Place-Based Redistribution

Abstract

Governments around the world redistribute to distressed areas by conditioning taxes and transfers on location in addition to income. Do the equity gains of place-based redistribution exceed its eciency costs? Working with a model of locational choice and labor supply decisions that nests workhorse specications in urban and public economics, we show that when disadvantaged households are spatially concentrated, transfers from one location to another can yield equity gains that outweigh the eciency costs of distorting location decisions. Eciency costs dominate equity gains, however, when workers are very mobile or subsidized areas are substantially less productive. We provide expressions for the optimal transfer size that depend on the mobility of households, the earnings responses of movers, and sorting patterns. We also provide conditions under which place-based redistribution can improve welfare when place-blind income taxes are set optimally. Place-based redistribution is more likely to improve on income taxes when society favors spatial equity among households with the same earnings levels, motives for which we provide some utilitarian microfoundations. To gauge the plausibility of such social preferences, we conduct a survey querying Americans about the desirability of place-based redistribution between identically poor households. Responses indicate strong support for targeting tax relief to poor households who live in distressed places. A calibration exercise suggests that placebased transfers that optimally lower the eciency cost of taxation may be of the same order of magnitude as found in prominent American \zone" policies.