions 169 patent licenses 172, 331 Pharmalicensing 294 process of 15

486 Research handbook on intellectual property and technology transfer revenues from 169 technology portals for online licensing 167 of technology transfer 171 terms for industry sponsored research 191–2 value of 193–4 Licensing Executives Society (LES) 180 licensing fee 207, 295 licensing requests, denial of 82, 305, 330, 356 Loise, Vicki 137–8, 140 loose coupling, concept of 100–102, 122, 126 LOT Network 301 Love, Brian J. 256–283, 296 Machlup, Fritz 252 Madison, Michael J. 364–390 Maker Movement 447, 449–54 Maker Movement Manifesto, The (2014) 451 Maker Revolution, The (2017) 451 MakerSpace movement 454 Management by Objectives (MBO) 126 Manhattan Project, of World War II 8, 105, 116, 238 march-in rights, use of 11, 39, 58–61, 239, 252–3, 255, 448 marijuana, legalization of 81–2 market failure, risk of 237, 416, 425 market university 49 Massachusetts Institute of Technology (MIT) 124 Invention Administration Agreement (1937) 8 patent policy 8 rights to an invention 8 material transfer agreements (MTAs) 17, 77, 132, 167, 173, 174, 412, 424 Merton, Robert 6–7, 30 Mireles, Michael S. 309–323 MN-IP: Minnesota Innovation Partnerships 192 moral hazard, problems of 90 Motion Picture Patents Company 287 MPEG LA consortium 286 multi-national corporations 168 Nader, Ralph 144 National Institute of Standards and Technology (NIST) 57 Return on Investment Initiative Request for Information 57, 62, 65 National Research Council (NRC) on Bayh-Dole Act 64 Nine Points document 64–5 National Venture Capital Association (NVCA) 190 Nicotra, Melita 412–433 non-practicing entity (NPE) 257, 266, 269, 325 Bayh-Dole Act (1980) 333 commercialization-plus theories 333 contributing to social welfare 328 impact on commercialization goals 328

independent invention and prior user rights 337–8 justifications for innovation-related 326–8 traditional 328 patent suits 332 policy implications 337–8 prospect theory 333 nonprofit generated patents 311, 318–22, 323 non-profit organizations 92, 122–5, 132, 301 nonprofit patent litigation 311–18 Obama, Barack 394, 458 Maker Movement 447 Office of Science and Technology Policy (OSTP) 65 Office of Sponsored Programs (OSP) 77 Oliver, E. 256–282 online licensing, technology portals for 167 Open Access (OA) 392 Open Access movement 386 Open Data (OD) 364 data protection regulation 393, 403–5 collection limitation principle 404 critical issues in 398–408 data quality principles 404 purpose limitation principle 404 transparency principle 404 use limitation principle 404 detrimental effect of 404 dissemination of 403–4 EU Horizon 2020 research project 405–8 and intellectual property rights (IPRs) 395–8 “one size fits all” solution 409 opportunities, interests and needs 393–5 specific purpose versus any purpose 408–10 Open Government Data 394–5 Open Innovation (OI) 93, 290, 388, 392, 448, 449, 478–9 Open Knowledge International 394 open “letters patent” 285 Open Research Data Pilot 405–7 open science and data-driven society concept of 391–2 open data 393–5 premise 391–3 Open Science movement 365, 386 option agreements 172–3 Organisation for Economic Cooperation and Development (OECD) 394 data privacy regulation 400 organized anarchies, characteristics of 101 organized research units (ORUs) 104, 107 organized technology markets 285–8 Osage University Partners 180

Index ownerships of copyright 72 creations by universities 24–5 copyright policy and 25 rules concerning 24 of institutional works 74 of Intellectual Property 15, 25, 75, 118, 417 of patents 332–7 of works of non-employees 74 patent activity, in Europe 422–4 patent assertion entities (PAEs) 53, 55, 65, 244, 250, 253, 291, 295, 422 patent budgets, funding of 169–70 patent commodity exchanges 305–6 Patent Cooperation Treaty 322 patent enforcement 2, 54, 309–11, 318–22, 325–6, 328, 338, 340, 356, 448, 467–8 patent exchange broad exchanges, marketplaces and aggregators 296–301 brokerages 293–6 building blocks of 298 Green Patent exchange 307 rise of 288–302 social responsibility-oriented platforms 301–2 subject specific transfer marketplaces limited exchanges and brokerages 293–6 single-entity marketplaces or portals 290–93 patent inventions 239, 250, 414–15 government funding to 69 in universities 236 patent licensing and enforcement companies (PLECs) 172, 295 patent litigation system 52, 56 patent marketplace 257–8 analysis of academic patent market 272–5 foreign entities 270–72 findings 262–70 buyers 266 sellers 263–5 technology and litigation 266–70 litigation experience and strategy 271 patent portfolios 272 university patent assignments 258–60 proceeds from 262 sales 260–62, 263 patent monetization entity 315 patent rights 7–10, 19, 44, 53, 56, 166–7, 172–3, 178, 180, 193, 228, 239, 245, 256–7, 259, 271–3, 275, 296, 305, 325, 345, 377, 422, 448, 457, 474, 476

