The quality of innovation “Booms” during “Busts”

Models of creative destruction assume that recessions are periods of reallocation and disruption, generating new ideas that catapult new firms to the frontier. However, current empirical evidence suggests that research and development (R&D) expenditures and patenting are procyclical, not countercyclical. In this paper, we introduce new insights to enrich this debate. First, using panel data on the quality of innovation between 1980 and 2019, we document that the quality of innovation is countercyclical: Innovations produced during busts have a larger effect on the path of future research than those developed during booms. Second, we investigate several additional patterns on the composition of R&D. We find that the procyclicality of R&D is concentrated among firms that are more financially constrained, and that time allocated towards basic science (applied) research is countercyclical (procyclical). These results highlight the importance of composition effects among not only the organizations that contribute to innovation over the business cycle, but also the tasks that employees within these organizations perform.