Research

Fragmented land ownership and ill-defined property rights are key obstacles to large-scale industrial development in emerging economies. To facilitate private investment, governments increasingly rely on compulsory land acquisition policies. I study the effect of this practice on industrial composition and local employment, exploiting an unexpected reform that prohibited compulsory acquisition for Indian Special Economic Zones (SEZs). I pin down the effect on entry by constructing a novel dataset on the universe of SEZ proposals before and after the reform. First, I find that the share of manufacturing SEZ proposals, and the share of developed manufacturing SEZs, decreases by almost 50 percent. This effect is most pronounced for (a) more land-intensive industries and (b) areas with higher land fragmentation. Second, I study how restricting compulsory acquisition affects employment in and around SEZs, and find that manufacturing SEZs after the reform are associated with significantly higher local employment than SEZs before the reform. Despite the reduction in entry, post-reform manufacturing SEZs contributed four times more to aggregate employment than their pre-reform counterparts.

This paper develops a general equilibrium model of trade intermediation with search frictions and firm heterogeneity. Firms decide whether to export, and if so, whether to use an intermediary. Both firms and intermediaries need to invest in market penetration to reach foreign consumers, but the intermediary can transport multiple goods through the same advertising network. A firm that opts for intermediation saves the cost of market penetration but loses out on demand since the intermediary charges a higher price. The corresponding lost profit increases more quickly in productivity than the optimal investment in advertising. Therefore, the equilibrium exhibits productivity sorting in the export mode: the most productive firms export directly, the least productive do not export and the moderately productive firms use an intermediary. In line with empirical evidence, I find that the resulting trade network is characterized by negative assortative matching both in terms of number of connections and size. Moreover, intermediaries make up a higher share of exports when goods are more homogeneous and when the costs of exporting are higher. Finally, the model predicts that sales of an intermediated firm are less responsive to trade liberalization than direct exporters. Trade liberalization reduces the diversity of intermediated products but increases the number of intermediated firms.

We use detailed historical data on India's domestic infrastructure to show how high domestic transport costs have conditioned the local labor market consequences of its drastic import tariff liberalization in the early 1990s. We find that districts located at larger distances from the country's main international gateways are better shielded from increased foreign import competition: non-agricultural employment falls less than in otherwise similarly exposed districts closer to one of the country's main ports. They, however, also benefit less from improved access to foreign intermediates: non-agricultural employment increases less than in districts with a similar input-output structure, but located closer to one of the country's main ports. Furthermore, we find substantial heterogeneity in these effects across the firm size distribution in district closer to India's main ports: the negative employment response to increased import competition is concentrated at small to medium firms, whereas employment at medium and large firms benefits most from better access to foreign intermediates.