The Barra Credit Series: Default and Information

Jan 1, 2003

The recent string of accounting scandals highlights the need for bond investors to reconsider the informational assumptions underlying traditional models of default. In these models it is implicitly assumed that the information used to calibrate and run the model is publicly available. In reality, model inputs and parameters are uncertain. We develop a class of cause and effect default models based on incomplete information, in which investors can specify the degree of confidence they have around inputs and parameters. These models are predicated on the surprise nature of default and typically forecast positive short­term credit spreads. The level and shape of the credit spread term structure is determined by investors' information level. We introduce the pricing trend and provide a reduced­form representation of our cause and effect model class in terms of this trend. Closed­form expressions of the pricing trend, conditional default rates, default probabilities, prices of default­sensitive securities, and credit spreads are obtained.