487

patents 80 age at litigation 313 auctions 303–5 Bayh-Dole Act 420 Boston University litigation 316–17 California Institute of Technology 317–18 currents and crosscurrents 318–22 definition of 72 double-patenting 63 enforcement practices 54 against multiple defendants 312 for technology transfer 328–32 of university 318–22 filing strategies 346–9 financial incentive from royalties 63 “first-inventor-to-file” system 49 “first-to-invent” system 49 FRAND-licensed 287 in higher education 449 infringement of 261, 313, 461, 478 in Australia 474–8 in Canada 472–3 in European Union 473–4 litigation against 344 in United States 467–71 Institutional Patent Agreements (IPAs) 45–6 licensing portfolios of 331, 351 litigation system 52 “micro-entity” patent application status 52 nonprofit generated patents 318–22 nonprofit related 322–3 open “letters patent” 285 ownership of 34, 286, 332–7, 448 protection of 326 Kitch’s theory for 333 of public goods 237 ThermoLife litigation 315–16 for three-dimensional (3D) printing and education 454–7 to-be-patented technology 329 by type of organization 312 US patent system 49–50 of US universities see university patents Patent Transparency and Improvements Act (2013) 53 Patent Trial and Appeal Board (PTAB) 56, 262, 343 patent troll 53, 65, 248, 253, 295, 310–15, 320–22, 325 patentability circumstantial impairment of 73 general criteria for 73 international variation of 73 loss of 73 patentable inventions 7, 10, 23, 33, 70, 72–3, 88, 236, 371, 374, 377, 381, 390

488 Research handbook on intellectual property and technology transfer patenting of invention 420 process of 15 “personal connection” problem 307 Pew Charitable Trusts 152 Pharmalicensing 294 Pilz, Bryce 166–194, Polanyi, Michael 215 portfolio licensing 351 post-AIA Patent Trial and Appeal Board (PTAB) 56 Powell, Walter 225–6 pre-negotiated Startup License 191 presidential associations 42 Principal Investigator (PI) 12, 93, 159, 358, 432 “private” market 257 problem solving 96, 222, 476 product-implementing businesses 332 product life cycles 266 professional bureaucracies 100 professional development, strategies for 138 professionalization, theories of 114–15 proof-of-concept centers 94, 204, 211, 302 “proof of concept” programs, for university research 62 property rights 5, 56, 77, 237, 372, 391, 416, 420–21, 432, 439, 460 “proprietary data” sensibility 372 Protecting American Talent and Entrepreneurship (PATENT) Act 54 public accountability, act of 64 public domain technologies 334 public education 132–3 public investment, in university research 240–41 public-private partnerships 83, 456, 459 Public Research Organizations (PROs) 257, 412, 414, 426, 454, 455, 459 Public Sector Information (PSI) Directive, Europe 395 public-sector organizations 307, 455, 466 public stock offerings 257 pure research university, model of 87–8, 91 Pyne, Christopher 466 quasi-public “brokered” market 257 Radio Frequency ID (RFID) 286 “Reclaim Invention” program 321 Republic of Letters 375 Republic of Science 378 Research Corporation for Science Advancement 253–4 research and development (R&D) 10, 25, 38, 87, 93, 120, 193, 199, 207, 238, 241, 243–4, 246, 254, 300, 302, 319, 331, 377, 412 research innovation

