A horse race of models and estimation methods for predicting bankruptcy

We use a comprehensive set of performance metrics to analyze the improvement in the classification power and prediction accuracy of various bankruptcy prediction models after adding governance variables and/or varying the estimation method used. In a sample covering bankruptcies of U.S. public firms in the period 2000 to 2015, we find that the addition of governance variables significantly improves the performance of all bankruptcy prediction models. We also find that the additional explanatory power provided by governance measures improves the further the firm is from bankruptcy, which suggests that governance variables may provide earlier and more accurate warning of the firm's bankruptcy potential. Our findings show that the performance of any bankruptcy prediction model is significantly affected by the estimation method used. We find that regardless of the bankruptcy model, hazard analysis provides the best classification and out-of-sample forecast accuracy among the parametric methods. Furthermore, non-parametric methods such as neural networks, data envelopment analysis or classification and regression trees appear to provide comparable and sometimes superior classification accuracy to hazard analysis. Lastly, we use the dynamic panel generalized methods of moments model to address concerns raised in prior studies about the susceptibility of similar studies to endogeneity issues and find that our findings continue to hold.     

Innovate or die: Corporate innovation and bankruptcy forecasts

We investigate the relationship between a firm’s innovation performance and its probability of bankruptcy. Estimating the discrete hazard model with a comprehensive set of bankruptcies spanning the period of 1980–2009, we find several previously neglected innovation-based variables are important determinants of bankruptcy probability, especially for firms belonging to technology-intensive industries. R&D productivity demonstrates persistent significance across different prediction horizons while the predictive power of patent count becomes larger and more significant at longer prediction horizons. We also find that a firm’s organization capital intensity correlates positively with future bankruptcy.     

Bankruptcy prediction of financially stressed firms: an examination of the predictive accuracy of artificial neural networks

This is an extension of prior studies that have used artificial neural networks to predict bankruptcy. The incremental contribution of this study is threefold. First, we use only financially stressed firms in our control sample. This enables the models to more closely approximate the actual decision processes of auditors and other interested parties. Second, we develop a more parsimonious model using qualitative ‘bad news’ variables that prior research indicates measure financial distress. Past research has focused on the ‘usefulness’ of accounting numbers and therefore often ignored non‐accounting variables that may contribute to the classification accuracy of the distress prediction models. In addition, rather than use multiple financial ratios, we include a single variable of financial distress using the Zmijewski distress score that incorporates ratios measuring profitability, liquidity, and solvency. Finally, we develop and test a genetic algorithm neural network model. We examine its predictive ability to that of a backpropagation neural network and a model using multiple discriminant analysis. The results indicate that the misclassification cost of the genetic algorithm‐based neural network was the lowest among the models. Copyright © 2001 John Wiley & Sons, Ltd.     

A re-evaluation of auditors’ opinions versus statistical models in bankruptcy prediction

Existent empirical evidence on the relative performance of auditors’ going concern opinions versus statistical models in predicting bankruptcy is mixed. This study attempts to add new reliable evidence on this important issue by conducting the comparison based upon an improved statistical model. The improved statistical model incorporates some new developments advocated by recent bankruptcy prediction research (e.g., Shumway, 2001). First, the following non-traditional variables are added: a composite measure of financial distress, industry failure rate, abnormal stock returns, and market capitalization. Secondly, a hazard model is employed. The prediction ability of the hazard model with incorporation of non-financial-ratio variables is superior to that of auditors’ going concern opinions in the holdout sample. This suggests that a well-developed statistical model could serve as a decision aid for auditors to better make going-concern judgments. Further analyses reveal some evidence that industry failure rate does not have a significant impact upon auditors’ going concern judgments as it should be; auditors could improve their going concern judgments by considering industry-level information in addition to firm-specific information. Finally, we find that auditors’ opinions do have incremental contribution beyond stock-market information and industry failure rate in predicting bankruptcy.

Are Auditors Professionally Skeptical? Evidence from Auditors’ Going‐Concern Opinions and Management Earnings Forecasts

We examine whether auditors exercise professional skepticism about management earnings forecasts when making going‐concern decisions. Using publicly issued management earnings forecasts as a proxy for earnings forecasts provided by managers to auditors, we find that management earnings forecasts are negatively associated with both auditors’ going‐concern opinions and subsequent bankruptcy. The weight auditors put on management forecasts in the going‐concern decision is not significantly different from the weight implied in the bankruptcy prediction model. Moreover, compared with the bankruptcy model, auditors assign a lower weight to management forecasts they perceive as being less credible, including those (1) issued by managers who issued optimistic forecasts in the previous two years, and (2) predicting high earnings increases or high earnings. Taken together, our evidence is consistent with auditors being professionally skeptical about management earnings forecasts when making going‐concern decisions.     

