Computing Reliable Default Probabilities in Turbulent Times

We compare different methods for computing default probabilities using a sample of banks which experienced financial distresses during the 2007-2009 global financial crisis. The traditional KMV-Merton model for firm valuation, credit ratings by rating agencies and a recently proposed zero price probability (ZPP) model are discussed and compared. An empirical application with the acquired or bankrupt banks in the late 2000s financial crisis is presented to show the differences among the three approaches and to discuss their suitability during financial distresses