The best-known measures of prospective bond quality are the bond ratings assigned by the three agencies: Moody’s, Standard & Poor’s, and Fitch. Their ratings provide a judgment of the investment quality of a long-term obligation and a measure of default risk. Accordingly, they may affect the interest rate an organization pays on its bonds. Although each rating agency has defined the meaning of its ratings, the agencies have not explicitly specified the process they use to arrive at these ratings. Given the importance of ratings, various authors have attempted to explain and predict them based on the financial and/or statistical characteristics of the bonds and issuing firms. These rating-prediction studies were reviewed in the previous chapter. Although the models derived do an adequate job of capturing the human judgments of bond raters, they suffer from (a) a diversity of approaches used in selecting independent variables for the regression, discriminant, or multivariate probit models; (b) a lack of an ‘‘economic rationale’’ underlying the choice of these variables; 1 (c) a failure to account for the differences among the companies in their accounting for long-term leases; and (d) the confusion of ex ante predictive power with ex post discrimination. Consequently this study will correct for the above limitations to develop a multiple-discrimination bond-rating model.
A bond rating is primarily a judgment of the investment quality of a firm’s long-term obligation. It reflects the raters’ expectations and estimates of the relevant characteristics of the quality of the investment. To capture the deter-