Judges, juries, and patent cases - an empirical peek inside the black box.

• Author: Moore, Kimberly A.

Honest to God, I don't see how you could try a patent matter to a jury. Goodness, I've gotten involved in a few of these things. It's like somebody hit you between your eyes with a four-by-four. It's factually so complicated.

-- Judge Alfred V. Covello(1)

INTRODUCTION

The frequency with which juries participate in patent litigation has skyrocketed recently. At the same time, there is a popular perception that the increasing complexity of technology being patented (especially in the electronic, computer software, biological and chemical fields) has made patent trials extremely difficult for lay juries to understand. These developments have sparked extensive scholarly debate and increasing skepticism regarding the role of juries in patent cases.(2)

Juries have participated in some aspects of patent litigation since the enactment of the first patent statute in 1790, which provided for "such damages as shall be assessed by a jury."(3) The enactment of the Patent Act of 1870, however, which gave equity courts the power to award common law damages,(4) spawned an era in which patent cases were almost exclusively decided by the bench. This pattern has changed only recently -- and the change has been dramatic. In 1940, 2.5% of all patent cases tried in district court were heard by juries.(5) From 1968 to 1970, the figure was almost unchanged at 2.8%.(6) By contrast, from 1997 to 1999, 59% of all patent trials were tried to juries.(7) This surge in jury requests has prompted a flurry of recent litigation over the right to a jury trial in patent litigation.(8)

[ILLUSTRATION OMITTED]

Despite extensive debate over the role of the jury in patent cases, no comprehensive empirical research has been done to ascertain, to the extent possible, the differences between jury and judge resolution of patent cases and the cause of the increased demand for jury trials in recent years. Given that patent litigation is an expensive endeavor -- it routinely costs each party in excess of a million dollars(9) -- there is an urgent need for empirical evidence on patent litigation. This Article undertakes that task by providing the first large-scale comparison of patent-holder win rates and recoveries in cases tried before juries and judges. The data include all patent cases that went to trial in the period from 1983 through 1999 (seventeen years of data).(10) This time period was selected in order to analyze, among other things, the impact the creation of the Federal Circuit may have had on the resolution of patent trials in the district courts.(11) Accordingly, each of these cases was followed through to appeal to ascertain the issues appealed and the relative affirmance rate by issue, by adjudicator, and by year. The data were analyzed to determine whether popular perceptions or theoretical models about judges and juries can be validated, to provide descriptive statistical data, and to develop predictive models from the data through regression analysis.

At first blush, the results of the study suggest that complaints about jury bias and incompetency are unfounded. Judges and juries decide some issues differently. For example, juries are significantly more likely to find patents valid, infringed, and willfully infringed than judges. The differences, however, are not as profound or pervasive as one might expect. Judges and juries find patents enforceable with similar frequency. Additionally, juries seem as "accurate" in their decisionmaking as judges are, as measured by appellate affirmance rate.(12)

And yet, despite similar affirmance rates for judge and jury trials, there is some ground for concern with jury resolution of patent cases. To a greater degree than judges, juries tend to decide whole suits rather than delineate individual issues, even when separate issues are presented to them via special verdict forms or interrogatories. This finding suggests that judges are subtler at managing the complex nature of patent cases and the technical distinctions between patents and products. It may also affirm the popular perception that juries are unduly swayed by tangential factors.

In addition, who filed the suit is a significant predictor of win rate in jury trials. Juries are significantly pro-patentee in suits for infringement (68% patentee win rate); but when a possible infringer initiates a declaratory judgment action, the patentee only has a 38% win rate. If the same were true of judges, then one could attribute the difference in win rate to the strength of the cases -- namely, that alleged infringers only bring declaratory judgment suits when they have strong cases. But patentee win rates are substantially uniform in bench trials, regardless of who initiated the suit.

