How strong is the causal relationship between globalization and energy consumption in developed economies? A country-specific time-series and panel analysis

ABSTRACT

We examine the causal relationship between globalization, economic growth and energy consumption for 25 developed economies using both time series and panel data techniques for the period 1970–2014. Due to the presence of cross-sectional dependence in the panel (countries from Asia, North America, Western Europe and Oceania), we employ the cross-sectional augmented IPS test to ascertain unit root properties. The cointegration test results indicate the presence of a long-run association between globalization, economic growth and energy consumption. Long-run heterogeneous panel elasticities are estimated through the common correlated effects mean group estimator and the augmented mean group estimator. The empirical results reveal that, for most countries, globalization increases energy consumption. In the USA and UK, globalization is negatively correlated with energy consumption. The causality analysis indicates the presence of the globalization-driven energy consumption hypothesis. This empirical analysis suggests insightful policy guidelines for policy makers using globalization as an economic tool to utilize energy efficiently for sustainable economic development in the long run.

KEYWORDS:

• Globalization
• energy consumption
• causality
• developed economies

JEL CLASSIFICATION:

• Q43
• F49

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Data sourced for the World Bank World Development Indicators database (http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators).

² https://www3.epa.gov/climatechange/ghgemissions/global.html.

³ The null hypothesis is that slope coefficients (no heterogeneity) are homogenous against no homogeneity (heterogeneity).

⁴ There is no need to test the presence or absence of cointegration between the variables when applying the Toda and Yamamoto (1995) causality test. This methodology uses a two-step approach. In the first step, the lag length of the VAR is determined using standard selection criteria like AIC or SIC. In the second step, the order of the VAR lag length is augmented by a number reflecting the order of integration associated with the variables.

⁵ We are thankful to the anonymous referee for his suggestion to recheck our bivariate models, using globalization and energy consumption, for the omitted variable bias. Accordingly, we added economic growth in our energy demand function as a control variable and find that our results are robust and are not misspecified. To conserve space, we report and focus on the empirical linkages between globalization and energy consumption. The rest of the results are available upon request from the authors.