An Integrated Model to Predict Corporate Failure of Listed Companies in Sri Lanka

The primary objective of this study is to develop an integrated model to predict corporate failure of listed companies in Sri Lanka. The logistic regression analysis was employed to a data set of 70 matched-pairs of failed and non-failed companies listed in the Colombo Stock Exchange (CSE) in Sri Lanka over the period 2002 to 2010. A total of fifteen financial ratios and eight corporate governance variables were used as predictor variables of corporate failure. Analysis of the statistical testing results indicated that model consists with both corporate governance variables and financial ratios improved the prediction accuracy to reach 88.57 per cent one year prior to failure. Furthermore, predictive accuracy of this model in all three years prior to failure is above 80 per cent. Hence model is robust in obtaining accurate results for up to three years prior to failure. It was further found that two financial ratios, working capital to total assets and cash flow from operating activities to total assets, and two corporate governance variables, outside director ratio and company audit committee 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.