Author :
Abstract
Grounding the evaluation of Environmental, Social, and Governance (ESG) performance in Stakeholder Theory is increasingly vital, as sustainability practices strengthen firms' long-term value creation. Accordingly, this study examines the impact of ESG performance on key financial indicators for 50 firms listed on the Borsa Istanbul (BIST) Sustainability Index, with data continuity spanning the 2019-2020 period. The relationship between ESG scores and performance variables such as ROA, ROE, market capitalisation, financial leverage, net profit, EBIT, P/B ratio, current ratio, and Tobin’s Q was analysed using the XGBoost algorithm to overcome the nonlinear limitations of traditional econometric models. The findings indicate that ESG practices have a more pronounced effect, particularly on market based indicators (e.g., Market Value and Tobin’s Q). In contrast, their impact on accounting based indicators (e.g., ROA and ROE) remains more limited due to the complexity of internal operational transitions. By bridging the gap between machine learning and sustainability literature, this study provides a strategic roadmap for investors seeking to refine risk assessment through non-financial signals, for corporate managers aiming to boost market valuation via stakeholder-centric strategies, and for regulatory authorities in designing standardised ESG frameworks to enhance transparency and stability in emerging financial markets.
Keywords
Abstract
Grounding the evaluation of Environmental, Social, and Governance (ESG) performance in Stakeholder Theory is increasingly vital, as sustainability practices strengthen firms' long-term value creation. Accordingly, this study examines the impact of ESG performance on key financial indicators for 50 firms listed on the Borsa Istanbul (BIST) Sustainability Index, with data continuity spanning the 2019-2020 period. The relationship between ESG scores and performance variables such as ROA, ROE, market capitalisation, financial leverage, net profit, EBIT, P/B ratio, current ratio, and Tobin’s Q was analysed using the XGBoost algorithm to overcome the nonlinear limitations of traditional econometric models. The findings indicate that ESG practices have a more pronounced effect, particularly on market based indicators (e.g., Market Value and Tobin’s Q). In contrast, their impact on accounting based indicators (e.g., ROA and ROE) remains more limited due to the complexity of internal operational transitions. By bridging the gap between machine learning and sustainability literature, this study provides a strategic roadmap for investors seeking to refine risk assessment through non-financial signals, for corporate managers aiming to boost market valuation via stakeholder-centric strategies, and for regulatory authorities in designing standardised ESG frameworks to enhance transparency and stability in emerging financial markets.
