Ahmad Lashkaripour
Published: 2014
Total Pages:
Get eBook
Each of the three chapters in this dissertation is based on an empirical research paper. The collective goals of these, rather independent, papers is to present an alternative and unifying theory of international specialization. A theory suitable for analyzing international trade at various levels and between a broad set of countries.The first chapter develops a new theory of international specialization that tractably com- bines all aspects of North-North and North-South trade into one model. The new theory also provides an alternative explanation for many other well-established facts, most notably the "Washington apples" effect. The theory builds upon, and retains the central elements of Krug- man [1980]. In the new framework, North-North trade is governed by national product dif- ferentiation. North-South trade is governed by a new channel of across product specialization that has been overlooked in the literature. Specifically, there are many products and each prod- uct comes in different varieties. Products differ in how (horizontally) differentiated they are. Monopolistically competitive firms charge a higher markup for varieties of highly differenti- ated products. In equilibrium, rich countries specialize in highly differentiated--high markup products, while poor countries specialize in less differentiated--low markup products. To quan- tify the gains from trade, I estimate the structural parameters of the model using disaggregated data. Incorporating the new channel of across-product specialization into the Krugman model magnifies the gains from opening to trade by around 200%. Despite trading less, low-income countries experience the largest gains from trade liberalization.The second chapters provides the first empirical confirmation of the iceberg trade cost as- sumption. The assumption is embodied in all major models of International trade. However, empirical evidence to support this rather conventional assumption is lacking. This paper pro- vides such evidence by developing a simple model of international transportation. The model links shipping cost to the f.o.b. price of the shipment, and demonstrates that shipping cost per count is more iceberg-like than shipping cost per kilogram -- existing studies have generallylooked at shipping cost per kilogram for goods that are measured primarily in counts (e.g. TVs, cars). To address this finding, I first calculate price and shipping cost on a per-count basis for goods that report count as the primary unit of measurement in US import data. Then, I estimate the dependence of shipping costs on f.o.b. price. Estimation results strongly support the iceberg specification. Specifically, for every 1% increase in f.o.b. price (per count), the ship- ping cost (per count) increases by 0.91%. The paper then estimates the "Washington apples" effect: the dependance of export f.o.b prices on shipping costs. The effect is estimated to be stronger in industries where shipping costs are more iceberg-like. This suggests that, contrary to common belief, per-unit trade costs cannot be the only driving force behind the "Washing- ton apples" effect. The paper then proceeds to find strong empirical support for an alternative force.The third chapter provides a simple framework to analyze the three main components of international trade flows: (i) the number of goods traded, (ii) the quantity of each good that is shipped, and (iii) the prices they are sold for. While gravity equations are massively suc- cessful in explaining the overall value of trade, they do not provide much insight about the decomposition of trade. In this paper I develop a novel framework that provides, consistent with data, predictions about not only the value of trade but the composition of trade values. I relax the conventional assumption that consumers are identical, and allow for demand hetero- geneity across consumers. I also allow for quality heterogeneity across varieties. The model explains the effect of distance and per capita income on trade along the intensive margin, the extensive margin and the price margin--all of which are well-documented in the empirical literature. It also provides a novel theoretical foundation for the higher price of tradables in developed countries. To further asses the model, I evaluate two predictions, of the model, re- garding the price of traded goods and one prediction regarding the extensive margin of trade. The exercise confirms that all three predictions are borne out in the data. The model provides a framework to investigate the (across-consumer) distributional effects of trade liberalization. I show that, despite the aggregate gains, the poorest consumers experience losses in face of trade liberalization.