Within the geographically mobile U.S. labor force, highly educated workers are much more likely than others to migrate over long distances. Theoretical explanations for this disparity abound and generally hinge on assumptions about the
relative costs and benefits of moving, but empirical tests of the theories are essentially non-existent. I test competing explanations for the disparity by exploiting their conflicting implications for conditions under which workers successfully
arbitrage spatial differences in wages. To isolate relatively pure arbitrage opportunities, I confine the analysis to
responses to unanticipated changes in local labor demand at the time a worker first entered the labor market and control
flexibly for underlying trends in migration and wages.
I find that young college graduates are two to five times more likely than less educated workers to reside in a state with
high labor demand at the time they entered the market. Among college graduates, cross-state migration equalizes the wage impact of early career labor demand shocks in their home states. This is not true for less educated workers. The
lack of wage convergence is most severe for cohorts who entered the labor market during periods of high spatial variation in state conditions and low national employment growth. My results are consistent with a theory of educational differences in internal migration that assumes less educated workers are credit constrained, and cast doubt on several other explanations for the difference. The results imply that modest relocation subsidies could significantly
reduce the spatial mismatch between less educated workers and labor demand.