Cities and Labor Market Dynamics

Abstract

People live and work in local markets spatially distinct from one another, yet space is absent from most economic models of the national labor market. Workers choose the markets in which they will participate, but there are costs to mobility. Furthermore, cities are heterogeneous in a number of dimensions, including their local labor market productivity, their housing supply, and their offering of amenities.I examine the impact of these spatial considerations on the dynamics of local labor markets and the national market to which they aggregate. First I study the patterns of location choice through a gravity model of migration applied to rich panel data from the U.S. I find that location choices respond to temporal shocks to the labor market, but only after controlling for local heterogeneity. Next, with this result as motivation, I turn to development of a dynamic spatial equilibrium of the national labor market. I make a technical contribution to work in dynamic equilibrium modeling by empirically implementing an island economy model of worker mobility. I quantify the importance of worker mobility costs versus local housing prices for explaining spatial variation in the unemployment rate. I find that the link between the local housing market and the local labor market is important for explaining the spatial dispersion in unemployment, but mobility costs are not. Finally, I further exploit the dynamic equilibrium framework to examine the effect of local housing policy on labor market growth. I find that housing supply regulation is a constraint to growth, but is only binding on cities that are particularly desirable because of their labor market opportunities or amenities. I find that some lightly regulated markets have a contingent of population that has been pushed out of more regulated markets by high housing prices.