We use necessary cookies to make our site work (for example, to manage your session). We’d also like to use some non-essential cookies (including third-party cookies) to help us improve the site. By clicking ‘Accept recommended settings’ on this banner, you accept our use of optional cookies.
Necessary cookies
Analytics cookies
Yes
Yes
Yes
No
Necessary cookies
Necessary cookies enable core functionality on our website such as security, network management, and accessibility. You may disable these by changing your browser settings, but this may affect how the website functions.
Analytics cookies
We use analytics cookies so we can keep track of the number of visitors to various parts of the site and understand how our website is used. For more information on how these cookies work please see our Cookie policy.
Exploring the relationship between credit spreads and default probabilities
Working papers set out research in progress by our staff, with the aim of encouraging comments and debate.
Published on
17 August 2004
Working Paper No. 225
By Mark J Manning
Contrary to theory, recent empirical work suggests that changing default expectations can explain only a fraction of the variability in credit spreads. This paper takes a fresh look at this question, relating credit spreads for a sample of investment-grade bonds issued by UK industrial companies to default probabilities generated by the Bank of England’s Merton model of corporate failure. For the highest quality corporate issues, where the probability of default is low, this factor explains relatively little of the variation in credit spreads. For such bonds, common market factors – perhaps related to liquidity conditions – appear to be of greater importance. This is consistent with previous empirical work. For lower-rated investment-grade bonds, however, the probability of default is found to be a more important determinant of credit spreads, explaining around a third of variability in a pooled regression. When coefficients are allowed to vary at the level of the individual issue, explanatory power rises to 50% for this group. This is much higher than previous studies have found, reflecting both the more direct application of the Merton model and the recognition that idiosyncrasies in factors such as liquidity conditions and expected recovery rates are likely to undermine results from pooled estimation.