Export-led growth hypothesis: panel cointegration evidence from East Africa

Abstract

Economic growth continues to drive policy discussions across Africa and most developing countries. Different approaches have been employed including those that follow the market-led and export-led growth theories. Expansion of export base is expected to have significant impact on economic growth (Feder, 1983; Krueger, 1990). The study therefore examined the export-led growth (ELG) hypothesis for five East African countries of Uganda, Kenya, Rwanda, Tanzania and Burundi using panel co-integration and multivariate Granger causality tests. Also examined in the study was the impact of terms of trade volatility on exports and output. The results showed strong support for both long and short-run relationship between exports and outputs for all the East African countries. Feedback effects between exports and GDP are reported for Tanzania and Burundi while Uganda, Kenya and Rwanda reports unidirectional causality from exports to output. Overall the study supports Export-led growth hypothesis for all the five countries in the study. Using the Dynamic Ordinary Least Square (DOLS), the study shows that Terms of Trade Volatility has a Negative and Significant Effect on Exports in the long-run. It has also been shown using Vector Error Correction Model that Terms of Trade Volatility has a Negative and Significant Effect on both output and Exports in the short-run. These results supported the export-led growth hypothesis for the panel and for all individual countries in East Africa. Thus, to stimulate economic growth, policies that promote expansion of exports need to be emphasized; coupled with supportive fiscal and monetary policies xii