Does exchange rate risk affect exports asymmetrically? Evidence from Turkey

The effect of exchange rate fluctuations on macroeconomic performance is a heavily researched topic. The conventional Mundell-Fleming model clearly depicts the relationship between the exchange rate and the export level. According to the standard model, a depreciation of a country’s exchange rate promotes export; an appreciation of the exchange rate inversely affects export level. The main question is whether the export level of a country react differently to an exchange rate change (appreciation or deprecation). This paper examines the hypothesis of asymmetric effects of the exchange rate on the Turkish bilateral export using monthly data covering the time span from 2003:1 to 2014:12. The asymmetry means that exchange rate affect exports differently during appreciations and depreciations, respectively. Many empirical studies employed the standard methodology for the impulse response estimates in vector autoregressive (VAR) models, regardless of whether the data generating process (DGP) is symmetric or asymmetric. Therefore, this misspecification renders the parameter estimates inconsistent. However, this study uses asymmetric VAR methodology as proposed by Kilian and Vigfusson (Quant Econ 2(3):419-453, 2011) to analyze the asymmetric impact of exchange rate on the Turkish bilateral exports with its major trading countries.This model can be used without knowing the nature of the data generating process. To define the asymmetric effects of real exchange rate (REER) on exports, we consider a 2 variable symmetric VAR(p) model, with Y_t as export and X_t as REER, as in the Kilian and Vigfusson (2011)'s study. The empirical results show that real exchange rates have asymmetric effects on exports for all countries except Germany and the United Arab Emirates at the different periods. Therefore, policy makers need to consider the stability of the exchange rate.