ESG Engagement and Stock Volatility: The Moderating Role of Risk Disclosure Tone from Textual Analysis of Egyptian Annual Reports

 This study investigates the impact of Environmental, Social, and Governance (ESG) engagement on stock return volatility, and whether the risk disclosure tone in Egyptian mandatory corporate filings moderate this relationship. While prior research suggests that ESG stabilizes firm performance, this study challenges that assumption by investigating whether ESG uniformly reduces volatility or if its effects depend on risk communication strategies. Using publicly listed Egyptian firms, we employ fixed-effects panel regressions, Propensity Score Matching (PSM), standard deviation (SD)-based volatility, and Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models to assess static and dynamic volatility effects. Risk Tone is measured using textual analysis of Egyptian annual reports based on the Loughran-McDonald dictionary. Our findings indicate that before accounting for Risk Tone disclosure, ESG engagement is linked to higher stock volatility, where ESG initiatives could be perceived as resource-intensive and uncertain. Pillar-level analysis reveals that Environmental initiatives heighten volatility, Governance mitigates risk over the long term but has a weaker short-term effect, and social responsibility stabilizes short-term fluctuations in GARCH analyses. However, when controlling for Risk Tone and its interaction with ESG, the relationship shifts. ESG’s volatility-reducing effect strengthens, suggesting that firms with clear, measured risk disclosures experience lower volatility. In contrast, negative or uncertainty-laden risk disclosures overshadow ESG’s stabilizing benefits, amplifying stock fluctuations. These findings are important for investors, regulators, and managers by considering the firm’s ESG engagement and their risk disclosure tone, as investor responses are not only influenced by ESG performance, but also how companies communicate risks in mandatory reports.