Forecasting Bank Credit Ratings

We recently added Forecasting Bank Credit Ratings by Periklis Gogas, Theophilos Papadimitriou and Anna Agrapetidou from the Department of Economics, Democritus University of Thrace, Komotini, Greece to our policy library. The study was recently published in The Journal of Risk Finance (Volume 15, Issue 2, pp 195-209). Below is the abstract:

Purpose – This study aims to present an empirical model designed to forecast bank credit ratings using only quantitative and publicly available information from their financial statements. For this reason, the authors use the long-term ratings provided by Fitch in 2012. The sample consists of 92 US banks and publicly available information in annual frequency from their financial statements from 2008 to 2011.

Design/methodology/approach – First, in the effort to select the most informative regressors from a long list of financial variables and ratios, the authors use stepwise least squares and select several alternative sets of variables. Then, these sets of variables are used in an ordered probit regression setting to forecast the long-term credit ratings.

Findings – Under this scheme, the forecasting accuracy of the best model reaches 83.70 percent when nine explanatory variables are used.

Originality/value – The results indicate that bank credit ratings largely rely on historical data making them respond sluggishly and after any financial problems are already known to the public.