Conditional dynamic forecast of electrical energy consumption requirements in Ghana by 2020: A comparison of ARDL and PAM

The frequent power outages which characterises Ghana's electricity sector have raised serious concerns among policy makers and stakeholders. The causes of this problem have been cited as growing demand and inadequate investment. The objective of this study, therefore, was to identify the factors that affect aggregate electricity demand in Ghana both in the short and long-run as a guide for demand-side management based on two econometric approaches-ARDL and PAM. Also predictions of aggregate electricity demand based on these econometric methods were obtained to guide investment decisions in the electricity sector. The result revealed that the positive output, urbanisation, and income effects more than offset the negative efficiency effects leading to growth in electricity consumption both in the short-run and long-run. The result further establishes the superiority of the ARDL approach to forecasting over the Partial adjustment approach to forecasting. For all three scenarios, future domestic consumption levels of electrical energy are predicted to increase ad infinitum from 2012 to 2020. Specifically electricity demand is estimated to fall within 20,453 and 34,867 GWh by the year 2020 for all three scenarios. This requires huge capital investments in additional plant capacity.

► ARDL and PAM are used to forecast electricity demand.
► Income is the predominant factor in explaining electricity demand.
► Industry efficiency, output, and urbanization play secondary role.
► ARDL has a better forecasting performance than Partial Adjustment Model.
► Electricity demand is estimated to fall within 20,453 and 34,867 GWh by 2020.