ESG論争が企業財務パフォーマンスに与える影響：米国上場企業からの証拠

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

🔬研究者:ESG論争と財務パフォーマンスの因果的関連を固定効果モデルで検証し、情報環境の調整効果を示した実証的貢献。

🏢実務担当者:ESG論争がROA低下につながるリスクを認識し、ネガティブ情報の適切な開示とアナリスト対応の重要性を再確認できる。

🏛政策担当者:ESG論争の財務影響を踏まえ、開示規制や情報環境整備の政策設計に示唆を得られる。

This study examines the association between environmental, social and governance (ESG) controversies and the subsequent financial performance of firms listed in the United States and whether the information environment of the firm conditions this association. Most of the literature measures the sustainability conduct of the firm through its own disclosure and the ratings built from that disclosure, while negative events recorded by external media sources have received less attention. Using a panel of 14,210 firm-year observations for 1737 NASDAQ and NYSE firms over 2014 to 2025 from the LSEG Refinitiv database and a two-way fixed effects model with standard errors clustered by firm, the study finds that a firm that recorded an ESG controversy in the previous year reports a return on assets lower by about 0.50 percentage points and a lower Tobin’s Q relative to the same firm in years without a controversy. The return on assets result is the more consistent of the two across the alternative specifications, while the Tobin’s Q result is supportive but more sensitive to the specification and to the composition of the sample. The pre-event lead coefficient is not significant at the 5 percent level for the return on assets, and analyst coverage does not respond to a controversy, which speaks against a simple reverse-causality reading. The evidence is consistent with heterogeneity in the association with the return on assets, which is concentrated among firms followed by few financial analysts and is not detected among firms in the highest tercile of analyst coverage, whereas no such heterogeneity is found for Tobin’s Q. The results are robust to sector-by-year fixed effects, to the exclusion of the pandemic years, to lagged controls, and to alternative treatment definitions, and they are reported in full together with the specifications in which the estimates lose precision. The principal limitation is that a fixed effects design with observational data documents within-firm associations and cannot by itself establish causal effects.

• semanticscholar https://doi.org/10.3390/su18199901first seen 2026-10-01 05:23:21 · last seen 2026-10-02 05:22:29