Date of Award

August 2016

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Economics

First Advisor

Mohsen Bahmani-Oskooee

Committee Members

Niloy Bose, Kundan Kishor, Rebecca Neumann

Keywords

International Trade, J-Curve, Nonlinear Ardl, Trade Balance

Abstract

ABSTRACT

EXCHANGE RATE CHANGES AND THE TRADE BALANCE: IS THE LINK SYMMETRIC OR ASYMETRIC

By

Hadiseh Fariditavana

The University of Wisconsin-Milwaukee, June 2016

Under the Supervision of Professor Mohsen Bahmani-Oskooee

This dissertation consists of three essays in international trade. The J-Curve theory suggests that after currency depreciation, the trade balance continues to deterioration till some lags emerge, and then starts to improve. My contribution is in using a non-linear Autoregressive Distributed Lag model to examine if the effects of depreciation are different than the effects of appreciation of exchange rate on the trade balance. Using the sample data from thirteen developed and developing countries I show that when aggregate trade data are used, the effects of those two are asymmetric. In response to changes in the real exchange rate, a country’s trade balance could improve with respect to one trade partner and could deteriorate with respect to another trade partner. Testing the J-Curve using aggregate trade data might not capture both effects at the same time. Thus in section two of chapter four, using the non-linear ARDL model, the bilateral J-Curve phenomenon between two specific trade partners is tested. I use the bilateral trade data between the United States and its sixteen major trade partners and I find support for my claim that the effects of exchange rate changes are asymmetric. Section three of chapter four takes it one step further in a way that employs the trade data of 162 trading industries between the United States and Canada to investigate the asymmetric claim. By further disaggregating bilateral trade data my results show that in majority of the cases in my sample, the effects of depreciation are significantly different than the effects of appreciation. Due to the possible positive response of one bilateral commodity flow to the exchange rate changes and possible negative response of another flow at the same time, the commodity level trade data is considered to be able to solve any possible aggregation bias of the other types of data sets.

Included in

Economics Commons

Share

COinS