Pro-cyclical Aggregate R&D: A Comovement Phenomenon

Min Ouyang
2011-01-01
Abstract:Pro-cyclical aggregate R&D has been taken as evidence against the conventional Schumpeterian view on the optimal timing of innovation. We decompose aggregate R&D and real GDP in the U.S. into those by 22 industry groups. Surprisingly, we find only 5.67% of pro-cyclical aggregate R&D reflects withinindustry timing of innovation and production, but 94.37% arises from interindustry co-movement between R&D and output. We posit pro-cyclical aggregate R&D, just like the business cycle itself, is a comovement phenomenon. Email: mouyang@uci.edu. Hugo Hopenhayn, Tony Braun, and Seminar participants at USC, the Brookings Institution, 2010 Midwest Macro Meeting, 2010 LACEA, the Atlanta Fed, and UC-Irvine provided helpful discussions. All errors are mine.
What problem does this paper attempt to address?