Predicting corporate Failure: A study on selected Bangladeshi companies

Abstract

The prediction of financial distress holds significant importance for the stakeholders of a company, as it helps them to proactively implement preventive measures such as policy adjustments or restructuring both operational and financial frameworks. The timely prediction serves as a catalyst for informed decisions encompassing investments, credit extensions, bank loan approvals, and more. The occurrence of corporate insolvency imposes considerable costs upon diverse stakeholders, including debt providers, shareholders, suppliers, employees, auditors, customers, and others. Therefore, early detection plays a crucial role in enabling these stakeholders to make well-informed choices that drive effective decision-making processes. Henceforth, the focal point of this research resides in the prediction of corporate insolvency within the business entities of Bangladesh. The significance of this study is that it demonstrates the necessity of using prediction models to forecast the financial condition of entities classified under the Z category and OTC. It also emphasizes the importance of implementing alternative measures to protect the interests of various stakeholders. This is crucial because general investors are unaware of the true financial health of companies transferred to the Z category or OTC, as explained by the BSEC. Simply designating firms as Z category or OTC is insufficient. Despite being classified as Z category by the regulator, these firms do not experience any impact on trading and there is no reflection in stock prices. Instead, there is an upward movement in the prices of certain low-quality securities, which poses a risk to general investors when price corrections occur. Consequently, the capital market can become unstable. Additionally, there have been instances where the regulator was unable to trace certain companies in the OTC, which is detrimental to general investors. Therefore, utilizing a failure prediction model for distressed firms is necessary to initiate effective actions that protect the interests of general investors. When a firm reaches a distressed level that warrants insolvency declaration, there must be a robust infrastructure for bankruptcy, enabling immediate filing to mitigate losses associated with restructuring procedures or the bankruptcy process. Otherwise, if there is a delay in the bankruptcy or restructuring procedure, it creates three impacts (Grigaraviˇcius, 2003). First, it increases direct and indirect spending related to bankruptcy. Second, it decreases the recovery potentials of the indebted firms. Third, it reduces the reimbursements of obligations to creditors. Hence, it is imperative for the distressed firm to promptly initiate the bankruptcy appeal during the initial stage of their indebtedness; otherwise, these issues will exacerbate. To avert failure, it is crucial for the Chief Executive Officer to grasp the nature and facets of failure comprehensively. Subsequently, corrective measures need to be implemented to prevent such failure. Mistakes should be acknowledged, and precautionary actions should be taken to safeguard the organization from future errors. Based on the news report from bdnews24 (Only 2 out, 2006), it was revealed that a mere two out of thirty-three delisted companies opted to repurchase stocks from the public between 1994 and 2006. This particular situation serves as a testament to the fact that only a small fraction, six percent to be precise, of shareholders were able to reclaim their 6 investments through buyback arrangements initiated by the sponsor or director. Consequently, a staggering 94% of shareholders find themselves in a precarious position, their interests in these delisted companies left unaddressed. Hence, this thesis contends that the governing authority ought to adopt a more proactive approach rather than simply designating such firms as Z category or OTC. Instead, the authority should employ a failure prediction model to identify companies facing severe financial distress and take necessary steps to liquidate them forcibly. Following the mandatory liquidation, the funds recovered should be promptly redistributed to the shareholders and other stakeholders who are rightfully entitled to receive their dues. This study contributes in three aspects by addressing the following three gaps. First, the inclusion of OTC companies in predicting corporate failure fills a gap in this field of research, as no prior study has utilized data from OTC companies. Second, this study also addresses the gap of incorporating the recent data of Z category companies, as the previous study by Chowdhury & Barua (2009) only covered data up to 2009. In contrast, our study includes the most recent data of Z category firms, spanning up to 2019. Third, this study addresses the research gap by utilizing forward logistic regression to identify the most influential predictors in predicting corporate failure. The reason for choosing this method is the limited number of studies conducted using it. Therefore, this research will make a valuable contribution to the existing literature. There are two primary aims of this study: 1. To find out whether there are any financially unhealthy firms in the Z category and OTC companies; 2. To identify the predictors that impact the financial failures of the Z category and OTC companies. There are two secondary aims of this study: 1. To understand the financial characteristics of the Z category and OTC companies; 2. To determine whether the characteristics of financially unhealthy companies