This dissertation describes two essays in empirical international trade, focusing on trade costs and the pattern of trade along the intensive and extensive margins. In the first essay, I study the barriers that impede international trade. In the second paper, I examine the growth of U.S. agricultural trade in detail describing how U.S. agriculture and food trade has expanded along the margins. The first chapter introduces a relatively straightforward, yet empirically powerful, manipulation of the gravity equation. The gravity model has been dubbed the work horse model of empirical trade, and thus is a suitable foundation from which to derive an indirect measure of largely unobservable 'iceberg' trade costs. In this paper, I solve a sector level version of the gravity equation and study the pattern of agricultural trade costs and factors that impede world agricultural trade growth over a long time series, 1986-2011. In addition, I estimate sector-specific elasticity of substitution which is a key parameter in the computation of trade cost. In the second essay, I examine the growth of world and U.S. agricultural exports along the intensive and extensive margins of international trade over the period 1986 to 2010. The purpose of this essay is to decompose the growth of world and U.S. agricultural trade using qualitative methods from the marketing literature (i.e., market expansion grids) but modified to fit bilateral trade relationships and a theoretical index to measure the margins of trade at a single point in time. In addition, we examine often overlooked channels by which U.S. agricultural exports have expanded using very detailed agricultural product lines. Using information related to the pattern of a trade rather than trade volume itself, I estimate how much starting a trade relationship with a new partner or in a new product variety matters to agricultural trade growth and then conclude with a set of stylized facts to inform current theory. / Ph. D.

This dissertation investigates the effects of changes in technologies and trade-related policies on income inequality. The first chapter shows that an advancement in labor saving technologies, known as automation, raises the agglomeration of economic activity in large cities and increases wage inequality across regions. I show novel stylized facts about the relationship between city size and the routineness of tasks performed by workers. I develop a general equilibrium model of a spatial economy where automation affects the type of tasks performed by workers and is related to a firm's choice of production location. The model generates several predictions that are consistent with stylized facts and existing empirical evidence: larger cities have greater agglomerations of firms and grow larger when firms can automate more tasks in the production process. The model predicts that an increase in automation raises wage dispersion between larger and smaller cities. A 20% rise in automation increases wages in the top decile of largest cities by about 8% and lowers wages in smaller cities by about 2-8% and hence widens the wage gap by about 10 to 16 %. The second chapter investigates the effect of exchange rate volatility on the intensive and extensive margin of trade, and on income inequality within a country. It finds that the greater volatility in exchange rates lowers trade margins and income inequality. I derive testable predictions regarding the impact of exchange rate volatility on trade margins at the firm level and on income distribution at the industry level. I empirically test these predictions using firm-level microdata. Empirical results provide clear support in favor of the model's predictions about the effects of volatility on trade margins. Finally, in the third chapter, my coauthors and I investigate the effect of Bangladesh’s graduation from Least Developed Country (LDC) status on the price of insulin, an essential medicine for diabetes, and on households’ welfare and poverty. We find that upon Bangladesh’s graduation from LDC status, the price of insulin could rise as much as 11 times the current price for patented insulin if an unregulated monopoly is allowed. This would significantly reduce welfare and increase the incidence of poverty for households with members suffering from diabetes.