Trade balances:do exchange rates matter?

Abstract

The concern about the persistence of the US trade deficit has generated a lively debate over its causes and the role of exchange rates in restoring external balance. While some authors argue that the traditional adjustment process linking the behavior of the trade balance to movements in the real exchange rate and the domestic and foreign income levels has worked, recent research has found no empirical evidence in support of a stable long-run relationship between these variables, In this paper I use the approach in Johansen (1988,1991) to test for coiontegration between US bilateral trade balances, exchange rates and incomes during the floating period. I also apply the testing procedure developed by Hansen (1992) to detect potential structural breaks in a cointegrating relalionship in the bilatelal trade balance equations. The empirical results indicate the following: (1) the presence of stable long-run trade balance equations in all six models, (2) no significant evidence of structural breaks, (3) the nominal exchange rate has better explanatory power than the real rate, (4) the Marshall-Lerner condition is supported by the data, (5) the exchange rate is weakly exogenous in the trade balance relationship and, (6) the traditional belief that trade balances adjust slowly to exogenous shocks Is confinned; however, the speed of adjustment varies significantly across bilateral models.