in the Z category and OTC differ significantly from those in financially healthy positions. As a data collection method, primary data is adopted. For this purpose, the author contacted the particular stakeholders of Bangladesh Securities and Exchange Commission (BSEC) and Dhaka Stock Exchanges (DSE) for the data of failed or liquidated companies. But the concerned officers of both the offices do not maintain any data related to those companies. Both the authorities keep only data of active companies whose shares are trading on the market i.e., the stock exchange of Bangladesh. Later the author asked for the data of Over-The-Counter (OTC) companies because the annual reports are not available on the website of the OTC companies. Finally, the author decided to continue this study using the data of OTC companies because no study was done on the companies in the Over-the-Counter (OTC) trading platform. Besides using the data of OTC companies, this study will also include the data of Z-category companies. Although there was a previous study (Chowdhury &Barua, 2009) on Z-category companies, this study will consider the recent data for those companies. 7 As a means of collecting data, the researcher has employed primary data collection methodology. For that purpose, the author reached out to the specific stakeholders associated with the Bangladesh Securities and Exchange Commission (BSEC) and the Dhaka Stock Exchanges (DSE) in order to obtain data pertaining to companies that have experienced failure or liquidation. But, it was discovered that the officials in both offices do not preserve any records pertaining to such companies. Instead, they exclusively preserve data on active companies whose shares are actively traded on the Bangladeshi stock exchange market. Then, the author requested the data concerning Over-The-Counter (OTC) companies, as the annual reports of these companies were not available on their respective websites. Consequently, the author resolved to proceed with the study utilizing data from OTC companies, primarily due to the absence of previous research conducted on companies operating within the Over-the-Counter (OTC) trading platform. In addition to utilizing data from OTC companies, this study will also incorporate data from Z-category companies. While a prior study (Chowdhury & Barua, 2009) did examine Z-category companies, this present study will focus on the most recent data available for Z-category companies. To collect data, a sample of 35 companies was taken out of 46 Z-category companies. The selection was based on the availability of annual reports on the websites of those companies. Data from 2007 to 2019 were collected, considering their availability. Additionally, data from 13 companies in the OTC market were collected through hardcopy records obtained from Dhaka Stock Exchange. In total, the study utilized a dataset comprising 217 firm years, with 26 firm-years originating from OTC companies. Among the Z-category firms, there were 191 firm-years of data, with 142 firm-years belonging to manufacturing and service providing companies, while the remaining 49 firm-years pertained to bank and non-bank financial institutions (NBFI). In the analysis section of this study, the financial characteristics of the Z category and OTC companies are determined through the calculation of descriptive statistics. Subsequently, Altman's (1968) model is employed to calculate the Z score in order to determine the presence of failed, grey, and non-failed positions within the Z category and OTC companies. Subsequently, the application of One Way ANOVA and Independent Samples T-Test helps in identifying significant differences in the mean values of the financial position predictors among the failed, grey, and non-failed statuses. Finally, through the utilization of Forward Logistic Regression, the factors or predictors with the greatest impact on the financial failures of the Z category and OTC companies are determined. This research reveals that the overall failure rate among companies categorized as Z is 72%. These findings align with the results of a previous study conducted by Chowdhury and Barua (2009), which reported a 77% failure rate among companies. In a more specific context, an alarming 98% of Bank and Non-Bank Financial Institutions in the Z category are experiencing failure. This finding mirrors the conclusions drawn from a study conducted by Hamid et al. (2016), where a substantial 93% of companies were found to be in a failed position. 8 In order to fulfill the first primary objective of the study in assessing the financial health of firms categorized as Z and those traded over-the-counter (OTC), the results reveal significant insights. Manufacturing and servicing companies in the Z category encountered a failed financial position in 59% of firm years, while 19% of firm years were classified as non-failed and 22% as grey. Conversely, Z category banks and non-bank financial institutions experienced a failed financial position in 98% of firm years, with a mere 2% categorized as grey and none falling under the non-failed category. Regarding OTC manufacturing and servicing companies, 92% of firm years faced a failed financial position, while 8% were deemed as grey. Similar to Z category financial institutions, no firm years were classified as non-failed. These findings unequivocally indicate that Z category banks and non-bank financial institutions are entrenched in an exceedingly weakened state. In order to attain the second primary objective of this study, which involves identifying the predictors with the greatest impact on predicting