Abstract
We combine novel micro data with quasi-random timing of patent decisions over the business cycle to estimate the effects of the Great Recession on innovative startups. After purging ubiquitous selection biases and sorting effects, we find that recession startups experience better long-term outcomes in terms of employment and sales growth (both driven by lower mortality) and future inventiveness. While funding conditions cannot explain differences in outcomes, a labor market channel can: recession startups are better able to retain their founding inventors and build productive R\&D teams around them.