U.S. R&D and Japanese Medium Term Cycles

Abstract

In the thirty year period between 1960 and 1990 Japan saw labor productivity rise from a level of 27 percent of the U.S. to 87 percent of the U.S. This development miracle can be explained by an initial low capital stock and measured variations in TFP. These facts motivate our investigation into the sources of Japanese TFP variations. We consider Japanese and U.S. data that is filtered to retain medium cycle events such as the productivity slow down in the 1970's. An investigation of Japanese medium cycles reveals an important role for the diffusion of business ideas from the U.S. to Japan. U.S. R&D leads Japanese TFP by four years and accounts for as much as 60% of the variation in medium term cycle Japanese TFP. Japanese R&D, in contrast, is coincident with Japanese TFP. Simulations designed to isolate the roles of Japanese and U.S. R&D suggest that the diffusion of knowledge from the U.S. is a key driver of Japanese medium cycles. Interestingly, our theory also accounts for Japan's experience in the 1990s. Slow growth during this period was preceded by a sharp and persistent decline in U.S. R&D.