ESG, firm systematic risk, and economic policy uncertainty: analysis using deep learning-based ESG scores

Abstract This study explores the complex relationship between Environmental, Social, and Governance (ESG) performance, firm-level systematic risk, and economic policy uncertainty using data from 645 Vietnamese publicly listed firms from 2019 to 2023. A deep learning model is applied to compute ESG scores by analyzing data from annual reports disclosures, hence, reducing human subjectivity, enhancing cross-language scalability, and bridging the ESG data gap for emerging markets. The research reveals a U-shaped relationship between ESG and risk, where firms with moderate ESG involvement face heightened risk due to increased costs and volatility. However, when ESG efforts reach a certain threshold, the long-term benefits outweigh the costs, helping firms better manage risks associated with economic policy uncertainty. Furthermore, this nonlinear relationship is more pronounced during periods of high Economic Policy Uncertainty (EPU). The findings suggest that sustained and meaningful ESG initiatives can serve as a risk mitigation tool, particularly in environments with fluctuating economic policies. The findings offer valuable insights into ESG strategy and risk management in emerging markets like Vietnam.