Determinants of Non-Traditional Exports in Uganda (2000–2016)

Abstract

This study investigated determinants of Uganda’s non-traditional exports using quarterly data for the period 2000 to 2016. The main approach used was the ordinary least squares to test for the long-run relationship. The error correction model was applied to test for short run relationship while Granger (1987) model was applied to test for the long-run causal relationship between the variables. The variables that were investigated in this study included non-traditional exports, real GDP, foreign direct investment, real effective exchange rates and gross domestic savings. There was presence of a structural change in 2011 when Bank of Uganda enacted the monetary policy framework reform meant to meet the challenges of macroeconomic management generated by the transformation of the economy over the last 10 years. Going further to investigate the source of the structural change, it was concluded that there were differences in slopes for the period before and after the enactment of the 2011 monetary policy framework and necessitated the use of a model with a common intercept term but different slopes (concurrent regression). After the 2011 structural change, empirical results show that real GDP (p-value = 0.000) and foreign direct investments (p-value =0.000) positively and significantly affected Uganda’s non-traditional exports. For the period before the structural break, all study variables were positive and significant at 5 percent level that is real GDP (p-value = 0.000), foreign direct investments (p-value =0.000), real effective exchange rates (p-value =0.012) and gross domestic savings (p-value = 0.000). Results from the short-run model revealed that the Error Correction Term in the model was significant at 5 percent level and had a correct sign (negative). The Error Correction Term coefficient (= -0.085 and -0.426 for the period before and after the structural break respectively) implied that in each quarter, non-traditional exports adjusted by 8.500 and 42.600 percentage points between the current level and the long run equilibrium level. It was also revealed that non-traditional exports granger-caused real GDP, hence the direction of the relationship between non-traditional exports and real GDP ran from non-traditional exports to real GDP implying that variations in non-traditional exports are preceded by variations in real GDP but not vice versa. The study therefore recommends allocations of real GDP to primarily focus on productive sectors (agriculture) of the economy that boast non-traditional export growth in the country.