An Analysis of Financial Distress Prediction of Selected Listed Companies in Colombo Stock Exchange

Introduction

Financial distress of business organizations has been a widely discussed topic over the past few decades due to the magnitude of the consequences it has on its stakeholders. The business world witnessed several traumatic bankruptcies of business giants within the past couple of decades which had a calibre of affecting the economies as a whole. A major reason behind these bankruptcies were the non–identification of financial distress at an earlier stage which could have assisted these firms to mitigate such tragedies or reduce the consequences to its stakeholders. Thus, early identification of financial distress was becoming a spotlight in the field of finance.

As capital markets developed, the investors of companies became withdrawn from the day to day operations of the businesses. The shareholders of the company appointed the management of the business to a board of directors who were intended to act in the best interest of its shareholders. Hence, investors of public companies received information on the financial health of the company only when the financial statements were published. Apart from investors, early identification of financial distress became important to lenders such as financial institutions and creditors as well as government institutions and several other stakeholders considering that it was crucial for their financial decision making.

Researchers started studying about the phenomena as it became more challenging to identify using a mere simple method. This was highly affected by the complexity and scale of the business models and decision making within the growing economies. The common causes for financial distress were identified as inefficient debt management, absence of adequate financial knowledge, failure of long term capital plans etc. With such identification, the researchers further extended the studies in predicting financial distress and bankruptcy risk.

Among the numerous amounts of studies conducted by researchers all around the world, seminal studies by Altman (1968) and Beaver (1966) started a new era in the study of financial distress and bankruptcy risk. Altman Z-score is a multivariate approach based on financial ratios and discriminant analysis. Multivariate Discriminant Analysis (MDA) (has since then been the traditional methodology of predicting bankruptcy/ financial distress. This approach has been used in numerous number of researches on financial distress/ bankruptcy prediction including Altman Z-score. There have been several criticisms regarding the use of MDA such as assumption of multivariate normality, bias on extreme data. These criticisms have led to a search for various developed techniques which could be used for more complex predictability of financial distress.

After Altman’s study, several approaches have been used by researchers over the time including univariate technique, multivariate technique, linear multiple discriminant techniques, logit regressions, factor analysis, artificial neural networks etc. The studies used these various approaches to increase the accuracy of prediction and to avoid the weaknesses of the previous models.

Loopholes and weaknesses in existing studies have left opportunities for further development of financial distress prediction. A major limitation is the effective range of applicability of the models in different contexts. This study is an effort taken to improve an evolutionary technique in predicting financial distress to cope up with dynamic nature of the financial environment.

This study focuses on developments to the existing literature on financial distress prediction. The study uses a rather newer approach to distress prediction; Artificial Neural Network (ANN) and Panel Regression Analysis to bring a newer perspective on the matter. Further, it aimed on broadening the understanding of variables assisting prediction of financial distress and identifying a prediction model that is effective to listed companies within Sri Lankan economic setting. Focus is given to additional areas such as cash flow and market based variables in addition to the financial ratios used in Altman Z-score. Higher prediction accuracy is achieved through this study as Artificial Neural Networks is used as the approach for prediction. In addition, it also studies the relationship of leverage, market based variables and ownership structure to financial distress of the selected companies using panel regression. Use of a developing economy and addition of new dimensions in financial distress prediction are expected to provide a new investigable area for the topic.