What\u27s the Connection between Repeat Litigation and Patent Quality? A (Partial) Defense of the Most Litigated Patents

My inspiration for the article was two articles by John Allison, Mark Lemley and Joshua Walker analyzing the characteristics and litigation success of the most litigated patents. (Allison et al. (2009) published by the University of Pennsylvania Law Review and Allison et al. (2011) published by the Georgetown Law Journal) Understanding the most litigated patents ( MLPs ) is important because of the possibility James Bessen and Michael Meurer (2008) are correct that since the patent litigation surge of the 1990s the patent system has created net disincentives to innovate in most industries. The vast majority of patents are never litigated and of the few that are most are asserted in one or two different lawsuits. Thus, MLPs are responsible for a disproportionate share of costly patent litigation. Optimistically, both Allison et al. (2011) and I theorize that MLPs should have greater litigation success because their patents should tend to possess higher legal quality (though our reasons differ and my criticism of the basis for their prediction is interestingly connected with the new joinder provision of the America Invents Act). More success would indicate that the high costs MLP owners ( repeat patent plaintiffs ) impose are at least partially offset by the greater benefits they reap from their litigation. Allison et al. (2011) surprisingly find that repeat patent plaintiffs are overwhelmingly likely to lose. In my paper, I reanalyze the relationship between the number of lawsuits a patent is asserted and litigation success using a much larger (and I argue a more carefully constructed) sample of judgments. In contrast with Allison et al. (2011), I find that in general repeat patent plaintiffs are much more successful, supporting the optimistic view of their impact on innovation. However, this optimism does not extend to repeat SOFTWARE patent plaintiffs, who are not more likely to win adjudicated disputes because they are not more likely to win infringement judgments. Within the context of a selection effect model, I argue this finding is most likely explained by Bessen and Meurer\u27s (2008) theory that software patents possess more uncertain property boundaries. I hope that your editors will review this article and give it serious consideration. While I need to begin submitting it to other journals, I would prefer to have it published in the STLR because you have been so proactive in reaching out to me. I believe the article is publication-ready but there are several content changes I could easily make and that your readers might prefer. For example: 1) Limiting or removing my criticism of the basis for Allison et al.\u27s (2011) hypothesis that repeat patent plaintiffs should be more successful (Part II.A. and Appendix); 2) Replacing the formal selection effects model I include at the end of Part II.B with a qualitative discussion of the effect of different selection variables on patent owner win rates; 3) Including a more detailed explanation of the connection between control variables and litigation success in my regression analysis in Part II.B.2; and/or 4) Including more discussion of the possible impact of the America Invents Act\u27s joinder rule on repeat patent litigation. I look forward to hearing from you. Please feel free to call me at 989-600-2694 if you have questions or need more information to make a decision. Thank you again for your interest in my work