Firm Startups, Population Growth and Domestic Migration

In this article we causally link the effects of population growth to firm startups. We attempt to rule out endogeneity in multiple ways. First, we show that the effect of population in a fixed-effects panel is robust to the inclusion of employment growth, which is the most plausible mechanism through which startup rates would cause population growth. Second, we utilize an instrumental variable for population growth based on metro-to-metro migration rates from Howard (2017), which creates a proxy for annual migration into a specific metro area that is not subject to reverse causality by using historical migration patterns combined with annual changes in migration flows between other metro areas. Finally, we utilize dynamic panel models. Overall, we find a robust and consistent effect of population growth on startup rates. The effect appears to operate through the working-age population, consistent with labor supply rather than demand.