Structured Products and Their Ratings: Are They Sufficiently Stable for Regulatory Purposes?

Abstract

Structured products performed very badly during the subprime crisis. Not only they suffered massive ratings downgrades (even at the most senior levels of the capital structure) but also inflicted significant losses on investors. This study shows, using numerical simulations based on historical data, that such structures were highly unstable. In other words, minor errors in the basic assumptions could manifest themselves dramatically in the accuracy of their ratings. In fact, for all practical purposes, in many cases it was impossible to make a reliable statement regarding the future performance of a synthetic CDO tranche, regardless of the quality of the underlying assets. Moreover, the study demonstrates that single-point credit risk estimators (in which no attempt at specifying a confidence interval is made) could be specially misleading. Finally, the study suggests that a regulatory framework based on ratings “as we know them” is unlikely to be effective and probably much fresh thinking is needed in this area.