From words to action? Linking ESG reports to environmental performance
言葉から行動へ？ESG報告と環境パフォーマンスの連関 (AI 翻訳)
Ivan Savin, Mateo López Carel, Eva Schlindwein
🤖 gxceed AI 要約
日本語
本研究は、STOXX Europe 600構成企業のESG報告書1,477件を自然言語処理で分析し、34トピック（うち6つが環境関連）を特定。持続可能なバリューチェーンや再生可能エネルギーに関するトピックは環境KPI改善と関連するが、排出量や電気自動車は関連しないことを発見した。特定テーマに着目した開示と実際の環境行動の関係を明らかにした点が貢献。
English
This study applies computational linguistics to 1,477 ESG reports from STOXX Europe 600 companies, identifying 34 topics (six environmental). It finds that topics like sustainable value chains and renewable energy are associated with improvements in environmental KPIs, while others (emissions, EVs) show no significant link. The work advances topic-specific evaluation of ESG disclosure versus actual performance.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本ではSSBJ基準の策定が進み、ESG報告書の実質性が問われている。本論文は報告書で強調するテーマと環境パフォーマンスの関連を検証する手法を提供し、日本の開示実務や投資家対話に示唆を与える。特にグリーンウォッシング懸念への対応として有用。
In the global GX context
As global disclosure frameworks (ISSB, CSRD) emphasize connectivity between reporting and outcomes, this study provides a methodology to test whether specific environmental topics in reports correlate with actual performance. It offers regulators and companies a way to distinguish substantive reporting from rhetoric, relevant for transition finance and anti-greenwashing efforts.
👥 読者別の含意
🔬研究者:This paper demonstrates a topic-level analysis linking ESG text to KPI data, offering a replicable method for future research on disclosure quality.
🏢実務担当者:Corporate sustainability teams can use this framework to assess which environmental themes in their reports are most strongly tied to performance improvements, enabling more targeted disclosure.
🏛政策担当者:Regulators can apply this approach to evaluate whether disclosure requirements (e.g., under ISSB or SSBJ) lead to real environmental action, informing evidence-based policy.
📄 Abstract（原文）
Companies increasingly use ESG reports to communicate their environmental priorities, yet it remains unclear whether the issues they emphasise correspond to measurable improvements in environmental performance. Using computational linguistics, we analyse the full texts of 1,477 ESG reports from companies listed in the STOXX Europe 600 index, identifying 34 main topics—six of which are related to environment. We describe how these topics evolve over time and differ across sectors, cluster companies based on their thematic focus, and test whether specific topics are associated with tangible low-carbon transition efforts. Our findings show that the alignment between environmental communication and environmental performance is topic-dependent. Topics such as sustainable value chains and renewable energy are associated with improvements in environmental KPIs, whereas others, including emissions or electric vehicles, show no significant association. Methodologically, our study advances prior work by combining structural topic modelling with firm-level KPI analysis to evaluate ESG disclosure at the level of specific environmental themes rather than aggregate reporting measures. Our study demonstrates how topic-specific reporting patterns relate to environmental outcomes and offers a more nuanced understanding of when ESG reporting reflects substantive environmental action.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.1371/journal.pone.0350762first seen 2026-06-18 04:50:32
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。