An empirical comparison of bankruptcy models

Four types of bankruptcy prediction models based on financial statement ratios, cash flows, stock returns, and return standard deviations are compared. Based on a sample of bankruptcies from 1980 to 1991, results indicate that no existing model of bankruptcy adequately captures the data. During the last fiscal year preceding bankruptcy, none of the individual models may be excluded without a loss in explanatory power. If considered in isolation, the cash flow model discriminates most consistently two to three years before bankruptcy. By comparison, the ratio model is the best single model during the year immediately preceding bankruptcy. Quasi-jack-knifing procedures suggest that none of the models can reliably predict bankruptcy more than two years in advance.     

Low‐income,low‐asset debtors in the U.S. bankruptcy system

The United States' bankruptcy system faces a major problem: many consumers are too poor to file for bankruptcy, usually because they cannot afford the necessary attorney fees. Some consumers appear to spend months trying to save the funds to pay their attorneys, thus either delaying their bankruptcies or foregoing bankruptcy altogether when they fail to save enough money. Others file for repayment bankruptcy in order to pay attorney fees during the case, when liquidation bankruptcy is usually a better fit for consumers with low incomes and low asset levels. The most recent comprehensive bankruptcy reform, the Bankruptcy Abuse Prevention and Consumer Protection Act (BAPCPA), exacerbated these problems by implementing additional procedural requirements that resulted in attorneys raising their fees. These problems have led to calls for administrative bankruptcy, especially for low‐income, low‐asset (LILA)/no‐income, no‐asset (NINA) debtors. Administrative bankruptcy would make bankruptcy more accessible by lowering access costs, for example, by eliminating the need for consumers to hire attorneys. Administrative programs in the United States, however, have a history of long‐term decline, especially when these programs serve low‐income people. It has become a cliché that poor people's programs become poor programs. A better solution would be to eliminate the procedural requirements imposed by BAPCPA and simplify the decision consumers must make about which type of bankruptcy to use.     

Consumer Bankruptcies and the Bankruptcy Reform Act: A Time-Series Intervention Analysis, 1960–1997

The Bankruptcy Reform Act of 1978, effective on October 1, 1979, significantly altered the basic rules by which consumers file for bankruptcy. Between 1979 and 1997, the number of nonbusiness bankruptcies filed annually rose from 200,000 to 1.35 million, and the rate of bankruptcies per 100,000 adults increased from 129 to 715. A controversial aspect of bankruptcy is how much of this increase can be attributed to the 1978 act. Early empirical studies provide estimates ranging from a low of 6% to a high of 75% for the immediate post-act period. However, two recent studies using longer data series report that none of the increase was due to the act. Previous studies suffer from several econometric problems, including inadequate attention to stochastic properties and stationarity of the data series, as well as data errors due to reporting changes. This paper uses an ARIMA intervention analysis to estimate the impact of the 1978 act. Using adjusted quarterly data for 1960:3 to 1995:4, the data first are examined for unit roots. The tests reject the presence of seasonal unit roots but confirm the presence of a nonseasonal unit root. The empirical analysis therefore is based on logged first differences of bankruptcy filings and filing rates per capita. An ARIMA model is estimated using the preintervention data for 1960:3 to 1979:3. This model is re-estimated for 1960:3 to 1995:4 with the intervention terms included. The intervention model estimates indicate that the 1978 act increased consumer bankruptcies by 36% in the post-act period relative to the pre-act period, or about 72,400 additional bankruptcies per year. Overall, the net impact of the 1978 act was modest compared to the substantial rise in bankruptcies that has occurred since 1979.     

