The Performance of Structural Models in Pricing Credit Spreads

Abstract

Although structural models have been used as a standard in credit risk modelling for the last thirty years, there is a lack of consensus in the literature regarding their performance. This paper tests the performance of three structural models, namely the Merton (1974) model, the Collin-Dufresne and Goldstein (2001) stationary leverage ratio model and the CreditGrades™ (2002) model against the credit default swap spread using data from 2003 to 2010.

We find that, consistent with existing literature, while the structural models have significant explanatory power for the cross-section of CDS spreads, all three models underpredict the observed credit spreads. The underprediction of credit spreads is not driven by any ratings categories or sector. We also show that while the three models produce spreads that are highly correlated with each other, they carry information incremental to each other. However, our evidence demonstrates that other firm specific variables: CDS liquidity, and credit rating can explain a significant part of observed credit spreads. Finally, our results indicate that the observed underpricing is driven by actual spreads reflecting additional information such as liquidity and residual credit risk, and in fact, the actual spreads are lower than implied spreads once additional credit risk and CDS liquidity information is taken into consideration.