Abstract:

This paper examines the influence of ESG ratings in credit risk assessment. The role of sustainability reporting is growing within the European Union regulatory framework, so the paper explores ESG not only as a reporting outcome, but as a potential signal of firm-level risk and institutional adjustment. The research uses firm-level data from the Croatian Chamber of Economy (HGK) for 2024–2025, and combines machine-learning techniques (random forest prediction to assess out-of-sample performance) with traditional econometric approaches. It observes ESG as an aggregate score and through its Environmental, Social, and Governance components. Credit risk is measured using HGK creditworthiness indicators. The results point to a weak and unstable relationship between ESG measures and credit ratings. Within-firm changes in ESG scores do not systematically translate into changes in credit ratings over the observed period. From a predictive perspective, ESG variables have a modest influence on forecasting accuracy, and decomposing ESG into components does not significantly improve model performance. Overall, the findings show that ESG ratings reflect structural firm characteristics and reporting capacity, instead of a strong standalone indicator of credit risk. The reason for this might be an early-stage adjustment process to EU sustainability regulation, due to the fact that ESG metrics have not yet been fully integrated into credit risk assessment practices.