Significance
As the number of patents has grown rapidly over the past two decades, so has patent litigation. The rate of litigation per patent has remained roughly constant (a finding re-confirmed by this study). Also, survey evidence suggests that patent suits have become significantly more costly, causing concern that the total burden of patent enforcement is increasing substantially. This paper examines empirically the determinants of patent lawsuits and their outcomes.

By linking detailed information from the U.S. Patent and Trademark Office to data from the Federal court system and industry sources, the authors have assembled the most comprehensive data set to date covering all patent suits in the U.S. reported by the federal courts during the period 1978-1999. The data set enables them to construct a controlled random sample of the population of potential disputants and to evaluate systematically the variation in litigation risk across patents and their owners—a central issue for the enforcement of intellectual property rights and its economic consequences. The authors evaluate the characteristics that affect the decision to file a patent lawsuit as well as decisions of whether and when to settle a dispute or to proceed to adjudication at trial.

Results
The authors confirm and extend findings from their prior work (Lanjouw and Schankerman, 2001). Probabilities of litigation vary considerably, depending on the characteristics of patents, patent owners, and their technology field. The probability that a patent is litigated increases with its quality (measured by the number of claims, the number of citations made per claim, and the number of times other patents cite this one).

Regarding firm characteristics, there are considerable advantages to scale. The probability of litigation per patent is negatively related to firm size and the size of a firm’s patent portfolio. Patents held by individuals and small firms are more likely to be involved in litigation, although small firms are involved in absolutely fewer patent lawsuits than large firms. Also, characteristics that encourage "cooperative" interaction, such as concentration of firms within a technology field, reduce the likelihood of litigation. Litigation risk also varies widely across technology fields. While the overall average rate is 19.0 suits per thousand patents, rates vary from 11.8 per thousand for chemicals to 25-35 per thousand for computers, biotechnology, and non-drug health.

In contrast, key post-suit outcomes such as plaintiff win rates at trial and probability of settlement do not depend upon observed characteristics. The authors find win rates very close to 50 percent, supporting the view that the cases that actually go to trial are the ones where the two parties have very different views on the likely outcome (Priest and Klein, 1984). In addition, they find high post-suit settlement rates and that most settlement occurs soon after a suit is filed. These results reveal that enforcement of patent rights may rely more on the threat of action rather than on costly legal proceedings that consume firm and court resources.

References
Lanjouw, Jean O. and Mark Schankerman. 2001. "Characteristics of Patent Litigation: A Window on Competition," RAND Journal of Economics, vol. 32, issue 1, pp. 129-151.