Patent scholars tend to have backgrounds and interests that coincide with the skill set required for empirical scholarship, and so it is not surprising that many patent scholars have turned to empirical methods in recent years.  Many of their projects have been highly ambitious, taking advantage of the massive body of information generated by the patent system.  In recent years, research scientists and other stakeholders in the technological fields affected by the patent system have contributed their perspectives and insights to this work.  Today, empirical scholarship in patent law draws its data gathering and analytical methodologies not only from the social sciences, statistics, business, and economics, but also from computer science, the natural sciences, and engineering.

On October 23–24, 2008, the University of North Carolina School of Law community had the honor of welcoming a group of leading scholars whose work reflects the diverse interdisciplinary activities that are extending the frontiers of empirical patent law scholarship.  Four participants in particular exemplified this emergent interdisciplinarity:  Mark Calcagno, Senior Information Scientist at Procter & Gamble; Martin Campbell-Kelley, Professor of Computer Science at the University of Warwick; Xin Li, Ph.D. student in Management Information Systems at the University of Arizona; and Jon F. Merz, Associate Professor of Medical Ethics at the University of Pennsylvania.

Our keynote speaker, Judge Jay S. Plager of the U.S. Court of Appeals for the Federal Circuit, and our presenting sponsor Red Hat, Inc.’s vice president and assistant general counsel, Rob Tiller, highlighted a dinner welcoming an extraordinary collection of patent law scholars, gathered in the same spirit in which the nation’s first patent commissioner famously was said to have dined alone.   This was followed by a full day of panel presentations on patent litigation and policy, patents and information technology, patents and laboratory research, and patent examination and prior art.  All four sessions were well attended and followed by provocative and thoughtful exchanges.  I would like to thank Red Hat, Inc., and our program sponsors, the law firm of Kenyon & Kenyon LLP, the UNC Graduate and Professional Student Federation, and the law firm of Jenkins, Wilson, Taylor & Hunt, P.A., for their generous support of these proceedings.

Just as the symposium was getting underway, Flash of Genius—very likely Hollywood’s first major movie to center on a patent infringement trial—was closing its three-week run nationally.  The $16 million production grossed just over $4.5 million worldwide,  a gentle reminder to our distinguished guests that fame in the world of patent law is an esoteric phenomenon.  As data sets age, interest in particular empirical findings tends to be fleeting as well.  Still, in years to come, I hope these proceedings will come to be seen as a harbinger of a broader and deeper interdisciplinary engagement in the modes and methods of jurisprudence.

Recent proposals to amend the Patent Act  and the legal academy’s growing interest in empirical methods  have made the policy analysis of patent law a timely topic for colloquia  and symposia.   Empirical scholarship plays a vital role in these discussions, as it tests the theories, hypotheses, and characterizations that underlie legal rules and institutions, as well as proposals for their reform.

Patent scholars tend to have backgrounds and interests that coincide with the skill set required for empirical scholarship, and so it is not surprising that many patent scholars have turned to empirical methods in recent years.  Many of their projects have been highly ambitious, taking advantage of the massive body of information generated by the patent system.  In recent years, research scientists and other stakeholders in the technological fields affected by the patent system have contributed their perspectives and insights to this work.  Today, empirical scholarship in patent law draws its data gathering and analytical methodologies not only from the social sciences, statistics, business, and economics, but also from computer science, the natural sciences, and engineering.

On October 23–24, 2008, the University of North Carolina School of Law community had the honor of welcoming a group of leading scholars whose work reflects the diverse interdisciplinary activities that are extending the frontiers of empirical patent law scholarship.  Four participants in particular exemplified this emergent interdisciplinarity:  Mark Calcagno, Senior Information Scientist at Procter & Gamble; Martin Campbell-Kelley, Professor of Computer Science at the University of Warwick; Xin Li, Ph.D. student in Management Information Systems at the University of Arizona; and Jon F. Merz, Associate Professor of Medical Ethics at the University of Pennsylvania.

Our keynote speaker, Judge Jay S. Plager of the U.S. Court of Appeals for the Federal Circuit, and our presenting sponsor Red Hat, Inc.’s vice president and assistant general counsel, Rob Tiller, highlighted a dinner welcoming an extraordinary collection of patent law scholars, gathered in the same spirit in which the nation’s first patent commissioner famously was said to have dined alone.   This was followed by a full day of panel presentations on patent litigation and policy, patents and information technology, patents and laboratory research, and patent examination and prior art.  All four sessions were well attended and followed by provocative and thoughtful exchanges.  I would like to thank Red Hat, Inc., and our program sponsors, the law firm of Kenyon & Kenyon LLP, the UNC Graduate and Professional Student Federation, and the law firm of Jenkins, Wilson, Taylor & Hunt, P.A., for their generous support of these proceedings.

