Asymmetric effects of low-carbon transition disclosure on stock market pricing efficiency: evidence from a novel machine learning index

Information disclosure plays a crucial role in shaping pricing efficiency, with heightened relevance for firms in the midst of a low-carbon transition. This study examines the impact of firm-level low-carbon transition disclosure (LTD) on stock market pricing efficiency. Using semantic recognition from machine learning models and text analysis, we construct a novel LTD index at the level of listed Chinese companies. Our findings reveal that enhanced LTD significantly improves stock pricing efficiency. Notably, the influence of LTD on pricing efficiency is asymmetric – it is more effective in correcting overvaluation than in addressing undervaluation. Further analysis shows that this positive effect is more pronounced among firms with higher levels of AI adoption and greater investor attention.