The Geography of Lifecycle Human Capital Accumulation

We examine how place shapes the production of human capital across the lifecycle. We ask: do those places that most effectively produce human capital in childhood also have local labor markets that do so in adulthood? We consider the following determinants of wages across place: 1) location-specific wage premiums, 2) adult human capital accumulation due to local labor market exposure, and 3) childhood human capital accumulation. We construct estimates of location wage premiums using AKM-style estimates of movers across US commuting zones and validate these estimates using evidence from plausibly exogenous out-migration from New Orleans in response to Hurricane Katrina. Next, we examine differential earnings trajectories among movers to construct estimates of human capital accumulation due to labor market exposure. We validate these estimates using wage changes of multi-time movers. Finally, we estimate the impact of place on childhood human capital production using age variation in moves during childhood. Crucially, our estimates of location wage premiums and adult human capital accumulation allow us to construct estimates of the causal effect of place during childhood that are not confounded by correlated labor market exposure. Using these estimates, we show there is a tradeoff between those places that most effectively produce human capital in childhood and the local labor markets that do so in adulthood. We find that each 1-rank increase in earnings due to adult labor market exposure trades off with a 0.5 rank decrease in earnings due to the local childhood environment. This pattern is closely linked to city size, as adult human capital accumulation generally increases with city size, while childhood human capital accumulation falls. These divergent trajectories are associated with differences in both the physical structure of cities and the nature of social interaction therein.