Document Type


Publication Date



City living in the United States often requires paying high taxes and receiving low-quality services, partly as a result of the large number of poor residents in cities. This is the first paper to measure how much this feature of American local government affects individuals’ residential location choices. I do this by taking advantage of a natural experiment: the dramatic increase in state financing of local governments in some states’ poor cities, often due to court-ordered school finance equalization. I use the universe of over 20,000 Census-designated places to compare population growth of cities between states with large amount of redistribution to poor cities and states with little redistribution. The key threats to accurate measurement are that poor places may have grown differently from rich places in the absence of redistribution, and places in high-redistribution states may have grown differently from places in low-redistribution states. To address these concerns, I use the “differences-in-differences” econometric technique. The results show that redistribution had a large effect on population growth between 1980 and 2010, helping explain the “return to the city” in recent decades. I then do a case study on the local finances of Connecticut, which shows that the state transfers for education were used not only to increase education spending as intended, but also reduce tax rates. Since this paper shows that redistribution to poor areas mitigates the distortion to individuals’ location choices resulting from poor cities’ high tax rates and poor services, it suggests that further place-based redistribution may be not only equitable but also efficient.