Predicting corporate failure of listed companies in Sri Lanka

Abstract

The purpose of this research is to develop a model to predict corporate failure of listed companies in Sri Lanka. This study utilized publicly available data from annual repots of a sample of 70 failed firms and a sample of matched 70 non failed firms listed on Colombo stock market for a period covering the 2002 to 2008 financial years with logistic regression analysis. A total of seven corporate governance variables were used as predictor variables of corporate failure. Analysis of the statistical testing results indicated that Model consists with corporate governance variables improved the prediction accuracy to reach 82.86% one year prior to failure. Furthermore, predictive accuracy of the Model in all three years prior to failure is above 73%. Hence model is robust in obtaining accurate results for up to three years prior to failure. Final model includes four corporate governance variables, outside director ratio, CEO duality, remuneration of board of directors and company audit committee. These variables are having more explanatory power to predict corporate failure. Therefore, model developed in this study can assist investors, managers, shareholders, financial institutions, auditors and regulatory agents in Sri Lanka to forecast corporate failure of listed companies.