Long-run Asymmetric Association between Imports and Economic Growth in Turkey

Abstract

This study aims to analyze asymmetric association between economic growth rate and growth rate of imported goods and services in Turkey for the periods of 1988-2019 by using the Nonlinear Autoregressive Distributed Lag (NARDL) model and two distinct indicators of economic growth rate. According to the estimation results, there exists a nonlinear cointegration relationship between economic growth rate and growth rate of imports. NARDL (2,2) model is estimated to determine the presence of any asymmetric link in short-run/long-run. We identified an asymmetric relationship between the series in the long-run but not in the short-run. Estimation findings of the first model disclose that a rise in growth rate of imports by 1% causes a rise in economic growth rate by 0.606% while a decrease in growth rate of imports by 1% leads to a drop in economic growth rate by 0.565%. On the other hand, according to the estimation results of the second model, a jump in growth rate of imports by 1% causes to an increase in economic growth rate by 0.574% whereas a drop in growth rate of imports by 1% leads to a decrease in economic growth rate by 0.535%. Also model diagnostic tests imply that the model do not contain autocorrelation, heteroscedasticity, model misspecification and non-normality problems.