Environmental, social, and governance rating and firm energy efficiency: evidence from double machine learning algorithms in China
摘要
Environmental, social, and governance (ESG) ratings have emerged as a pivotal instrument of informal environmental regulation in advancing sustainable development. This study employs advanced double machine learning (DML) algorithms and utilizes a sample of Chinese A-share listed companies from 2009 to 2024 to examine the effect of ESG ratings on energy efficiency. The empirical results suggest that higher ESG ratings significantly promote energy efficiency, a finding that remains robust after controlling for endogeneity and considering alternative model specifications. Further analysis reveals that this positive influence is more pronounced among older firms, state-owned enterprises, and companies located in non-resource-based cities. Promoting green innovation and optimizing energy consumption structures are crucial pathways through which ESG ratings enhance energy efficiency. ESG ratings can effectively compensate for the institutional deficiencies in formal environmental regulations and play a critical substitutive role in advancing energy efficiency. This study provides fresh insights into the economic implications of ESG ratings and delineates a novel, actionable pathway toward environmental sustainability.