Electricity Consumption and Economic Activity in Malaysia: Co-integration, Causality and Assessing the Forecasting Ability of the Vector Error Correction Model

The paper empirically investigates the co-integration and causality relationship between electricity consumption (EC) and income based on Malaysia s data over the period 1980-2014 using VECM s framework. In the analysis, the paper used standard econometric approach, namely the unit root tests, the co-integration tests, and Granger causality tests. Meanwhile, to evaluate the forecasting ability, the paper used variance decomposition analysis and standard diagnostic tests for accessing the forecast ability. The paper provides empirical evidence on the existence of short-and long-run unidirectional causality relationship running from economic activity to EC in Malaysia. The results from variance decomposition analysis suggested that economic activity is an important variable in explaining future variation in EC of the country. The forecasting diagnostic tests showed that the VECM has a good forecasting ability to forecast EC for Malaysia. In this respect, the usage of more sophisticated econometric techniques such as the co-integration approach and the VECM should be considered. The existence of unidirectional causality running from economy activity to EC may suggest that the government can implement electricity conservation measures without putting economic development at risk.