Does Methodology Matter? Revisiting the Energy-growth Nexus in Asia Pacific Economies

Abstract

There have been many studies on the energy-growth nexus since the 1970s. However, the findings are mixed. Reasons for the inconsistent findings include different time spans and countries or regions examined, different frameworks, as well as different methodologies adopted, by various studies. In this paper, we examine how methodology affects findings regarding the energy-growth nexus employing a sample of Asia Pacific economies for the period of 1965–2019, using a supply-side framework that takes into consideration both physical capital and human capital in the production function. Methodologies considered in this paper include the widely used vector error correction model and autoregressive distributed lag model (ARDL) for time series and panel data, and Granger non- causality tests. Coefficients are estimated and compared using various methodologies including the cointegrating approach, Dynamic OLS, Fully- Modified OLS, and others. The conclusion obtained from this study will contribute to a better understanding of the varying findings in the literature and highlight the importance of choosing an appropriate methodology for the data analysis.