What Accounts for Time Variation in the Price of Default Risk?∗

We study the market for credit default swaps (CDS) between 2003 and 2008 in order to understand origins of the well documented tendency for credit spreads on diverse issues to periodically undergo large, common adjustments in the same direction and of similar magnitudes. Our methodology allows us to distinguish co-movements that reflect common revisions in the statistical default distribution from common factors driving time variation in the market price of default risk. We estimate the risk neutral default distribution using a latent variable model which assumes that defaults on a name follow a jump process where the log intensity of arrivals of defaults itself follows an Ornstein-Uhlenbeck process. Estimates of this model are used to find the implied times series of the risk neutral default intensity for each firm. A principal components analysis suggests that a very high fraction of time variations in the implied default intensities of diverse firms is explained by a single common factor. We then combine these estimates with estimates of the statistical default process based on a hazard model in order compute the implied market price of default risk. We show that a relatively high fraction of the observed variation of this market price of default risk can be accounted for by a linear model of the market price of default risk using as observed covariates macro indicators, firm indicators and indicators of equity market and credit market conditions. Our estimates show a strong association between that credit market conditions and the market price of risk. The estimated coefficients have the correct signs. Overall, our results provide some evidence of the partial segmentation of credit markets. ∗Preliminary and incomplete. Please do not circulate. Comments welcome †London School of Economics and CEPR. [email protected]. I would like to thank Poornima Sankar, Xiao Wang and Ping Zhou for research assistance. This research has been supported by the EPSRC Grant No. EPRC522958/1, “Integrating Historical Data and Market Expectations in Risk Assessment for Financial Institutions”. I have benefitted from comments by Aneel Keswani, K.G. Lim, Yves Nosbusch and participants at the FMG workshop on “Integrating Historical and Expectations Data in Financial Econometrics”. Responsibility for all views expressed and all errors is my own.