Predicting the Bond Ratings of S&P 500 Firms

Abstract

In this paper, we have developed models to find out as to what factors are important in determining the bond ratings of the non-financial firms which are included in S&P 500 index. Our analysis is different from other analyses in the literature because we have used the more recent data, i.e., the ratings belong to the years 2008, 2009 and 2010. We have performed two types of analyses. In the first analysis, all the variables are used as explanatory variables after eliminating some variables to avoid multicollinearity. In the second analysis, factor analysis is performed to group the variables into factors, and variables whose correlations with the factors are the highest are used as explanatory variables. In both the analyses, multiple discriminant analysis, ordered logit, and ordered probit models are estimated. The best model is the ordered logit model that used all the variables. The important factors that determine the bond ratings are long-term liabilities/total assets ratio, return on equity, net profit margin, trade payables, and operating income. The firms that need to improve their bond ratings must pay attention to these factors. Also, by using the models presented in the paper, investors can have an idea about the credibility of the issuers.