Credit Expansion and Asset Correlation: Asking for Trouble?

Abstract

After the recent financial crisis, it came to light that the correlation parameters in the evaluation models used by credit rating agencies to assess the default risk of structured financial products were largely inaccurate. This nonperformance was partly based on a lack of historical data to estimate correlation parameters in risk-evaluation models. However, the models also notably lacked the guidance of an overarching macroeconomic theoretical framework to place them in a context of central bank-induced bank credit expansion. This paper argues that had such a theoretical framework been adopted, it would have been possible to anticipate somewhat a wave of high correlations for credit-induced boom assets, as these are eventually driven by monetary expansion and other institutional factors. A full understanding of these effects would necessarily lead to more realistic and conservative asset correlation estimates and consequently improve the design of risk-evaluation models.