Direct, spillover and welfare effects of regional firm subsidies

We analyze the effects of a large place-based policy, subsidizing up to 50% of investment costs of manufacturing firms in East Germany after reunification. We show that a 1-percentage-point decrease in the subsidy rate leads to a 1% decrease in manufacturing employment. We document important spillovers for untreated sectors in treated counties, untreated counties connected via trade and local taxes, whereas we do not find spillovers on counties in the same local labor market. We show that the policy is at least as efficient as cash transfers to the unemployed, but is more effective in curbing regional inequality.