RATING AGENCIES SHOULD USE PROBABILITIES.

RATING AGENCIES SHOULD USE PROBABILITIES. I posted here that intelligence estimates by the CIA should be expressed in terms of probabilities just as weather forecasts are. More information is conveyed and people can compare and discuss opinions. Charles Calomiris makes a number of suggestions for reforming financial markets in this article and one of them is that rating agencies should express their ratings in terms of the probability of default and the loss given a default. The rating agencies have used letter rankings which have no direct meaning. This wikipedia article gives examples: “The Standard & Poor’s rating scale is as follows, from excellent to poor: AAA, AA, A, BBB, BB, B, CCC, CC, C, D. Anything lower than a BBB rating is considered a speculative or junk bond…. The Moody’s rating system is similar in concept but the naming is a little different. It is as follows, from excellent to poor: AAA, Aa1, Aa2, Aa3, A1, A2, A3, Baa1, Baa2, Baa3, Ba1, Ba2, Ba3, B1, B2, B3, Caa1, Caa2, Caa3, Ca, C.” These ratings have no intuitive meaning. Ratings in terms of probabilities of default would make the agencies more accountable by providing a yardstick to measure their performance. Saying that a rating agency incorrectly rated bonds as Baa1 when they should have been rated Ba1 or Caa1 is not the same as saying that their default rate was much higher than predicted. People are comfortable with probabilities and numbers and they should be used.