The Usefulness of Accounting Information, Economic Variables and Corporate Governance Measures in Predicting Corporate Failure - An Empirical Study

Abstract

The present study has the core aim of investigating the usefulness of employing accounting information, macroeconomic variables and corporate governance measures to predict corporate failure in an Egyptian setting. The research is directed to adapting a corporate failure prediction model applied to a sample of Egyptian companies listed in the Egyptian stock market. A sample of 79 companies drawn from the 100 most actively traded firms listed in the Egyptian stock market has been used for the empirical testing. A pooled sample is formed covering the period 2000-2005 inclusive. The empirical study emphasizes improving failure prediction accuracy by introducing two classes of variables besides financial ratios based on accounting information. These classes of variables are economic variables and corporate governance measures. Logistic regression analysis has been used to test the predictive accuracy of four models. Model I includes accounting information only. Model II adds economic variables to accounting information. Model III includes corporate governance measures and accounting information. Finally, model IV employs these three classes of variables together. Analysis of the testing results indicate that employing the three classes of variables together improved the prediction accuracy to reach 84.8% in the classification sample and 78.2% in the validation sample. Furthermore, model IV is used to predict failure up to three years prior to failure and therefore can provide a tool for failure prediction for enhancing auditing, investment and credit decisions in an Egyptian market setting.