Dynamic impacts of energy consumption on economic growth in China: Evidence from a non-parametric panel data model

To empirically gauge the efficacy of energy policies, we propose a non-parametric method to investigate the relationship between economic growth and energy consumption from both time and space perspectives. Specifically, we rely on the local linear dummy variable estimation (LLDVE) method to explore the time-varying province-specific trends, the common trend, and the coefficients based on panel data from 26 provinces in China from 1995 to 2017. We find that the promotion effect of energy consumption on economic growth changes over time, as evidenced by the inverted U shape of the relationship. Moreover, the non-parametric model captures such an effect better than the parametric model. With the dual goals of sustainable economic growth and carbon emissions reduction in mind, we classify the sample according to the degree of carbon intensity, which indicates that energy efficiency should be improved in high-carbon development areas, while more attention should be paid to investment and innovations in low-carbon development areas.

Automated summary

By using a non-parametric method, the relationship between economic growth and energy consumption is empirically examined from both time and space perspectives. The study finds that the promotion effect of energy consumption on economic growth changes over time, and the non-parametric model captures this effect better. Additionally, classification based on carbon intensity suggests that energy efficiency should be improved in high-carbon development areas, while investment and innovation should be emphasized in low-carbon development areas.