Drawing upon recent advances in machine learning and natural language processing, we introduce new tools that automatically ingest, parse, disambiguate, and build an updated database using U.S. patent data. The tools identify unique inventor, assignee, and location entities mentioned on each granted U.S. patent from 1976 to 2016. We describe data flow, algorithms, user interfaces, descriptive statistics, and a novelty measure based on the first appearance of a word in the patent corpus. We illustrate an automated coinventor network mapping tool and visualize trends in patenting over the last 40 years. Data and documentation can be found at https://console.cloud.google.com/launcher/partners/patents-public-data.

Much work on innovation strategy assumes or theorizes that competition in innovation elicits duplication of research and that disclosure decreases such duplication. We validate this empirically using the American Inventors Protection Act (AIPA), three complementary identification strategies, and a new measure of blocked future patent applications. We show that AIPA—intended to reduce duplication, through default disclosure of patent applications 18 months after filing—reduced duplication in the U.S. and European patent systems. The blocking measure provides a clear and micro measure of technological competition that can be aggregated to facilitate the empirical investigation of innovation, firm strategy, and the positive and negative externalities of patenting. This paper was accepted by Joshua Gans, business strategy.

We propose a new method to estimate and isolate the localization of knowledge spillovers due to the physical presence of a person, using after-application but pre-grant deaths of differently located coinventors of the same patent. The approach estimates the differences in local citations between the deceased and still-living inventors at increasingly distant radii. Patents receive 26 percent fewer citations from within a radius of 20 miles around the deceased, relative to still-living coinventors. Differences attenuate with time and distance, are stronger when still-living coinventors live farther from the deceased, and hold for a subsample of possibly premature deaths. (JEL O31, O33, O34, R32)