Semester

Spring

Date of Graduation

2004

Document Type

Dissertation

Degree Type

PhD

College

Chambers College of Business and Economics

Department

Economics

Committee Chair

William N. Trumbull.

Abstract

The study of empirical trade estimation has been a staple in international economics literature since the early 1960s. I focus on three current topics in empirical trade estimation. In chapter 1, I use a time-series approach and in the context of the effect of exchange-rate volatility on trade, provide both theoretical and empirical explanations for positive volatility effects. My results show a clear pattern; when the effect of exchange-rate uncertainty on trade volume is positive [negative] for importers, the effect for exporters is negative [positive]. In addition, the U.K. provides evidence that trade balance is an important factor in determining the effect of volatility. If changes in the sign of trade balance are not taken into account, the effect of volatility may go undetected. Finally, the sensitivity of imports and exports to exchange-rate volatility is affected similarly by changes in market conditions; trade is more sensitive to volatility when the trade terms are expected to improve. In chapter 2, I turn to cross-sectional analysis and following a growing trend in the literature, I examine the impact of adding nonstandard variables, ones that measure cultural or ideological differences, to the basic gravity model. In the context of United States trade with Latin America, I show that in the case of Brazil, Nicaragua, and El Salvador, as the number of foreign-born peoples in the United States increases, U.S. exports to their country of origin increases. This finding sheds important light on the fact that the determinants of trade may go beyond standard economic and descriptive variables. In Chapter 3 I compare the Hausman-Taylor method for estimating the unrealized US-Cuban trade potential to the OLS, fixed-effects, and random-effects methods using the out-of-sample approach. The Hausman Taylor method is ideal because it allows for the inclusion of time-invariant variables in trade projections and circumvents the problem of an ad hoc estimation of the country-specific dummy variable needed for a projection based on the fixed-effects estimator. In addition, it removes the correlation between the error term and included variables which often plagues random-effects estimation.

Download

Share

COinS