IP revenues 170 technology fee model of 170 research parks 205 research universities 375 allocation of IP interests in research products 383 capacity building 107 concept of design 99–102 knowledge-based model 380 resource allocation 115, 120–21, 385 return on investment, to technology transfer 196, 206–9 Revenue Centered Management (RCM) 112, 119 revenue distribution policy (RDP) 168 revenue sharing across the institution 26–7 amongst co-creators 27 IP Policies on 27 Rinehart, Marc Daniel 143–165 Rimmer, Matthew 447–79 Rockefeller Foundation 44, 246 Romano, M. 412–433 royalty buyouts 183–4 payments 171, 193 stacking 182, 286 Rooksby, Jacob H. 1–2, 61, 69–71, 80–84, 124, 131–142, 309, 448, 454, 478 Ryan, Christopher J. 236–255 Saint Regis Mohawk Tribe 355 sales-based royalty 171–3, 177, 192–3 science-based entrepreneurial firms 232 scientific research, in Chinese universities 440 Sebeok, Jessica A. 41–68 self-governance, of the university 379 shareholder investment 228 Sherer, Todd 196–213, Sherwood Group 137, 140 Singer, Peter 286, 459 Small Business Innovation Development Act (1982) 49 Small Business Innovation Research (SBIR) Program 49, 136, 139 Small Business Technology Transfer (STTR) program 49 smart cities 398 social responsibility 301–2, 307 versus licensing for income 37–9 social science research 119, 388 social values 212, 236–7, 246, 251–3, 388, 438 of academic institutions 118 Society of University Patent Administrators (SUPA) 46–7, 84, 131, 132, 134–5, 137, 141 software distribution licenses 173

Index software startups, equity-only framework for 191, 192 sovereign immunity, defense of 345–6, 349, 352–5, 362 sovereign patent funds (SPFs) 272 “special” government employees (SGEs) 151 “spinout” companies 259 sponsored research 17, 25, 39–40, 79, 107, 142, 159, 170, 183, 191–2, 224, 228–9, 242 standard setting organizations (SSOs) 284, 287–8 Stanford Nonpracticing Entity Database (SNE Database) 311 Stanford v. Roche 10, 22, 34 start-ups 4, 49, 197 creation of 200 failed startups 325 university-supported 200 STEM-oriented units (SOUs) 92, 119, 125, 128 design-based strategies 94 design of 94, 102 incentives-based planning and budgeting 119 Sui Generis Database Right (SGDR) 396–7, 408 sui generis law 408 Support Technology and Research for Our Nation’s Growth (STRONG) Patents Act 56–7 Support Technology & Research for Our Nation’s Growth and Economic Resilience (STRONGER) Patents Act (2018) 363 sustainable development 119, 456 sustainable technology 302 tacit knowledge, in university-industry technology transfer 215–24 academic engagement in 226 “best mode” requirement 217 challenges of 222–4 commercial value of 219–22 dimensions of 215–19 importance of 233–4 mechanisms for concept centers and incubators 229–30 consulting engagements and corporate positions 226–8 networks 224–6 sponsored research 228–9 university spinoffs 231–2 nature of 218, 223 Targeting Rogue and Opaque Letters (TROL) Act 54–5 technical “know-how” 73 Technische Hochschulen 376 technological innovation 199, 438, 442–3 technology commercialization 69, 144, 203, 229, 273, 275, 454 technology fees 170

489

technology innovation 138, 219, 384 technology licensing 132, 140, 285, 292 Stanford’s mission statement on 77–8 technology portals, for online licensing 167 technology portfolios, management of 131 technology transfer 1, 4, 131 academic 144 challenges of 222–4 in Chinese universities see Chinese universities, technology transfer in communalism-based objections to 30 drug pricing, issue of 58–61 economic benefits of 211 ecosystem of 49 effectiveness of 232 within European Union see European technology transfer evolution of 167 future politics and political future of 61–7 human interaction in 307 ‘informal’ transfers 5 in joint research projects 18 legislation involving 49 patent enforcement and 328–32 political history and historical politics of 43–57 political projects 41 politics of 43 case study in 58–61 process of 5, 12–13 return on investment to 206–9 role of 198–202 services to ensure effective 78–9 structure of 12–19 through the sale and licensing of intellectual property 283 university-industry technology transfer 214 university licensing and 232–4 Technology Transfer Committee (TTC) 86 technology transfer deals general structures of 171 types of 170 biomaterials licenses 173–4 end user licenses 173 material transfer agreements 174 option agreements 172–3 patent licenses 172 software distribution licenses 173 Technology Transfer Evolution Working Group 167 technology transfer offices (TTOs) 4, 77, 94, 139, 166, 197, 215, 236, 309, 339–40 AAAS Report (1934) 6–7 advantages and disadvantages of 245–6 Bayh-Dole Act (1980) 9–12 Covered Business Method Review (CBM) 341 creation of 244–5

