Does green finance improve energy security in Chinese Provinces? Evidence from machine learning approaches

As a cornerstone of Sustainable Development Goal 7, green finance (GF) has emerged as a strategic driver of China's transition to a low-carbon energy system. Despite its growing relevance, the provincial-level effects of GF on energy security (ES) remain underexplored. Hence, this study aims to evaluate how GF interacts with ES across Chinese provinces, spanning the period from 2007 to 2022. We hypothesize that GF plays a crucial role in fostering ES by mitigating both energy supply and transition risks. This study employs a novel machine-learning-based panel quantile-quantile Kernel Regularized Least Squares method, along with a quantile random forest (QRF) technique, for enhanced forecasting accuracy. The results reveal that while GF introduces some short-term disruptions, it significantly strengthens provincial ES in the long run. Financial development, industrialization, and GDP are consistently associated with improved energy security across most quantiles, whereas artificial intelligence shows mixed impacts. Risk mechanism analysis confirms that GF reduces both energy supply risk and energy transition risk. Additionally, the QRF outcomes indicate that ES can gain significant benefits, particularly through the implementation of green finance policies in the future. These findings suggest that green finance plays a crucial role in achieving SDG 7 by enhancing energy security across China. Drawing on these findings, the study recommends a comprehensive regulatory framework to address existing challenges and enhance the resilience of China's energy security system through an improved green finance mechanism aligned with the objectives of SDG 7.