The role of Environmental, Social, and Governance (ESG) in predicting bank financial distress

Abstract

We analyze the predictive power of Environmental, Social, and Governance (ESG) indicators to forecast bank financial distress using a sample of 362 commercial banks headquartered in the US and EU-28 members states from 2012 to 2019. Our results demonstrate that ESG improves the predictive capability of our model to correctly identify distress. Notably, ESG strongly reduces the likelihood of misclassifying distressed/defaulted banks as healthy. Our model, which we estimate using six alternative approaches, including traditional statistical techniques, machine learning approaches, and ensemble methods, has implications for both practical implications by banking sector supervisors, as well as literature on default prediction.