Credit Spread Bounds and Their Implications for Credit Risk Modeling
44 Pages Posted: 22 Jul 2001
Date Written: June 28, 2001
A basic requirement for a credit risk model is that it should not imply negative default probabilities. In this paper, we explore the implications of this condition for credit risk modeling. More specifically, we use the condition as a diagnostic tool to investigate if a particular model is consistent with a given set of credit spreads. We show that under this condition, each model has two credit spread boundaries which can be calculated analytically, and the model is correctly specified if and only if the observed credit spread curve lies within the two boundaries. These analytical formulas for the boundaries also allow us to derive some general properties of a large class of credit risk models.
Our study also adds to the literature on pricing defaultable claims off the default probability curve, a method widely used in practice. It is well-known that negative default probabilities frequently occur in constructions of default probability curves. Our analysis provides one possible explanation of why this problem happens and also suggests how the problem may be solved (or at least alleviated).
Keywords: Credit risk modeling, Credit spread bounds
JEL Classification: G0, G1
Suggested Citation: Suggested Citation