Essays on Patterns of International Trade.

This dissertation studies the roles of comparative advantage, monopolistic competition, and firm-level heterogeneity in international trade. The first essay investigates cross-country and cross-industry variation in the fractions of exporters among domestic firms. The paper presents a model of an economy in which countries are asymmetrically endowed with two production factors, industries vary in the relative intensity of these factors, and firms differ in productivity. The model predicts that the shares of exporting firms in the number of domestic producers are ranked in order of the industry’s relative intensity of the factor with which the country is relatively well-endowed. This quasi-Heckscher-Ohlin prediction is empirically tested using data from the manufacturing censuses of Chile, Colombia, India, and the United States. The result shows that the correlation between the exporter fractions and industry skill intensities is larger for a country with higher skilled-labor abundance, giving evidence of the suggested role of comparative advantage in exporter selection. The second essay examines the importance of factor proportions in explaining the number of product varieties in exports of countries. The introduced model of a two-factor, two-country and multi-industry economy with productivity-heterogeneous firms suggests that countries export more varieties in industries in which the countries have a comparative advantage. This theoretical prediction is confirmed by empirical tests that use disaggregated data on the U.S. imports and show that relatively (un)skilled-labor abundant countries tend to export more varieties in more (un)skilled-labor intensive industries. The third essay proposes an alternative test of the monopolistic competition model of international trade that implies a positive correlation between the volume of trade and the similarity among trading countries in economic size. In contrast to the preceding studies testing this implication for aggregate trade, this paper focuses on trade of differentiated products that the model describes more directly. The amended prediction is tested with disaggregated data on manufacturing trade and production, using various estimation methods. The result confirms the predicted relationship between trade and country size similarity for all country groups, but it also demonstrates that for non-rich countries this correlation is driven more strongly by other forces than by horizontal product differentiation.Ph.D.EconomicsUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/61759/1/ikamata_1.pd