490 Research handbook on intellectual property and technology transfer development of 5–12 economic development model of 17 efficiency of 419 within European Union 412, 418–19 expanded and evolving roles of 167–8 expansion of 244 experimental structures of 18–19 facets of Michigan’s TTO operations 78–9 innovations handled by 167 internal structure of 15 inter partes review (IPR) 341 key objective indicators of 4 manual of the goals of 84 methodologies for IP valuation 174–83 missions of 17 necessity of 13–14 obligations of 13 Office of Technology Licensing (OTL) 70 Office of Technology Transfer (OTT) 70 operation of 16–18 organization of 15–16 patent/licensing structure of 31 portfolio approach to valuing university IP assets 190–92 Post Grant Review (PGR) 341 problems for institutions with determining inventorship and allocating revenue 33–7 incentivizing disclosure 29–33 industry partnerships 39–40 social responsibility versus licensing for income 37–9 profitability of 418 and public good 246–8 restructuring and repositioning of 124 and the role of IP valuation 166–70 structural location of university-housed 124 types of 124 universities setting up 420 University IP Rights 6–9 university patents and 236 TechTransfer Central publication 191 Text and Data Mining (TDM) 396 Thermolife International 315–16 three-dimensional (3D) printing 449–54 academic patenting of 455 applications of 456 civil liability in respect of 464 disruptive technologies of 454 investments in respect of 455 patent exceptions for 471 patent infringement to Australia 474–8 Canada 472–3 European Union 473–4

United States 467–71 patent landscapes of 454–7 patent ownership in Australia 464–7 Canada 459–61 European Union 462–4 United States 457–9 remote printing services 463 use in educational institutions 463 value of 468 Total Quality Management 126 trademarks 80 license use of 81–2 registration of 82 state university trademarks 82 trade secrets 73, 80, 328, 408 confidential information as 83 protection of 82–3 trade secret law 83 university-based information as 83 translational gap funding 176 transparency, principle of 405 trans-research institution 385 trans-university 385 Trudeau, Justin 459, 472 Trump, Donald J. 310 trustees, of public institutions 123 Turnbull, Malcolm 465 Unified Patent Court (UPC) 417, 422 Unitaid Medicines Patent Pool 302, 307 Unitary Patent (UP) 422 United Kingdom Research and Innovation (UKRI) 386–7 United Nations Declaration of Human Rights 398 United States-Mexico-Canada-Agreement (2018) 461 unit license rights 305 Universities Allied for Essential Medicines (UAEM) 458 university as commercial entity with public purpose 88–9 as pure commercial entity 89–90 University and Small Business Patent Procedures Act (1980) see Bayh-Dole Act (1980) university-based inventing 365 university-based technologies, commercialization of 200 university budgets 14, 115, 119 university economic engagement 167 university entrepreneurship 4 university-firm networks 123 university generated patent see university patents university-generated technology 38, 207 university-government partnership 45

Index university-industry-government relations 49 university-industry relations 63, 118 university-industry technology transfer 205, 214 encouragement of 200 role of tacit knowledge in 215–24 university intellectual properties 77–8, 424–6 university inventions 37, 51, 60, 215, 218–20, 224, 226, 228–9, 231–4, 330, 333, 346, 435 commercialization of 232 ownership of patents 19 University IP Rights, history of 6–9 university knowledge and discoveries 208 University of California 111, 113 Department of Plant and Microbial Biology (PMB) 117–18 University of Florida (UF) 5, 21, 107, 242 University of Illinois at Urbana-Champaign (UIUC) 111 University of Minnesota 39, 190–91, 246 University of Wisconsin 7, 9, 11, 292, 353, 376 licensing program 292 university patents 322, 420–22 see also technology transfer offices (TTOs) assignments 258–60 Bayh-Dole Act and 240–41 brief history of 238–44 gamble of 242–4 government funded university research 238–9 government patents 238–9 licensing and technology transfer 232–4 patent portfolio 243 policy of 252–3 pools of 253–4 public investment 241–2 sales 260–62 solving failures of 248–54 clarity 249–50 tacit knowledge see tacit knowledge, in university-industry technology transfer

491

university research, transformation of 438, 439–40 university-run enterprises, initiation period of 438–9 university spinoffs 136, 139, 215, 231–2, 234 university technology transfer see technology transfer US Patent and Trademark Office (USPTO) 49, 51, 310, 340 assignment database 260–61 Examination Guidelines interpret AIA 52 Vertinsky, Liza 196–213 venture capital 125, 136, 171, 186, 189, 228–9, 259, 332, 361 Vietnam War 45 Warshaw, Jarrett B. 92–130 Weinberg, Michael 471 Whitaker Foundation 116, 129 Wilbur, Ray Lyman 43–4 Wisconsin Alumni Research Foundation (WARF) 7, 9, 15, 139, 292 Wissenschaft 375–7 World Intellectual Property Organization (WIPO) 449 Green Patent exchange 307 World Trade Organization 456, 472 Canada-Patent Protection case 472 Xiaoxue, Shi 434–446 Yale Startup License 191 Zerit® (HIV anti-retroviral drug) 38 zero-based budgeting 126 Zero Marginal Cost Society, The 452