Energy intensity and its differences across China’s regions: Combining econometric and decomposition analysis

Abstract

To achieve the energy intensity reduction targets set by the Chinese government policymakers need to understand the key drivers that contribute to regional variations in energy intensity. Understanding regional differences will contribute to the design of more effective energy policies. To facilitate this understanding we estimate a penalized panel quantile regression model that accounts for unobserved individual heterogeneity and distributional heterogeneity across the regions of China. The effects of economic growth, urbanization, foreign direct investment, energy structure, and industrialization, on energy intensity differ across quantiles. The effects of economic growth and foreign direct investment on energy intensity are negative and significant at every quantile. A 1% increase in foreign direct investment decreases energy intensity along the entire conditional distribution, ranging from 7.5% at 10th quantile to 3.7% at 90th quantile. The effects of urbanization and industrialization on energy intensity are positive and significant at every quantile. Moreover, a 1% increase in industrialization lifts energy intensity by 54% at 10th quantile and 33% per cent at 90th quantile. The results from a Shapley decomposition model further show that economic growth is the most prominent factor that contributes to energy intensity differences, following by industrialization, foreign direct investment and energy structure.