Probability of Default and Credit Spreads in Banks: Examining a Modified Merton Model for Assessing Bank Risk

Sammanfattning: We examine the modified Merton model, as proposed by Nagel and Purnanandam (2019), and its ability to explain bank credit risk by comparing it to the standard Merton model. Previous structural models of default risk build on the assumption that assets follow a log-normal distribution, which is not applicable to banks. The proposed model takes this into account by assuming that banks' assets are mezzanine claims on collateral assets that follow a log-normal distribution. We are able to replicate the modified Merton model. However, our comparison between the two models does not suggest that the modified Merton model is better at explaining credit risk in banks. Additionally, our findings indicate that the modified model does not provide sufficient explanatory power when applied to non-financial companies. We see a need for further empirical validation of the proposed model.