Determinants of trade and specialization: A quantitative analysis

This dissertation studies the effects of technology, factor endowments, tastes, and transport costs on international trade and specialization. The analysis is performed using two methods: estimation and simulation, both using data from OECD countries. First, the effects of technology and factor endowments on specialization, measured by industry shares in GDP, are jointly estimated. The methodology allows for variable capital utilization and sector-specific capital. Capital utilization is calculated from the electricity consumption, and the efficiency of electricity consumption is calculated for each industry and country. The results show that ignoring variable capital utilization leads to approximately 30--40% overestimation of the effects of technology on specialization. It is also found that once variable capital utilization is accounted for, different assumptions about mobility of capital have little effect on the technology coefficients. Second, a computable model of trade is developed in order to account for the volume, composition, and pattern of international trade. The model incorporates technological, factor endowment, and transport cost differences at the industry level. An important feature of the model is that it relies on within-industry heterogeneity of producers in order to explain intra-industry trade. The model is parametrized and evaluated. It is found that the model accurately predicts trade flows and sizes of intra-industry trade. It is also found that the effects of capital on specialization predicted by the model are close to those obtained by the regression analysis. The model is then used to identify the effects of main determinants of trade on the volume and composition of trade. The results show that, given current trade barriers, within-industry heterogeneity accounts for 93% of the volume of trade, industry-level comparative advantage accounts for 5%, while factor-endowment differences have a negligible effect on the volume of trade. A labor-only version of the model is used to study the consequences of trade wars and trade liberalization. It is found that targeting different industries with these policies produces very different results with respect to employment and welfare.