These data suggest that there may be some problems with juror adjudication of patent suits, though the system masks them. Deferential standards of review leave the Federal Circuit with little ability to disturb potentially flawed jury decisions. Moreover, the system lacks sufficient transparency to ascertain flaws in jury verdicts. The "black box" nature of jury verdicts leaves the Federal Circuit unable to correct inaccuracy or bias on the part of jurors. This reality -- particularly in light of the increase in jury adjudication of patent disputes and the potential for jury error where increasingly technical inventions are involved -- highlights the value of a peek inside the black box.

Part I of the Article presents popular impressions of judge and jury outcomes in patent cases -- in particular, popular perceptions of juror incompetence and bias -- and considers how win-rate data might confirm or refute these beliefs. Part I also discusses selection effect theory, an economic model of the case selection process, and how win-rate data may be affected by parties' knowledge of adjudicator biases. Part II describes the data set, its acquisition, and the methodology used to analyze the data. Part III tests the impressions of judge and jury outcomes in patent cases against the empirical data. It presents descriptive statistics, the hypotheses, and the results of the regression models. Part III also discusses what insight the data lend on the role of the jury in the adjudication of patent disputes.

1. IMPRESSIONS OF JUDGE AND JURY OUTCOMES

   1. Popular View: Juror Incompetence

      The increased participation by juries in patent cases and the detailed attention given by the judiciary has caused a number of scholars and other commentators to question the propriety of jury resolution of patent cases. In this Part, I discuss the perceived wisdom regarding juries in patent cases. Typical complaints about the use of juries in patent cases include: juries are unable to comprehend the technology(13) or the nuances of the legal standards for patent validity and infringement; juries are pro-patentee(14) -- they favor inventors and have a high regard for the U.S. Patent & Trademark Office ("PTO");(15) juries are biased in favor of domestic companies;(16) juries award excessively high damage awards;(17) and juries are swayed too easily by tangential factors.(18) Jury consultants believe that juries do have distinct biases and preconceptions in patent cases.(19) These popular perceptions of juror incompetence and bias have caused commentators to argue that the role of the jury in patent litigation should be severely limited, and many alternatives have been proposed.(20)

      Although purely anecdotal, the following transpired between a judge and the jury in an antitrust case which involved, what was at the time, complex technology after the jury returned its verdict:

      The Court: Do you know what demand substitutability is, [Juror A]?

      Juror A: Well, I would like to kind of look into that.

      The Court: Okay. And how about the barriers to entry, [Juror B]?

      Juror B: I would have to read about it....

      The Court: All right. And how about reverse engineering, [Juror C]?

      Juror C: That's when you would take a product and you would alter it in a, or modify it for your own purpose; that is, you would reverse its function and use it in your own method.

      The Court: And [Juror D], what is software?

      Juror D: It's software.

      The Court: Well, what is software?

      Juror D: That's the paper software.

      The Court: What's the hardware?

      Juror D: That's the wires and hardware.

      The Court: And what is -- do you know what an interface is?(21)

      Juror D: Yes.

      The Court: Can you given me an example of that?

      Juror D: Well, if you take a blivet, turn it off one thing and drop it down, its an interface change, right?(22)

      This apparent lack of comprehension of the underlying technology exemplifies the fears many harbor about jury resolution of patent cases. As technology becomes increasingly complex, especially in the software and biotech fields, concerns would naturally escalate over a lay jury's ability to comprehend the technology in order to resolve the suits. These complaints often revolve around the educational make up of the jury.(23) For example, after a jury ruled that AT&T had infringed a small company's patent, lawyers for AT&T complained that the jury consisted of "unemployed laborers and housewives [who] did not understand that stuff."(24) Despite increasing complexity of technology and the corresponding patents that protect it, there is no minimum educational requirement for serving on a jury in a patent case -- "blue ribbon" or expert juries are not mandated, regardless of technical complexity.

      If juries are unable to understand the technology or apply the law, then their decisions will be based on emotional or other irrelevant factors.(25) Who tells the better story? Who is the more likeable or sympathetic party? Many commentators suspect that the party who demands a jury in a patent case has a weaker case(26) and therefore prefers an adjudicator less likely to focus on the merits.

      Many attorneys believe that juries are mesmerized by the inventor's story and tend to favor the patentee.(27) Juries respond well to...