Benchmarking Model of Default Probabilities of Listed Companies

This paper presents a benchmarking model for validation of default probabilities of listed companies for Basel II purposes. The model is based on the recent studies on the predictive capability of structural credit risk models. Benchmark ratings and one-year default probabilities are assigned to companies by mapping the term structures of default probabilities of the companies generated by a structural model based on stochastic leverage ratios to the term structures of default rates reported by rating agencies. The empirical results show that the benchmarking model have adequate discriminatory power of ranking credit risk. The association between the benchmark ratings and external credit ratings is statistically significant. Benchmark default probabilities obtained from the model could thus be used as external and independent estimates for comparisons with banks' internal default probability estimates. Significant deviations from this benchmark provide a reason to review the internal estimates.