TRESHOLDS FOR RATINGS' FORECAST DEFAULT PROBABILITIES: SOME QUANTITATIVE EVIDENCES

DE LISA, RICCARDO;MARCHESI M;ZEDDA, STEFANO

2009-01-01

---

Abstract

Basel Committee for Banking Supervision has proposed a procedure to “map” the ratings of different rating agencies into the risk weights that determine credit institutions’ capital requirements. Some observers notice that this procedure seems to lead to very broad ranges, and worry that this could prevent the prompt detection of on-going weaknesses in the rat-ing assignment standards followed by some rating agencies. This paper aims at providing an evaluation of whether and to what extent this worry is justified presenting an approach based on the binomial distribution to estimate rating thresholds across the various rating classes. On the basis of the theoretical characteristics of the binomial distribution, it is in fact possible to test the dis-criminatory power of the mapping methodology recommended by Basel II to supervisors when associating risk weights to rating agencies ratings in the Standardized Approach.