Revisiting the Income Elasticity of Energy Consumption : A Heterogeneous, Common Factor, Dynamic OECD & non-OECD Country Panel Analysis

Estimating the relationship between economic development and energy demand and determining whether that relationship changes as levels of development change have been popular questions in energy economics. The current paper contributes to the literature by assembling a wide panel dataset of energy consumption and prices for 37 OECD and 41 non-OECD countries—a particularly large dataset considering the inclusion of country-specific energy prices. The unbalanced data spans 1960-2016, with the full 56 years of data for 17 countries and all countries having at least 18 years. In addition, our dynamic panel estimates address nonstationarity, heterogeneity, and cross-sectional dependence. Most results suggest that the GDP elasticity is less than unity—i.e., energy intensity will fall with economic growth; however, that difference from unity is not always significant at the 95% level. A “grand” average (average of the many panel averages) suggests a GDP elasticity around 0.7. Most evidence suggests that the GDP elasticity is similar for OECD and non-OECD countries, and for non-OECD countries, similar across income-bands. Also, there is no evidence that individual country elasticity estimates (for GDP or prices) vary systematically according to income. The price elasticity is larger (in absolute terms) for OECD than for non-OECD countries—indeed, it is typically insignificant for non-OECD countries