Abstract As developed countries continue to update and strengthen corporate climate risk disclosure regulations, developing countries remain in a wait-and-see phase regarding related actions. However, the lack of legal and financial regulations does not dampen investors’ enthusiasm for integrating climate risk information into their investment and financing decisions. In the face of a severe climate crisis, corporate climate risk disclosure has become a key focus for investors. This study, which draws on textual data from Management Discussion and Analysis (MD&A) reports and corporate social responsibility (CSR) reports of China’s A-share listed companies from 2007 to 2022, uses machine learning and natural language processing to construct a corporate climate risk disclosure index. It examines the impact of corporate climate risk disclosure on stock mispricing from both theoretical and empirical perspectives. Research has revealed that corporate climate risk disclosure reduces stock mispricing, with a more pronounced effect on alleviating stock overvaluation. Further analysis shows that when disclosed climate risk information concerns “acute risk” and “transaction risk,” the mitigating effect on stock mispricing is stronger. The mechanism analysis reveals that corporate climate risk disclosure primarily reduces stock mispricing by decreasing information asymmetry between investors and companies, curbing irrational investor behavior, and reducing analyst opinion divergence. Heterogeneity tests indicate that the effect of climate risk disclosure on mitigating stock mispricing is stronger in companies with dispersed ownership structures, lower strategic differentiation, greater online attention, and greater focus from institutional investors. This study broadens the research field related to climate risk information disclosure and investor financial behavior, offering empirical insights for improving capital market development in emerging markets and supporting rational decision-making by investors under climate risk. Graphical Abstract
Mo et al. (Mon,) studied this question.