A Dynamic Multivariate Causality Analysis of Energy–Growth Nexus Using ARDL Approach: A Malaysian Energy Policy Perspective

Abstract

This study analyzes a dynamic long-run and short-run causal nexus between energy consumption and economic growth in the presence of capital, labor, and urbanization over the period 1971–2014, in Malaysia. The stationarity issue was tested using augmented Dickey–Fuller (ADF), Kwiatkowski–Phillips–Schmidt–Shin (KPSS), and Ng–Perron tests. However, a dynamic long-run co-integration relation between variables was checked through the ARDL technique. An unrestricted vector error correction model (MUVECM) was used to estimate the short-run and long-run dynamic relations between the parameters and the Engle–Granger method was used for causality analysis. Results of statistical analysis confirmed that all variables were found to be I (1) except variable labour was I(0) and none of the variables was I(2). Total energy consumption Granger caused GDP in one direction over the period 1992-2010 in the case of Croatia. However, labor and urbanization impacts were mixed. The Granger causality analysis confirmed mixed results in the short run and the long run. Moreover, estimated results confirmed a feedback hypothesis between income and capital was in the short term and the long run. The short-run and long-run causal effects of labor force on economic growth were confirmed. This study provides important insights to policymakers and energy economists. Prudent energy conservation policies and economically improved measures would be of great help. However, demand-side management-based policies would have no adverse effect on the economic performance of Malaysia.