Can natural disasters affect innovation? Evidence from Hurricane Katrina

Abstract

Studies of the geography of innovation have focused on how spatial proximity to human and material resources and institutions affect collaboration, knowledge flows, the demand and competition for invention, and the economic value of invention. I study a different source of the geographical determinants of innovation: exposure to large shocks. I conduct an inductive assessment, theoretically grounded on recent evidence that large exogenous shocks produce enduring fluctuations in risk aversion, to explain why Hurricane Katrina in the U.S. could have changed innovation outcomes. The difference-in-difference estimates show that, after an immediate fall, affected counties exhibit substantial increases in the growth rate of patenting and the quality of innovation compared to counterfactual counties. This correlation persists 10 years after the shock and is robust to measures of agglomeration and urbanization, firm resources, wealth, actual and expected income, education, external assistance, public policy, business cycles, and other county- and state-level factors. To account for the confounds of selective migration and network affiliation, I use narrowly georeferenced information to construct histories of inventors between 1999 and 2015 that allow me to follow the “Katrina effect” across geographies. The estimates imply that shock-affected individuals not only were more likely to patent, but became more skewed toward high-technology sectors.

Keywords: innovation, disasters, Hurricane Katrina, risk-taking behavior, risk preferences

JEL Classification: A12, D12, D91, O30, 031, R12