Social Sharing
Extended Viewer
On the White Board - October 2008
Oct 15, 2008
CDS and CreditGrades
Back in 2005, when we were lucky enough to have only a handful of firms in distress, we explored (Better Ingredients) the notion of incorporating implied volatility into CreditGrades, our equity-to-credit model. The model is a simple implementation of the Merton framework, wherein credit and equity are jointly considered as options on a firm's assets. The model takes a firm's equity price and leverage, along with an estimate of future equity volatility, and outputs an indication of where credit should be pricing. Comparing the model credit spread (what we call the CreditGrade) to the actual price of credit in the market gives us some sense of relative value between equity and credit, as well as a sense of which market is first to recognize trouble ahead.
A significant challenge in implementing the model lies in estimating the equity volatility. Traditionally, we have used a very long-dated, and therefore stable, estimate of historical volatility. If 2005 was any lesson though, the CreditGrades model reacts much more quickly to extreme market events when the historical volatility is replaced by an implied volatility from the equity options market. We have recently examined the behavior of the model on two firms in distress, AIG and Washington Mutual. In both cases, the standard CreditGrade model lagged the actual spread widening in the summer of 2007, while the addition of implied volatility resulted in a close anticipation of the spread widening then and at the end of 2007. Since that time, implied volatilities have generally been elevated, and for both firms, the actual spread spent most of 2008 ranging somewhere between the standard CreditGrade and the enhanced version. In the case of AIG, the enhanced CreditGrade seemed to lead the CDS spread, while with Washington Mutual, the CDS and CreditGrades were mostly in synch as the credit deteriorated through 2008.