Impact Of Electricity Consumption On Economic Growth: An Application Of Vector Error Correction Model and Artificial Neural Networks

Electricity consumption is vital for the industry, business, and policy matters on macro-level as well as on micro-level in developing countries having inadequate resources. Furthermore, electrical generation is also one of the dominating issues of less developed countries because of limited resources for low-cost electricity production, inefficient distribution, and hindrances to the implementation of policies for long-run objectives. This research uses time series variables of GDP per capita, electricity consumption, and electricity generation as endogenous variables while shortage as an exogenous variable. This study uses the dataset of the annual interval, starting from 1961 to 2015, gathered from reliable sources. Economic growth related to proxies’ data has been taken from World Bank Indicators 2015 (WBI) while electrical related variables’ data collected from the handouts of Power System Statistics of different years. This study utilizes the Johansen Cointegration (JC) approach to find a long-run relationship along with a Vector Error Correction Model (VECM) methodology to identify equilibrium nexus between electricity consumption and economic growth in Pakistan. Moreover, Artificial Neural Networks (ANN) forecast electricity consumption with a higher degree of accuracy. The JC approach finds the long-run relationship between economic growth, electrical consumption, electrical generation, and electricity shortage. VECM methodology reveals the short-run as well as a long-run nexus among the variables. The deviation from equilibrium swiftly adjusted in this model with no feedback effect of energy consumption-generation. Pairwise Granger Causality method discovers one-way causal relationship running from electricity consumption to economic growth. The findings suggest that electricity consumption and economic growth have a long-run equilibrium relationship, and electricity consumption leads to economic growth in Pakistan. Artificial Neural Networks predict electricity consumption with 99% accuracy, which represents the better-fitted picture of prediction and limited residuals. Furthermore, it also calculates composite coefficient contrast to time series individual coefficient. Policy recommendations of this study are to improve electricity generation through low-cost and environment-friendly options along with discouragement of electricity conservation policy because of a positive and causal relationship between electricity consumption and economic growth. Pakistan may enhance economic growth by minimizing line-losses, construction of dams on a priority basis, and by discouraging petroleum-based electricity.