Predicting Railroad Bankruptcies in America

Abstract

The purpose of this paper is to discuss the urgent need for an early-warning system covering the historically failure-prone railroad industry and to develop a tool for providing such a system. A multivariate statistical technique called linear discriminant analysis is utilized to identify and quantify those financial measures which are effective indicators of bankruptcies. A model which combined several financial statement ratios proved to be extremely accurate in predicting railroad bankruptcies at one and two annual financial statement dates prior to failure. Subsequent tests on additional railroad samples confirm the validity of the model. Finally, currently existing railroads in America are assessed for their bankruptcy potential by this diagnostic model.