The Stationary Spatial Equilibrium with Migration Costs

We extend the standard spatial equilibrium framework in that we model how utilitymaximizing migration decisions lead to spatial arbitrage, allowing us to solve for rather than assume a long-run spatial equilibrium. To this end, we develop a quantitative spatial model in which heterogeneous workers make location decisions facing region-group-specific labour-market-related agglomeration benefits, region-specific housing-market-related agglomeration costs, and group-specific bilateral migration costs. The model remains tractable and amenable to empirical analysis because of stochastic amenity shocks with group-specific variance. We estimate the structural parameters of the model exploiting comprehensive German labour and housing market micro data and exogenous variation that originates from trade shocks, Germany’s division, and deep history. Our quantitative framework can be used to evaluate the aggregate and distributional effects of arbitrary spatial shocks to productivity, amenity, or housing supply in general equilibrium, with and without migration costs. Using trade shocks, land use regulations, and regional transfers as cases in point, we illustrate how spatial shocks lead to winning and losing workers within winning and losing regions and place-based policies affect the spatial distribution of worker welfare.

Joint work: Fabian Bald, Duncan Roth and Tobias Seidel

Preliminary draft: http://personal.lse.ac.uk/ahlfeldg/WP/GA_FB_DR_TS_-_SEMC.pdf

Key words: Migration, gravity, amenity, productivity, spatial equilibrium, welfare
JEL: R38, R52, R58