Linking Financing To Decisionmaking In The U.S. Patent System

Michael Frakes & Melissa F. Wasserman, Does Agency Funding Affect Decisionmaking?: An Empirical Assessment of the PTO's Granting Patterns, 66 Vand. L. Rev. (forthcoming, 2013), available at SSRN.

As patent law has grown in social and economic importance, a growing number of scholars have given attention to analyzing the structure of the patent system.  The number of patents issued per year has increased significantly in the past few decades, and, in at least some fields, the breadth of products or activities on which these issued patent claims read has also increased.  Scholars studying the system have given the design and functioning of the U.S. Patent and Trademark Office (PTO) increased scrutiny on a number of fronts.  Among its many duties, central is the PTO’s power and obligation to decide whether the rights to control a putative invention belong to the patent applicant, to a different applicant, or to the public.

Although patent prosecution is an ex parte proceeding, the patent examiner stands between the applicant and the public in deciding who shall receive the entitlement to use, or exclude others from using, the invention.  Due process norms usually require such decisionmakers to be unbiased and free from conflicts of interest.  However, since 1991, the PTO has been paid to make this decision from fees paid by one of the parties to the decision:  the applicants or patentees.  Not surprisingly, senior PTO officials have on more than one occasion referred to applicants and patentees as the office’s “customers.” Might this financing arrangement affect the agency’s interpretation and application of the law?  Most scholars have assumed that it does to some extent, but this extent and the operation of this assumed bias has been underexamined.  Until now.

Michael Frakes and Melissa Wasserman rightly decided that the hypothesis of PTO bias was testable.  They have done a very nice job in identifying which aspects of the agency’s funding scheme exert the greatest incentives toward favoring a patent grant in a close case.  In particular, more than half of the PTO’s budget is funded from issuance and renewal fees, which are paid only after the agency has made a positive patentability determination. The key takeaways from their paper, Does Agency Funding Affect Decisionmaking?: An Empirical Assessment of the PTO’s Granting Patterns, are these:

[O]ur findings suggest the PTO is preferentially granting patents on technologies with high renewal rates and patents filed by large entities, as the PTO stands to earn the most revenue by granting additional patents of these types. Furthermore, we also find that these distortions are more likely to occur when markers indicative of an underfunded PTO are present.

By using the 1991 change in financing to test for decisionmaker bias, the authors had to control for a number of other variables. They detail how they did so in Appendix B of the paper.  Further details on the methodology are well explained in the paper. (A note to the reader:  I am conversant in basic statistics, but I admit that I lack the qualifications to conduct rigorous peer review of the statistical methodology used.  The authors appear to have considered and accounted for the likely confounding influences that would undermine confidence in their causality inferences.  There may be some minor corrections needed, but given the care and detail with which the authors have designed their study and reported their findings, I am sufficiently confident in the soundness of their work to recommend their article.)

To those who may object to the selection of this article for review because it primarily reports the result of empirical analysis rather than legal analysis, I have two responses.  First, while most legal scholars lack the training and expertise to conduct empirical studies of the legal system, some do, and that’s a good thing.  Those who study law and policy are particularly well-suited to design studies such as this.  While many other legal scholars were willing to simply assume bias, these authors dug in to examine the evidence.  Second, legal scholars will always have to make certain empirical assumptions when conducting policy or legal analysis because not all empirical hypotheses are testable, and, even when they are, time and research funding constrain our collective ability to test these.  Meanwhile, the world turns on.  So, it becomes a matter of judgment to determine which hypotheses are testable and worth testing.  Kudos to Professors Frakes and Wasserman for judging this to be one such hypothesis.