International trade and regional inequality

This paper studies how openness to trade can increase inequality across regions and the spatial concentration of economic activity due to the endogenous sorting of heterogeneous firms and industries across space. I embed selection into exporting and trade due to endowment-driven comparative advantage into a multi-sector economic geography model with heterogeneous firms. The model yields two novel mechanisms through which opening to trade increases the spatial concentration of economic activity, one at the firm level and another at the industry level. Firstly, firms in larger cities are more productive and will expand due to trade-induced within-industry re-allocation. Secondly, sectors located in larger cities are less labour intensive and will expand due to trade-induced across-industry reallocation according to comparative advantage in capital-rich andvanced economies. I test the model predictions using the rise in Chinese import competition vis-Ã -vis the United States. I find that on average a sector in a large commuting zone (at the 75th percentile of the city size distribution) loses 1.07 percentage points of employment due to the increase in Chinese import competition while the average sector in a small commuting zone (25th percentile) loses 2.09 percentage points, a loss almost twice as large. Decomposing the effect of trade on cities of different sizes I find that 53% of this differential impact on employment is driven by firm heterogeneity, while 27% is driven by across industry heterogeneity, and the remainder by the positive interaction of both effects. The relative importance of the firm-level channel implies that trade integration, even among similar countries, can substantially increase regional inequality even when there is no reallocation according to comparative advantage.