Credit Risk and Risk Neutral Default Probabilities: Information About Rating Migrations and Defaults

38 Pages Posted: 28 Jul 2003

See all articles by Gordon Delianedis

Gordon Delianedis

University of California, Los Angeles (UCLA) - Anderson School of Management

Robert L. Geske

University of California, Los Angeles (UCLA) - Finance Area

Date Written: February 2003

Abstract

This paper is a study of the properties and uses of option based estimates of risk neutral default probabilities (RNDP). We argue that because RNDP are the pricing probabilities, their changes should possess the information embedded in price changes. In this paper we provide the first evidence on the distribution of these important pricing probabilities. We compute RNDP's using the diffusion models of Merton (1974) and Geske (1977). While the simpler Merton model is estimated for the assumed debt structure of one pure discount bond, the Geske model is estimated for multiple default options, and boundary conditions implying either debt or equity refinancing. The refinancing condition in the Geske model is also shown to directly imply a term structure of default probabilities. First we demonstrate that the RNDP's estimated from both Merton and Geske models successfully categorize firms by actual credit rating, they will generally serve as upper bounds to risk adjusted default probabilities, and they are easier to estimate than risk adjusted default probabilities. We also show that RNDP's have the same sensitivities as risk adjusted default probabilities. Furthermore, we argue that the time series of RNDP's allows easy estimation of default probability correlations, while actual rare event default correlations are very difficult to estimate. Next, in order to further examine their information content, an event study using RNDP's and rating migrations is conducted. We are the first to show that the RNDP's estimated from both the Merton and Geske models do possess significant early information about credit rating migrations. Both models are able to distinguish, for example, which BBB's are more likely to migrate either up or down. Furthermore, the slope of the term structure of default probabilities from the Geske model is shown to contain information about impending migrations to actual default. Finally, the term structure of risk neutral default probabilities can be useful for pricing forward debt securities of different maturities such as delayed start credit default swaps.

Suggested Citation

Delianedis, Gordon and Geske, Robert L., Credit Risk and Risk Neutral Default Probabilities: Information About Rating Migrations and Defaults (February 2003). Available at SSRN: https://ssrn.com/abstract=424301 or http://dx.doi.org/10.2139/ssrn.424301

Gordon Delianedis

University of California, Los Angeles (UCLA) - Anderson School of Management

110 Westwood Plaza
Los Angeles, CA 90095-1481
United States

Robert L. Geske (Contact Author)

University of California, Los Angeles (UCLA) - Finance Area ( email )

University of California UCLA)-Financial Economics
Los Angeles, CA 90095-1481
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
2,413
Abstract Views
9,739
rank
8,202
PlumX Metrics