Financial stability and the transition: integrating environmental and social factors into a solvency model for SMEs

Policymakers seek to promote a sustainable economy through responsible investments while ensuring financial stability and a resilient banking sector. The financial sector is key in supporting sustainable firms, but uncertainty remains regarding their financial resilience and ability to repay debt due to the costs of sustainability efforts. This study builds a solvency prediction model that combines traditional financial ratios with both environmental and social sustainability metrics, employing Logistic Regression and Machine Learning techniques. To address missing data on sustainability indicators for insolvent firms, we develop a customized multiple imputation procedure. We perform an out-of-sample evaluation to assess model performance and apply a Monte Carlo framework to test the robustness of the results. Our findings reveal that incorporating sustainability indicators improves the model’s ability to distinguish solvent and insolvent firms and increases the likelihood of a firm being solvent. Sustainability data gives banks a valuable predictive edge in solvency assessments, indicating that financial stability and broader sustainability goals can be aligned rather than treated as competing objectives.