Modelling the volatility dynamics of China's regional carbon markets: The heterogeneous effects of the fossil and clean energy electricity generation

This paper models the volatility dynamics of China's regional carbon markets and examines the impact of the electricity structure on China's regional carbon markets based on the Auto-Regressive and Moving Average-Generalized Autoregressive Conditional Heteroscedasticity with Mixed Data Sampling (ARMA-GARCH-MIDAS) model. Particularly, how the electricity structure affects carbon market volatility is investigated using total electricity generation and decomposed electricity generation as low-frequency variables in the model, respectively. Subsequently, the varying responses of volatilities in China's three emission trading scheme pilots to both total and decomposed electricity generation are empirically analyzed. The results reveal significant and negative impacts of thermal and renewable electricity generation growth on China's carbon market volatility, with the most pronounced effect observed in Guangdong, followed by Hubei and Shenzhen; however, the impact of renewable electricity generation seems to be relatively weak. Additionally, heterogeneous effects of the decomposed electricity generation on market-specific volatilities are unveiled, with certain components exhibiting positive, negative, or no effects. Finally, the introduction of the electricity structure variable can enhance the accuracy of estimating carbon allowance price volatility, and avoid a potential biased measure of volatility.