ESG確信對企業漂綠行為影響之實證研究

Abstract

近年來，隨著投資人、監管機關與利害關係人對 ESG 資訊需求增加，企業永續揭露已成為資本市場評估企業風險、治理品質與長期價值的重要依據。然而，ESG 揭露程度提高並不必然代表實際 ESG 表現同步改善，企業可能透過大量揭露塑造正面永續形象，卻未能以相應績效支撐揭露內容，進而形成漂綠問題。第三方 ESG 確信被視為提升永續資訊可信度的重要機制，但其降低漂綠之效果仍有待檢驗。 本研究以 2018 年至 2023 年美國上市公司為研究對象，建構公司年度縱橫資料，探討第三方 ESG 確信與企業漂綠程度之關聯，並進一步分析確信提供者類型與確信層級之差異效果。本研究以 Bloomberg ESG Disclosure Score 衡量 ESG 揭露程度，並以 LSEG Workspace ESG Performance Score 衡量實際 ESG 績效，透過兩者之相對差距建構企業相對漂綠程度。實證分析分別檢驗確信有無、確信提供者類型與確信層級，並採用 CSDID 與多期多處置組 DID 進行穩健性檢查。 研究結果顯示，企業取得第三方 ESG 確信本身並未與較低漂綠程度呈現穩定關聯，審計專業提供者相較於非審計專業提供者亦未呈現更顯著的漂綠抑制效果。相較之下，合理確信相較於有限確信，更可能與較低企業漂綠程度相關，且穩健性檢查亦支持此一結果。 整體而言，本研究指出，ESG 確信不應僅被視為企業是否取得外部背書之二元問題，而應進一步檢視其確信品質與程序強度。若確信僅停留於有限程序或形式性安排，其降低漂綠之效果可能有限；相反地，合理確信因具備較高查核強度與證據蒐集要求，較可能發揮降低 ESG 揭露與實際績效落差之治理功能。

In recent years, the growing demand for ESG information from investors, regulators, and stakeholders has made corporate sustainability disclosure an important basis for assessing firm risk, governance quality, and long-term value in capital markets. However, a higher level of ESG disclosure does not necessarily indicate a corresponding improvement in actual ESG performance. Firms may use extensive disclosure to construct a positive sustainability image without supporting such disclosure with corresponding performance, thereby creating greenwashing concerns. Third-party ESG assurance is regarded as an important mechanism for enhancing the credibility of sustainability information, but its effectiveness in reducing greenwashing remains to be examined. This study examines U.S. listed companies from 2018 to 2023 and constructs a firm-year panel dataset to investigate the relationship between third-party ESG assurance and corporate greenwashing. It further analyzes the differential effects of assurance provider type and assurance level. This study uses Bloomberg ESG Disclosure Score to measure ESG disclosure and LSEG Workspace ESG Performance Score to measure actual ESG performance. Corporate relative greenwashing is constructed based on the relative gap between the two measures. The empirical analysis examines the presence of assurance, assurance provider type, and assurance level, and further conducts robustness checks using CSDID and multi-period and multi-treatment difference-in-differences model. The results show that obtaining third-party ESG assurance is not consistently associated with lower greenwashing. Audit professional assurance providers also do not exhibit a stronger greenwashing-mitigation effect than non-audit professional providers. In contrast, reasonable assurance is more likely to be associated with lower corporate greenwashing than limited assurance, and the robustness checks support this finding. Overall, this study indicates that ESG assurance should not be viewed merely as a binary issue of whether a firm obtains external endorsement. Instead, greater attention should be paid to assurance quality and procedural rigor. If assurance remains limited in procedures or primarily formal in nature, its effect in reducing greenwashing may be limited. By contrast, reasonable assurance, with higher verification intensity and stronger evidence-gathering requirements, is more likely to function as a governance mechanism that reduces the gap between ESG disclosure and actual ESG performance