Stressed Distance to Default and Default Risk

Distance to default (DTD) is a strong predictor of default risk derived from structural models. This paper specifies a stressed version of DTD ("stressed DTD'') to measure time-varying corporate default risk in the event that a systematic stress scenario occurs. Compared with the ordinary DTD, probability-of-default estimates from the Campbell, Hilscher and Szilagyi (2008) model, and a systematic default risk measure ("failure beta") proposed by Hilscher and Wilson (2017), the stressed DTD is more capable in predicting corporate defaults during the 2007--2009 crisis. The stressed DTD can also explain a substantial portion of variation in credit ratings after controlling for raw default probability and failure beta, which is consistent with the understanding that credit stability under stress should be a key attribute of ratings. Finally, we show that the stressed DTD can explain variation in CDS spreads, and the common variation in stressed DTDs can explain variation in senior CDO tranche spreads, providing supporting evidence that investors demand compensation for exposure to stressed default risk