Annual report textual complexity and ESG rating disagreement

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

🔬研究者:This study offers a novel information-processing lens to understand ESG rating disagreement, complementing agency-method focused literature.

🏢実務担当者:Corporate disclosure teams can use these findings to craft clearer annual reports, potentially reducing ESG rating divergence and improving investor communication.

🏛政策担当者:Regulators should consider disclosure complexity when designing ESG reporting guidelines, as it affects rating consistency across agencies.

With the global rise of environmental, social, and governance (ESG) evaluation, corporate ESG rating disagreement has attracted increasing attention from academics and practitioners. Existing studies primarily explain ESG rating disagreement from the perspective of methodological differences among rating agencies, but pay relatively limited attention to how the complexity of input information shapes the extent to which such methodological differences are reflected in rating outcomes. Drawing on information processing theory, this study uses Chinese A-share listed companies from 2013 to 2022 as the research sample to systematically examine the relationship between annual report textual complexity and ESG rating disagreement, as well as the boundary conditions in the information input stage and information interpretation stage. The results show that: (1) Annual report textual complexity is significantly positively associated with ESG rating disagreement. More textually complex annual reports are associated with higher information interpretation costs and greater subjective information completion among rating agencies, which may be related to wider rating disagreement. (2) The relationship between annual report textual complexity and ESG rating disagreement exhibits heterogeneity across ESG dimensions and rating agency types. Specifically, this association is more pronounced in the social and governance dimensions and among domestic rating agencies. This study reveals that ESG rating disagreement stems not only from methodological differences among rating agencies but also from the way these differences are amplified under complex and ambiguous information environments. (3) Earnings conference call tone and institutional investor site visits serve as boundary conditions in the information input and information interpretation stages, respectively. The positive relationship between annual report textual complexity and ESG rating disagreement is less pronounced when earnings conference call tone is more positive and when institutional investor site visits are more frequent. The findings provide a new information-processing perspective for understanding ESG rating disagreement and offer empirical evidence for regulators to develop differentiated disclosure guidelines and for listed companies to improve the interpretability of annual report disclosures.

• openalex https://doi.org/10.1016/j.iref.2026.105609first seen 2026-07-13 05:32:15