A closer look at the probability of default taking into account the current regulatory considerations

Abstract Credit risk is one of the most significant risks for banks, with the probability of default (PD) playing a central role in risk management. Logistic regression models are widely used to estimate PDs based on historical default data and firm characteristics. However, such models rely on restrictive assumptions that may abstract from complex structures in firm-specific risk drivers, such as multimodality in the distribution of the risk. This paper introduces a scenario-based, forward-looking framework for firm-level risk assessment that builds on corporate planning models and Monte Carlo simulation. By simulating alternative business plan scenarios and aggregating firm-specific risks, the approach generates a scenario-implied likelihood of default (LD) under explicitly stated assumptions. Using an illustrative case study of a medium-sized company, we demonstrate how the proposed framework can be used to explore risk figures. The analysis shows that scenario-implied LD can provide complementary insights to standard PD models, particularly for internal risk assessment and scenario analysis. Rather than replacing outcome-based PD estimation or supervisory validation procedures, the proposed framework is positioned as a transparent, firm-specific tool that supports internal risk management and forward-looking decision-making.