Abstract:
African countries have over the years experienced persistent current account deficits. The role of asymmetries in explaining the response of trade balance to exchange rate movement has not received adequate attention as linear models dominate extant empirical literature. In this paper, we examined the impact of exchange rate on the trade balance in five African countries using both linear and nonlinear autoregressive distributed lag models to analyze data for the period 1980-2018. The linear model revealed that the J-curve holds in Uganda in the short run, whereas evidence of long-run J-curve effect was found only in Algeria. However, the nonlinear analysis showed that the short-run J-curve holds for South Africa and Uganda whereas a long-run J-curve effect was found in Algeria and Uganda. The results make a case for modeling asymmetries as the nonlinear model performed relatively better. An important policy implication is the need to address structural imbalances in the economy to leverage on the exchange rate and trade policies to improve trade outcomes.