Language Barriers and the Speed of Knowledge Diffusion

         
Author Name Kyle HIGHAM (Institute of Innovation Research, Hitotsubashi University) / NAGAOKA Sadao (Faculty Fellow, RIETI)
Creation Date/NO. August 2022 22-E-074
Research Project Building innovation capability and incentive: evidence from micro data
Download / Links

Abstract

While language barriers are well-known obstacles to knowledge diffusion, quantitative research on this topic is sparse. In this work, we attempt to fill this gap by providing causal evidence on their effects on the speed of knowledge diffusion by exploiting the introduction of pre-grant publications by the American Inventors Protection Act (AIPA) in 2000. We find that this policy significantly accelerated, relative to Japanese inventors, US inventors’ use of Japan-originating technical knowledge in their patents. Our analysis controls for biases of patent citations as proxies of knowledge flow, including preference for citing local prior art. Consistent with incentives for translation, this acceleration is much larger for small firms and the firms with little investment in the Japanese market. Consistent with high uncertainty of foreign patents before translation, we see much larger effects of the AIPA on the patent applications with higher quality. Our findings suggest that pre-grant publication provides a significant public good for cumulative innovation through earlier translations of foreign patents.