Uncertainty in asset correlation for portfolio credit risk: the shortcomings of the Basel II framework

Moody’s databases of corporate issuers of long term bonds and structured products are used to estimate asset correlations across a group of sectors, world regions and products. The estimation of a dynamic factor model for default risk is performed using Bayesian methods. Results indicate that a two factor model rather than the one factor model, as proposed by the Basel II framework, better represents the historical default data. Furthermore, the results reinforce the importance of unobserved factors in this type of models and point out that the levels of the implied asset correlations critically depend on the latent state variable used to capture the dynamics of default, as well as other assumptions on the statistical model. Finally, the posterior distributions of the asset correlations show that the Basel recommended bounds, for this parameter, undermine the level of systemic risk. JEL Classification: G32, G33, C01.