Using a hybrid evolution approach to forecast financial failures for Taiwan-listed companies

Bankruptcy has been an important topic in finance and accounting research for a long time. Recent major bankruptcies have included seemingly robust companies such as Enron, Kmart, Global Crossing, WorldCom, and Lehman Brothers. These cases have become of serious public concern due to the huge influence these companies have on the real economy. This research proposes a hybrid evolution approach to integrate particle swarm optimization (PSO) with the support vector machine (SVM) technique for the purpose of predicting financial failures. The preparation phase collected an initial sample of 68 companies listed by the Taiwan Stock Exchange Corporation (TSEC). The financial datasets were constructed based on 33 financial ratios, four non-financial ratios and one combined macroeconomic index. To select suitable indicators for the input vector, the principle component analysis (PCA) technique was applied to reduce the data and determine how groupings of indicators measure the same concept. In the swarming phase, PSO was applied to obtain suitable parameters for SVM modeling without reducing the classification accuracy rate. In the modeling phase, the SVM model was used to build a training set that was used to calculate the model's accuracy and fitness value. Finally, these optimized parameters were used in the hybrid PSO–SVM model to evaluate the model's predictive accuracy. This paper provides four critical contributions. (1) Using the PCA technique, the statistical results indicate that the financial prediction performance is mainly affected by financial ratios rather than non-financial and macroeconomic ratios. (2) Even with the input of nearly 70% fewer indicators, our approach is still able to provide highly accurate forecasts of financial bankruptcy. (3) The empirical results show that the PSO–SVM model provides better classification accuracy (i.e. normal vs. bankrupt) than the grid search (Grid–SVM) approach. (4) For six well-known UCI datasets, the PSO–SVM model also provides better prediction accuracy than the Grid–SVM, GA–SVM, SVM, SOM, and SVR–SOM approaches. Therefore, this paper proposes that the PSO–SVM approach is better suited for predicting potential financial distress.     

CEO turnover and bankrupt firms’ emergence

This paper examines whether CEO turnover within a bankrupt firm predicts the firm's likelihood to reemerge from bankruptcy proceedings as a reorganized entity. Using 836 bankruptcy cases filed under Chapter 11 of the United States Bankruptcy Code from 1989 through 2016, we show that firms that undergo CEO turnover are significantly more likely to emerge from Chapter 11 proceedings. We conduct further analyses to examine the potential mechanisms through which CEO turnover is linked to a firm's chance of emergence. Consistent with the perspective that CEO turnover constitutes an observable event that can signal creditor support, we find that CEO turnover in bankrupt firms is positively associated with debtor-in-possession financing. Additionally, there is a significant increase in managerial quality post-turnover. Further, we document that the predictive power of CEO turnover is stronger in bankruptcy cases with greater uncertainty, such as in free-fall bankruptcies, where there is less preexisting agreement between the firm and its creditors. Overall, our findings provide valuable insight into external investors and stakeholder groups, whose interests are significantly impacted by corporate bankruptcies.     

Model Predictions and Auditor Assessments of Going Concern Status

This paper compares the predictions of a bankruptcy prediction model and the assessments of auditors on the going concern status of a sample of 165 bankrupt companies and 165 matched non-bankrupt companies. Data from US companies for the period 1978 to 1985 were used. Probit analysis (with the weighted exogenous sampling maximum likelihood procedure) was applied to estimate the model parameters. The Lachenbruch U method hold-out accuracy rates of the model are 85.45% for bankrupt firms, 100.00% for non-bankrupt firms, and 99.91% overall. The corresponding accuracy rates of the auditors based on their audit reports are 54.37% for bankrupt firms, 100.00% for non-bankrupt firms, and 99.73% overall. The sensitivity of optimal cut-off points to misclassification costs of Type I and Type II errors was also considered. Results of the study suggest that bankruptcy prediction models can be useful to auditors in making going concern assessments. Further, such models can serve as analytical tools and defensive devices.     

The Bankruptcy System's Chapter 22 Recidivism Problem: How Serious is It?

This paper is adapted from the keynote address from the Eastern Finance Association's 2014 meeting in Pittsburg, Pennsylvania. We highlight a recidivism problem: about 15% of debtors who emerge as continuing entities under Chapter 11, or are acquired as part of the bankruptcy process, ultimately file for bankruptcy protection again (18.25% when considering only those firms which emerge as a continuing, independent entity). We argue that the “Chapter 22” issue should not be dismissed by the bankruptcy community just because no interested party objects during the confirmation hearing. Applying the Z”‐Score model to a large sample of Chapter 11 cases reveals highly different and significant expected survival profiles at emergence. Credible distress prediction techniques can effectively predict the future success of firms emerging from bankruptcy and be used by the bankruptcy court to assess the feasibility of the reorganization plan, a requirement mandated by the Bankruptcy Code. Branch reviews, discusses, and critiques in this follow‐up article to Altman's original thesis.     

