Causal relationship between macro-economic indicators and coffee exports in Uganda: A timeseries analysis: 2001-2011

Abstract

The objective of this study was to ascertain the causal relationship between volume of coffee exports and other macroeconomic variables (real effective exchange rates and international coffee prices) from 2001 to 2011, using the Vector Auto Regressive Framework (VAR) and Granger (1969) approach. The trend of the series was observed by means of graphs which demonstrated an upward and downward trend hence not being stationary. The ADF tests indicated that the macroeconomic indicators and coffee exports were not stationary at levels but became stationary after taking first differences. Empirical tests using the chow (1960) test indicated presence of a structural break which was caused by the global financial crisis. The best optimal lag length of order one was established among the series. The Engel and Granger approach indicated presence of a long run equilibrium relationship between the macroeconomic indicators and volume of coffee exports. Results of the VAR model indicated that the best relationship was when volume of coffee exports was explained by international coffee prices and real effective exchange rates (Adjusted R2=0.661). Granger (1969) and Engel and Granger approach demonstrated presence of a bi-directional causality. The VAR residual serial correlation LM test indicated that the error terms were not auto correlated.

The bi-directional causality between macroeconomic indicators and volume of coffee exports, called for policies that make Uganda’s coffee export volumes competitive on the international market such as maintaining an appropriate real effective exchange rate and increasing Public and Private Investments especially for those involved in production of coffee.