BeschreibungRating agencies report ordinal ratings in discrete classes. We question the market's implicit assumption that agencies deﬁne their classes on identical scales. To this end, we develop a non-parametric method to estimate the relation of rating scales for pairs of raters. This scale relation identiﬁes for every rating class of one rater the extent to which it corresponds to any rating class of another, and hence enables a rating-class speciﬁc re-mapping of one agency's ratings to another's scale. Our method is based purely on ordinal co-ratings to obviate error-prone estimation of PDs and disputable assumptions involved, and exploits structure in the joint estimation of all rating classes' relations from a pair of raters.
We ﬁnd evidence against the hypothesis of identical scales for the three major rating agencies Fitch, Moody's and Standard & Poor's, provide the relations of their rating classes and illustrate the importance of correcting for scale relations in benchmarking.
|11 Mai 2011 → 13 Mai 2011
|Association Française de Finance 2011 Spring Conference