financial failures of Z category and OTC companies, the application of Forward Logistic Regression has yielded significant findings. It has been observed that when considering the single impact, a substantial 78.0% correct variation in the dependent variable (i.e., failed and non-failed positions) can be explained by the ratio of Earnings before Interest and Taxes to Total Assets. When the combined impact is taken into account, the dependent variable's correct variation is explained by four independent variables (X1, X3, X4, X5), amounting to 95.8%. Thus, it can be deduced that the prediction of failure can be enhanced by considering the following variables: X1 (Current assets minus current liabilities divided by total assets), X3 (Earnings before interest and taxes divided by total assets), X4 (Book value equity divided by book value of total debt or liability), and X5 (Sales divided by total assets). Furthermore, it has been determined that X2 (Retained Earnings divided by total assets) does not serve as a reliable predictor when it comes to forecasting corporate failure. The findings derived from the secondary objectives of this study, which aimed to explore the financial characteristics of Z category and OTC (Over-the-Counter) companies, reveal imperative insights. When examining the gross financial data of Z category companies, the descriptive statistics demonstrate that the minimum balance of Retained Earnings, Earnings before Interest and Taxes, and Book Value of Equity are all situated in negative territory. Furthermore, the mean value of Retained Earnings also showcases a negative figure. Conversely, when analyzing the ratios-based descriptive statistics of Z category companies, we observe that the mean value of the net working capital ratio and the Retained Earnings/Total assets ratio both exhibit negative figures. Remarkably, the descriptive statistics for OTC companies exhibit similar trends to those of Z category companies. In order to address another secondary objective of the study, which involves discerning notable distinctions in the attributes between financially unstable companies classified under the Z category and OTC, and those in a sound financial position, two statistical tests were employed: the Independent Samples T-Test and One Way ANOVA. The results indicate that when applying the Altman Z score to Z category Bank and NBFI companies 9 as well as OTC companies, only two outcome groups, namely "failed" and "grey," were observed, with no instances of non-failed firm years being identified. To compare the means of these two groups, the Independent Samples T-Test was utilized. Based on the findings from the Independent Samples T-Test, it was determined that only one ratio, specifically EBIT/Total assets, exhibited significant differences when comparing the "failed" and "grey" firms. On the other hand, the outcomes of the one-way analysis of variance (ANOVA) reveal the following mean values for X1: -0.0773 for Failed firms, 0.4152 for Non-Failed firms, and 0.2605 for Grey firms. Consequently, it can be inferred that Failed firms tend to exhibit a negative mean value for the net working capital ratio. Similarly, the mean values for X2 are as follows: -0.3204 for Failed firms, 0.1915 for Non Failed firms, and 0.1043 for Grey firms. Hence, it can be deduced that failed firms tend to display a negative mean value for the Retained Earnings/Total assets ratio. Moreover, the mean values for X3 are 0.0008 for failed firms, 0.0725 for Non-Failed firms, and 0.0786 for Grey firms. Thus, it can be inferred that the mean value of the Earnings before interest and taxes/Total Assets ratio for failed firms tends to be considerably lower compared to Non-Failed firms. Similarly, the mean values for X4 are 0.9879 for failed firms, 10.9985 for Non-Failed firms and 1.6087 for Grey firms. Consequently, it can be deduced that the mean value of the Book value equity/Book value of total debt or Liability ratio for failed firms tends to be significantly lower compared to Non-Failed firms. However, in terms of X5 (Sales/Total assets), there are no significant differences observed among Failed, Non Failed, and Grey firms. Hence, this finding indicates that only one ratio, specifically Earnings before interest and taxes divided by Total assets, exhibits significant differences when comparing Failed and Grey firms. It is worth noting that no firm-year falls under the category of "Non-Failed" within the OTC companies. In conclusion, the study asserts that the mere classification of certain firms into either the Z category or OTC category falls short in addressing the underlying issues. The findings of the study indicate a staggering failure rate of up to 98% and 92% for firms in the Z category and OTC category, respectively. Consequently, it becomes imperative to employ a failure prediction model in order to identify extremely distressed firms and implement proactive measures to safeguard the interests of general investors and other stakeholders. When a firm reaches a state of distress that necessitates an insolvency declaration, it becomes crucial to establish a robust infrastructure for bankruptcy proceedings. This would enable swift filing, thereby mitigating losses associated with the restructuring or bankruptcy procedures. Based on the findings of the study, it is recommended that employing Forward Logistic Regression can effectively uncover the key variables that play a significant role in predicting corporate failure. These insights can be invaluable for decision makers, enabling them to identify the factors with the greatest predictive power for corporate failure.