The asymmetric effect of renewable and non-renewable energy on carbon emissions in OECD: new evidence from non-linear panel ARDL model

Abstract

The level of economic income, population density and sources of energy supply is critical in assessing environmental quality. Recent empirical studies paid limited attention to the role of renewable (RE) and fossil energy (NRE) supply in carbon pollution regarding the Environmental Kuznets Hypothesis (EKC). Therefore, this study investigates the asymmetric relationships between carbon emissions and energy sources on the one hand and the environmental Kuznets hypothesis on the other hand for OECD countries, comprising countries with significant renewable energy supplies. The study includes the annual data from 1990 to 2021 and performs panel non-linear ARDL regression. The empirical results clearly show that RE and NRE have asymmetric effects on emissions in the long run but not in the short run. Both positive and negative shocks in RE reduce CO2 emissions in OECD economies, while asymmetric shocks in NRE substantially increase them. Increasing RE supply is clearly effective in reducing emissions. However, unlike most previous studies, this study shows that RE does not significantly reduce CO2 emissions in OECD countries. The error correction term (ect.) in the NARDL model is negative and significant. The magnitude of the term indicates that the system will return to long-term equilibrium about 4.2 years after any shock. Furthermore, we show that the EKC Hypothesis is supported in OECD countries. The turning point of the EKC is at $4085.77 per capita. Besides, regression with Driscoll-Kraay standard errors and Augmented Mean Group (AMG) estimator approach were used for robustness checks. The findings from the robustness check are consistent with the NARDL findings. Policies based on the promotion of a low-carbon and sustainable green environment should place greater emphasis on renewable resources even in OECD countries. Moreover, while many studies in the literature address asymmetric effects and EKC as energy consumption or utilisation, the novelty of this study is that it approaches the issue regarding energy supply with asymmetric effects for RE and NRE.