Just as the symposium was getting underway, Flash of Genius—very likely Hollywood’s first major movie to center on a patent infringement trial—was closing its three-week run nationally.  The $16 million production grossed just over $4.5 million worldwide,  a gentle reminder to our distinguished guests that fame in the world of patent law is an esoteric phenomenon.  As data sets age, interest in particular empirical findings tends to be fleeting as well.  Still, in years to come, I hope these proceedings will come to be seen as a harbinger of a broader and deeper interdisciplinary engagement in the modes and methods of jurisprudence.

In thinking about the range of issues we confront in empirical legal research, including research into patent law and policy, it is well to begin with fundamentals.  There are three inquiries that define what scholarship into the human condition cares about—knowledge, conduct, and governance.  Typically, scholarship regarding patent law and policy is concerned with conduct—who has the right, and under what circumstances, to make, use, or sell an invention that has been given recognition under the Patent Act  (more technically, who has the right to exclude others from making, using, or selling the invention)?  By contrast, empirical legal scholarship, including empirical studies of patent law, involves the process by which human knowledge is obtained—what we know and how we know it.  And when called upon to adjudicate patent disputes, courts necessarily engage both the knowledge and conduct questions in the context of governance, read broadly.

The question of conduct, what is permitted or prohibited under the patent laws, is a subject of much learning, but it is not the question to be addressed here.  Rather, our concern here is the question of knowledge, including some of the basic issues empirical scholarship confronts.  And, because it is empirical studies of law that are our subject of inquiry, a relevant question is how, if at all, does empirical legal scholarship relate to the method of governance, and in particular to the judicial process and the ways in which judges decide cases?

In recent years, many reform proposals have been presented in Congress for changing the patent system in the United States.  Most of these proposals have been normative in nature and based on overcoming the many perceived shortcomings of the United States Patent and Trademark Office’s (“Patent Office”) performance.  Nonetheless, actual legislative reforms have failed to materialize.

In this Article, we claim that in order to understand the chances of success of any reform to the patent system, we should take a closer look at the patent system’s political economy.  In particular, we should be aware of the different pressure groups with a stake in the system and their ability to influence congressional committees through which reform legislation is enacted.  We study the different constituencies favoring or opposing the reform and the strength of their bargaining power based on publicly available empirical data on political contributions by different groups and analyze the impact of political contributions to individual congressional representatives on individual floor votes on the Patent Reform Bill of 2007.  In addition, we also take into account the effect of the patent system on different technology industries and economic sectors.  As we show, each proposal will generate winners and losers who will try to push reforms forward or prevent them from being enacted.  In order to succeed, any reform will need a minimum consensus among these stakeholder groups:  firms in different technology sectors, inventors, the patent bar (divided separately into patent prosecutors and litigators), the Patent Office, and the courts.

Given the political processes, the final result of any reform will depart from any theoretical blueprint we describe in this Article.  As a consequence, a deeper understanding of this political process allows us to better understand the dynamics of reforms and the resultant characteristics of the patent system.  In the end, as in any other institutional device, the characteristics and performance of the patent system, plus its sustainability or reforms over time, depend on the preferences of the polity, specifically on the preferences of the groups with a definite stake in the performance of the system.  We also determine that the effects of the patent system on different technology and economic sectors will ultimately shape the different constituencies favoring or opposing the reform.  We observe that support for any patent reform will depend on the specific factors that define the structure of each economic sector.  Furthermore, in each sector, firms have different preferences depending on their economic power and particular stakes in the patent system.

Patent law is virtually alone in intellectual property (IP) in punishing independent development.  To infringe a copyright or trade secret, defendants must copy the protected IP from the plaintiff, directly or indirectly.  But patent infringement requires only that the defendant’s product falls within the scope of the patent claims.  Not only doesn’t the defendant need to intend to infringe, but the defendant may be entirely unaware of the patent or the patentee and still face liability.

Nonetheless, copying does play a role in some subsidiary patent doctrines, including damages rules, willfulness, and obviousness.  More significantly, the rhetoric of patent law (and of IP law more generally) often seems to presuppose that defendants in patent cases are in fact engaged in copying.  Similarly, the outcome of public policy debates over patent reform may well turn on the perception of patent infringers as either bad actors or as innocent businesspeople who accidentally ran afoul of a patent.

