This thesis consists of three essays on spatial labor markets and public policies. I study successively the interactions of space with job search, demography and housing policy. In the first essay, I develop a framework to study theoretically and quantitatively the welfare attributes of spatial mismatch, defined as a misalignment between where job seekers reside and suitable employment opportunities. In a quantitative urban model with frictional labor markets, the structure of the city interacts with labor markets because commuting is costly and information about job offers decays with distance. The decentralized equilibrium might feature too much or too little spatial mismatch, depending on whether commuting costs or information decay dominates. When commuting costs prevail, the constrained-efficient allocation may be restored by a mix of moving-to-opportunity and enterprise zone interventions that bring jobs and workers together. The second essay, joint with David Autor, studies the relationship between population age and population density in the United States. We document the inversion of the rural-urban age gradient between 1950 and 2019. Whereas in 1950, residents in the least dense counties were on average 4.5 years younger than their counterparts in the most dense counties, by 2019 residents of the most rural counties were 2.7 years older than those in the most urban counties, a swing of 7.2 years. We show that sharp temporal changes in age-specific migration rates were the predominant contributor to this reversal. In the third essay, Hector Blanco and I examine the distributional implications of the shift from public housing to subsidized private housing initiated by the U.S. government over the past few decades. We build a quantitative urban framework where housing assistance complements income taxation to redistribute across workers. We argue that provision of affordable housing involves a trade-off between indirect pecuniary redistribution and direct amenity spin-offs. On the one hand, public housing drives local rents down, while amplifying the spatial concentration of poverty. On the other hand, project- and tenant-based rental assistance enhances the local amenities of subsidized households by promoting mixed-income communities, but pushes private landowners’ rents up.Ph.D.