Predicting Bankruptcy Resolution

This study examines classification and prediction of the bankruptcy resolution event. Filing of bankruptcy is resolved through one of three alternative resolutions: acquisition, emergence or liquidation. Predicting the final bankruptcy resolution has not been examined in the prior accounting and finance literature. This post-bankruptcy classification and prediction of the final resolution is harder than discriminating between healthy and bankrupt firms because all filing firms are already in financial distress. Motivation for predicting the final resolution is developed and enhanced. A sample of 237 firms filing for bankruptcy is used. Classification and prediction accuracies are determined using a logit model. A ten-variable, three-group resolution logit model, which includes five accounting and five non-accounting variables is developed. The model correctly classifies 62 percent of the firms, significantly better than a random classification. We conclude that non-accounting data add relevant information to financial accounting data for predicting post bankruptcy resolution. Further, public policy implications for investors, researchers, bankruptcy judges, claimants and other stakeholders are discussed.     

基于非线性组合模型的P2P网贷平台危机预警研究

针对P2P网贷平台现金流较大、利润率较低和财务数据获取困难的特点,构建基于平台交易真实数据的危机预警评价指标体系和组合预测模型.将传统的财务评价指标转换成网贷平台交易数据指标,运用邻域粗糙集属性约简的方法对采集的数据指标进行降噪和约减处理,再基于机器学习理论引入神经网络、支持向量机和Logit回归等模型对数据进行训练.通过分组进行单模型和组合模拟预测,提高了新的破产指标下各模型预测的准确率.     

The Accuracy and Incremental Information Content of Audit Reports in Predicting Bankruptcy

A series of corporate failures in which auditors failed to warn about impending bankruptcy led to widespread criticism of the UK auditing profession during the last recession. For a sample of 976 quoted companies (1987–94), this paper shows that there are two reasons why audit reports were not accurate or informative indicators of bankruptcy. First, audit reports poorly reflected publicly available information about the probability of bankruptcy. Secondly, strong persistence in audit reporting reduced the accuracy of audit reports     

Investment opportunities and bankruptcy prediction

A firm's mix of growth options and assets in place is an important determinant of its optimal default strategy. Our simple model shows that shareholders of a firm with valuable investment opportunities would be able/willing to wait longer before defaulting on their contractual debt obligations than shareholders of an otherwise identical firm without such opportunities. More importantly, we show empirically using a dataset of recent corporate bankruptcies that measures of investment opportunities are significantly related to the likelihood of bankruptcy. Augmenting existing bankruptcy prediction models by these measures improves their out-of-sample forecasting ability.     

Are Corporate Restructuring Events Driven by Common Factors? Implications for Takeover Prediction

Abstract:  The paper shows that variables commonly used in takeover prediction models also help to explain the likelihood of several other restructuring events, including divestitures, bankruptcies and significant employee layoffs. This finding helps to explain the larger misclassification errors in binomial takeover prediction models commonly used in prior research. The results show that modelling takeover prediction models in a binomial setting is likely to lead to misspecification in the parameter estimates and, further, result in erroneous conclusions about the determinants of takeover likelihood. The paper shows that controlling for other restructuring events by using a multinomial framework results in consistently lower misclassification errors in out-of-sample prediction tests, when compared to the benchmark of a typical binomial model.     

United States Bankruptcy Code,Section 304(b)(3): “Other Appropriate Relif” for Multinational Bankruptcy

Section 304 of the United States Bankruptcy Code was enacted to provide a statutory solution to the “murky and uncharted waters” of multinational bankruptcies. Part I of this article provides a brief background and analysis of section 304; Part II canvasses court decisions fashioning relief to multinational bankruptcies; and in Part III, the author attempts to synthesize the legislative goal of efficient multinational bankruptcy administration with examples of “appropriate relief” that American bankruptcy courts should consider. L'article 304 du code américain sur l'insolvabilité a été passéafin de fournir une solution légale aux “eaux troubles et inconnues” des faillites multinationales. La première partie de cet article donne une brève histoire et analyse de l'article 304; la deuxieme partie couvre les décisions juridiques qui détermient comment assister les compagnies mutinationales en faillite; et dans la troisième partie, l'auteur essaye de synthétiser le but législatif d'une administration éfficace des faillites multinationales avec des exemples de “secours adéquats” que les tribunaux américains de faillite devraient prendre en compte.