Land Acquisition and Sectoral Composition: Evidence from India

Abstract
In many emerging economies, compulsory acquisition has become a critical instrument for industrial policy. In theory, compulsory acquisition might increase development in settings where private land acquisition is associated with substantial transaction costs. Developing countries are generally characterised by fragmented land ownership, such that land assembly entails negotiating with a large number of landowners -- a time consuming and costly process. With compulsory acquisition, the government takes on the burden of these negotiation costs, such that more productive projects can be realized, eventually fostering economic growth. This paper studies whether compulsory acquisition actually stimulates industrial development, both in terms of entry and employment, in the context of Special Economic Zones (SEZs) in India. I exploit an unexpected reform which increased the cost of compulsory land acquisition, and compare the effects on SEZs across Indian states depending on their take-up of this instrument, arguing that states with specific compulsory acquisition policies are more impacted by the reform than states where expropriation was not officially used. In the first analysis, I study the effect of restricting compulsory acquisition on entry in a difference-in-differences design. I find that this increase in land acquisition costs changes the industrial composition of SEZs, with the share of manufacturing decreasing by almost 40 percent. Using novel data on SEZ proposals, I show that this effect is mostly driven by lower intentions to entry, with no significant differences in actual SEZ operations. Secondly, I study the effect of restricting compulsory acquisition on entrant quality. I compare how pre- and post-reform SEZs affect local non-agricultural employment using a spatial difference-in-differences design. I find that manufacturing SEZs after the reform are associated with significantly higher local employment than their older counterparts. These findings suggests that ``too much'' compulsory acquisition can act as a subsidy for less productive developers.