Time-varying analysis of aggregate electricity demand in Ghana: a rolling analysis

Using the rolling regression technique, the study investigated how the effects of income, economic structure and industry efficiency on aggregate electricity demand vary with time. The Quandt-Andrews test was used to examine the presence or otherwise of structural breaks in the data set. The long-run estimate of aggregate electricity demand based on fully modified ordinary least squares reveals that the variables carried their a priori expectations with the positive income effect playing the most significant role followed by industry efficiency with structural effect playing the least role. The result of the Quandt-Andrews test showed rejection of the null hypothesis of no structural breaks with themaximumstatistic occurring in 1987, which is themostlikelybreakpoint location. Therolling regression result revealed evidence of constant shifting of poles of the estimated demand model. The plot of the evolution of demand elasticities showed variations over time with some evidence of strong linkages with periods of policy regime changes. Changing economic structure, improving energy efficiency and integrating electricity production, distribution and transmission into the overall macroeconomic planning will go a longway to stabilise the situation in the electricity market.