Bivariate Modelling of the Influence of Trade on Aggregate and Disaggregate Energy Use in Ghana

Abstract

The current research has explored the trade-energy nexus for Ghana using annual data within 1970-2011, using the Autoregressive distributed lag model (ARDL) and the Granger predictive causality test in bivariate modelling. The research findings provide evidence of a cointegration association between trade liberalisation (proxied by trade openness) and energy resources (proxied by aggregate energy use, electricity energy use, and fossil fuel energy use). However, there is an insignificant positive long-run effect of trade on energy use for all the energy resources under investigation. In the short-term, there is a positive significant effect of trade on electricity energy use and fossil fuel energy use, but an insignificant positive effect of trade on aggregate energy use. Concerning aggregate energy use, the research findings show that for aggregate energy use, there is causality from trade to aggregate energy use without feedback. However, in the case of electricity use and fossil fuel energy use, trade predicts energy use with feedback. For policy, energy conservation (fossil fuel energy use, and electricity energy use) may not have a deleterious effect on trade in Ghana. Future research may focus on how structural breaks and panel analysis improves the current study, controlling for the effect of other variables that influence energy use.