Predicting Bankruptcy via Cross-Sectional Earnings Forecasts

Abstract

We develop a model to predict bankruptcies, exploiting that negative book equity is a strong indicator of financial distress. Accordingly, our key predictor of bankruptcy is the probability that future losses will deplete a firm’s book equity. To calculate this probability, we use earnings forecasts and their standard deviations obtained from cross-sectional regression models in the spirit of Hou, van Dijk, and Zhang (2012). We add variables that we find to discriminate between bankrupt and non-bankrupt firms. As our model requires only accounting data, we can provide bankruptcy predictions for a wide range of firms, including firms that have no access to capital markets. In strictly out-of-sample tests, we show that our accounting model performs better than alternative corporate failure models that use only accounting information. If we additionally allow for stock market information, our approach also outperforms leading alternatives that require market data.

Keywords: bankruptcy prediction, negative book equity, mechanical earnings forecasts, financial distress

JEL Classification: G17, G33, M41, C25