Unfortunately, no one seems to know whether patent infringement defendants are in fact unscrupulous copyists or independent developers.  In this paper, we seek to answer that question.  We look both at the allegations made in a random sample of complaints and at the treatment of copying in recent reported decisions, including willfulness decisions.  We find that a surprisingly small percentage of patent cases involve even allegations of copying, much less proof of copying.  Copying in patent law seems to be very much the exception, not the rule, except in the pharmaceutical industry.

While each patent dispute is unique, most fit the profile of one of a limited number of patent litigation stories.  A dispute between an independent inventor and a large company, for instance, is often cast in “David v. Goliath” terms.  When two large companies fight over patents, in contrast, they are said to be playing the “sport of kings.”  Some corporations engage in “defensive patenting” in order to deter others from suing them.  Patent licensing and enforcement entities who sue have been labeled “trolls.”  Finally, observers of the patent system call the use of patent litigation to impose or exploit financial distress “patent predation.”

These stories, routinely invoked by the press, advocates, and academics, shape public understanding of the patent system.  In this Article, I describe, then match, these stories to data on patent litigations to determine which types of suits are most prevalent.  I focus exclusively on the litigation of high-tech patents—covering hardware, software, and financial inventions—using data from the Stanford Intellectual Property Clearinghouse for cases initiated in U.S. District Courts from January 2000 through March 2008.

The data shows that the reality is more complicated than the rhetoric regarding patent litigation suggests.  For instance, many blame nonpracticing entities (“NPEs”) for a majority of the problems with the patent system.  But they bring only a minority of patent suits:  17% of high-tech patent suits in the last eight years.  However, NPEs often name multiple defendants and sometimes, rather than sue, are sued, for declaratory judgment (“DJ”).  Counting suits based on the number of defendants and including DJ cases, the NPE share rises to 28% of all high-tech patent suits.  This average reflects an increase in NPE suits as a proportion of all suits over an eight-year period, from 22% in 2000–2001 to 36% in 2006 to March 2008, counting defendants, or from 10% to 20%, counting cases.  I also report the variation by industry based on the absolute number of suits—the share of hardware patent NPE suits (26%) was nearly triple that of financial patent NPE suits (9%).  These numbers provide a richer context for understanding the NPE phenomenon.

Another widespread perception of the patent system is that large companies dread going to court, carefully constructing portfolios of patents to avoid doing so.  The practice of “defensive patenting” has been well-documented and theorized.  Yet, I found that public and large private companies initiated 42% of all lawsuits studied, 28% of the time against other large companies—the largest single category.  They also defend against many other suits, brought by NPEs, small inventors, and individuals.  These data suggest that defensive patenting, which is supposed to keep large firms out of court, is at least an incomplete—and perhaps a failing—strategy for many companies.

This Article also reports on the other major narratives of patent litigation.  4% of the suits were initiated by individual inventors (David v. Goliath), 18% of the suits were brought by small private companies against public or large private ones (small v. large), 16% pit one small- or medium-sized company against another (limited stakes), and in 8% of the suits, a large firm sued a small one (predation profile).  These and other findings provide a snapshot of patent litigation that should both inform current efforts to reform the patent system and serve as a basis for further investigation into its functioning.

The United States Patent and Trademark Office’s (“Patent Office”) move to a paperless search facility and the public’s growing involvement in prior art search have recently elevated the role of search engine technology in the patent examination process.  This Article reports on an empirical study that examines how this technology has systematically changed not only how patent references are found, but also which patents are cited as prior art.

Publicly available records do not provide information identifying the method by which each of the prior art references cited by a patent was found, such as keyword search, citation tracking, or classification search.  The main methodological contribution of this Article is to identify large sets of patent citations that are likely to exhibit characteristics similar to those of citations actually found through a particular search technique.  By applying this synthetic approach to a comprehensive citation database, this study compiles a large set of patent citations that can reasonably be imputed to keyword search.

A longitudinal analysis of this imputed data set indicates that examiners became increasingly reliant on keyword full-text search in the late 1990s, as the technology became accessible from their desktop computers.  This change in examination practice appears to have had a substantive effect on the choice of patents to be cited as prior art.  Specifically, patent citations imputed to keyword search tend to be co-classified (according to the Patent Office classification system) more frequently than patent citations in general and patent citations imputed to citation tracking methods.

These findings support the concerns of some commentators about Patent Office automation and the outsourcing of prior art search.  In particular, it appears that the Patent Office classification system is not being fully utilized to improve the precision of search results.  This Article concludes with a survey of some initiatives and techniques that have recently emerged to address this problem.