Artificial Intelligence and Green Transformation: Human Capital Upgrading, Green Finance, and ESG Assessment as Drivers of Sustainable Productivity

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Offers a comprehensive empirical framework linking AI adoption, green finance, ESG, and productivity, with DiD and mediation analysis for policy evaluation.

🏢実務担当者:Chinese listed firms can use findings to prioritize AI adoption and green finance to enhance sustainable productivity, especially for SMEs.

🏛政策担当者:The negative productivity impact of Green Credit Guidelines on polluting industries suggests careful policy design to balance environmental and economic goals.

This study investigates the interrelationship between the adoption of artificial intelligence (AI), green finance, and Environmental, Social, and Governance (ESG) performance in relation to sustainable productivity within Chinese publicly listed companies. Utilizing comprehensive firm-level panel data that encompasses 1,000 listed firms from 2012 to 2024, amounting to 13,000 firm-year observations sourced from the CSMAR database, corporate disclosures, and reputable ESG rating agencies, the study applies pooled OLS and fixed-effects models, a difference-in-differences (DiD) framework, and mediation analyses to fill existing gaps in the comprehension of sustainable transformation mechanisms. The findings reveal a positive correlation between AI adoption and sustainable productivity (β = 0.115, p < 0.01), as well as with the intensity of green finance (β = 11.16, p < 0.05 in fixed effects). However, the ESG effects seem to indicate cross-sectional selection rather than improvements within firms. Mediation analysis indicates that green finance and ESG collectively account for approximately 18% of AI’s overall association (5.9% and 12.0%, respectively), with the remaining 82% functioning through direct operational channels. The 2012 Green Credit Guidelines are associated with a relative decrease in measured productivity among polluting industries (DiD: −16.48, p < 0.01), aligning with the policy’s restrictions on ‘Two-High’ sectors. Heterogeneity analysis shows that small firms benefit disproportionately from AI adoption (β = 0.131 compared to β = 0.112 for larger firms), challenging traditional beliefs regarding the technology-adoption benefits of large organizations. Key limitations include potential endogeneity, dependence on text-based and rating-based proxies, and the focus on a single-country context and this is why the future studies should consider employing instrumental-variable approaches, alternative metrics, and cross-country analyses. This study enhances the understanding of sustainable transformation as a synergistic process that integrates technological, financial, and governance aspects.

• semanticscholar https://doi.org/10.15291/oec.5026first seen 2026-06-24 05:28:57