Working Paper: CEPR ID: DP12977
Authors: Joan Monras
Abstract: Internal migration can respond to local shocks through either changes in in- or out-migration rates. This paper documents that most of the response of internal migration is accounted for by variation in in-migration. I develop and estimate a parsimonious multi-location dynamic model around this fact. I then use the model to evaluate the speed of convergence and long run change in welfare across metropolitan areas given the heterogeneous local incidence of the Great Recession. Results suggest that while there are some lasting effects of the Great Recession across locations, around 60 percent of the initial differences potentially dissipate across space within 10 years. This is true even when locals from the most affected metropolitan areas do not out-migrate in higher proportions in response to local shocks.
Keywords: Internal Migration; Local Labor Market Dynamics
JEL Codes: J61; J20; J30; F22; J43; R23; R58
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| initial differences in welfare across metropolitan areas (H73) | dissipate within 10 years (D15) |
| 1% decrease in wages (J31) | 2 percentage point decrease in net inmigration rates (J11) |
| 1 percentage point increase in unemployment rate (F66) | 3 percentage point decrease in net inmigration (J11) |
| local economic conditions (R11) | net inmigration rates (J61) |
| local shocks (F41) | wages and employment (J31) |