Globalisation, economic growth and energy consumption in the BRICS region: The importance of asymmetries

Globalization may have both favorable and unfavorable effects on energy consumption demand, which can be analyzed in three modes: scale effect, technique effect, and composition effect. Recent econometric literature (Shin et al., 2014) highlights that factors such as structural reforms, policy shifts, real and financial shocks, and regional and global imbalances may affect the variables under consideration and, hence, induce asymmetries in their dynamic relationships. Thus, it does not seem unreasonable that these factors may have induced changes in the type of relationship across the relevant variables. Additionally, given that asymmetry and non-linearity are two important stylized facts of many economic time series, a non-linear model that characterizes short-run and long-run linkages between globalization, economic growth and energy consumption is deemed appropriate. Practically, assuming a strictly linear relationship in the presence of significant asymmetries can lead to inefficient and biased results, which invalidate the usefulness of the linear specification. Notably, both economic and financial development depend mainly on macroeconomic factors (e.g., business cycles, monetary policy adjustments, and product market regulations), while energy consumption seems to be more sensitive to specific conditions in the domestic and global energy markets.

This paper contributes to the existing literature by examining the asymmetric relationship between globalization, economic growth and energy consumption using data from Brazil, Russia, India, China and South Africa (BRICS) for the 1970-2015 period. Existing energy literature confirms that none of the studies so far have used any comprehensive measure of globalization that encompasses the economic, social and political dimensions in BRICS. In doing so, we apply augmented Dickey-Fuller (ADF) (Kim and Perron, 2009) unit root test that contains information single structural break in the series. Further, we apply the multivariate non-linear autoregressive distributed lags (NARDL) framework proposed by Shin et al. (2014) to examine asymmetric dynamics between the variables. The data source for all variables except globalization is the world development indicators (CD-ROM, 2016). The globalization index is borrowed from Dreher (2006). Our empirical results report the presence of asymmetric cointegration across the variables for the case of the BRICS economies. Positive globalization shocks decrease (increase) energy consumption in Brazil, Russia and India (China), whereas negative globalization shocks decrease (increase) energy consumption in Russia, India and China (South Africa). Overall, energy consumption declines with globalization in the case of Brazil, Russia, India and China. Moreover, positive and negative shocks in economic growth and capital increase energy consumption in BRICS region.