In this line of research I investigate the impact of patent races and patent scope on firm outcomes.

Patent Scope

(Kuhn & Thompson, working paper, data access)

The economic consequences of patents—both good and bad—largely depend on their scope, and some patents are much broader than others. Broader patents cover a wider range of competing products, enabling innovators to capture more of the social returns from their inventions through higher prices. Broader patents also cover more subsequent improvements to the inventions, which affects the incentives for follow-on innovation. The government sets the scope of each patent individually through a negotiation between the innovator and patent office. The legal rules governing patent scope are complicated, vague, and subject to change through judicial decisions, which gives patent examiners and judges significant discretion when deciding a patent’s scope. The resulting variation in patent scope may have important public policy implications, since the differential treatment of inventions is probably distortionary in some cases.

This paper presents and validates a simple, practitioner-grounded metric for patent scope: the number of words in a patent’s first claim. We examine why this metric works and its possible limitations and then test it against the judgments of legal experts. We find that patent examiners adding words to a patent’s first claim strongly predicts expert views of changes to the patent’s scope across most (but not all) technology classes. In our validation exercise, our measure outperforms other measures commonly used in the literature, some of which have virtually no predictive power. We then extend our measure to show how it can be used to draw causal inferences about patent scope based on random assignment of patents to patent examiners. Lastly, we implement this causal inference approach to show that an exogenous increase in patent scope significantly increases the likelihood of a patent becoming standards essential.

I have presented this paper at the NBER Productivity Seminar (2016), the EPFL MTEI Research Seminar (2015), and the MIT Laboratory for Innovation Speaker Series (2015). It was also presented at the Searle Center Conference on Innovation Economics (2016).

Patent Races

(Thompson & Kuhn, working paper)

 This project uses “Patent Twins” – cases where two discoverers file patent applications at the US Patent and Trademark Office (USPTO) for the same invention at the same time – to see how firms react to getting (not-getting) patent protection for their ideas. A large theoretical literature has emphasized the importance of these Patent Races between firms, but there have been empirical challenges to studying them in practice (detailed below).

This research takes advantage of a new data set from the USPTO to address these concerns. In genetics research, it is common to disentangle the impact of nature vs. nurture using twin studies. Scholars such as Michael Bikard have imported this concept into innovation research with “Paper Twins” – areas where a single discovery is made by multiple research teams and published simultaneously in the scientific literature (Bikard, 2014).

Our research takes this concept a step further and considers “Patent Twins” – instances where the same invention is filed with the patent office within a narrow window of time. Perhaps the most famous example of this is the patent filings by Alexander Graham Bell and Elisha Gray on the invention of the telephone within a few hours of each other.