In this study, we estimate a cointegrating vector with U.S. trade balance with Brazil as the dependent variable, Brazil’s real GDP, U.S. real GDP, and U.S. dollar-real exchange rate – defined as the number of Brazilian currency, real, needed to purchase one U.S. dollar – as the independent variables with all the variables measured in natural log form over the data ranging from 1994 to 2016. The estimated value of the coefficient associated with the exchange rate variable turned out to be greater than 1 in absolute value satisfying the Marshall-Lerner condition. The satisfaction of the Marshall-Lerner condition led us to conclude that an increase in U.S. dollar-real exchange rate (i.e. an appreciation of U.S. dollar or a depreciation of Brazilian real) will lower the U.S. trade balance with Brazil or equivalently will improve Brazil’s trade balance with the U.S. It also means that if the imposition of a tariff on the import of steel and aluminum from Brazil, as announced by President Trump, lowers the supply of U.S. dollar and raises the value of U.S. dollar in Brazil, then the tariff may actually worsen the overall U.S. trade balance with Brazil.
Marshall-Lerner condition, net export, exchange rate elasticity of export, exchange rate elasticity of import, exchange rate, unit root, cointegration