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.     

Strategic Bankruptcies and Price Reactions to Bankruptcy Filings

This study investigates whether the stock market differentiates between firms that file bankruptcy petitions for strategic reasons and firms that file bankruptcy petitions for financial reasons. We perform both univariate and regression tests on a sample of 245 firms that filed Chapter 11 bankruptcy petitions between 1981 and 1996. After controlling for bankruptcy outcome, probability of bankruptcy, firm financial condition, and firm size, we find that, in the period around bankruptcy filing, firms that file bankruptcy petitions for financial reasons have significantly larger stock price declines than firms that file bankruptcy petitions for strategic reasons.     

Impaired bank health and default risk

Empirical studies in corporate finance have long been focused on the role of banks in reducing the costs of financial distress. The environment and events in Japan provide a “natural experiment” that allows such empirical studies. The number of bankruptcies steadily increased throughout the 1990s, and peaked in 2000. During this period, Japan's banking sector, in contrast, faced considerable problems regarding the disposal of their bad loans. The purpose of this paper is to investigate how various measures of bank health and how defaults of major trading partners affected the probability of bankruptcy among medium-size firms in Japan. Using probit models, we examine the causes of bankruptcy for unlisted Japanese companies in the late 1990s and early 2000s. We find that several measures of bank-specific financial health have had significant impacts on a borrower's probability of bankruptcy, even when observable characteristics relating to these borrower's financial variables are controlled. In particular, a close bank–firm relationship—which usually reduces the probability of bankruptcy—exacerbates the impacts of a financial crisis, which substantially damages other bank health measures as well.     

Prior Wall Street Journal Announcements of Possible Bankruptcy Filings and Price Reactions to Subsequent Bankruptcy Filings

This study examines the relation between prior Wall Street Journal (WSJ) announcements of possible bankruptcy filings and price reactions to subsequent bankruptcy filings for 336 firms that filed for bankruptcy between 1980 and 1993. Extant research indicates that price reactions to announcements of economic events are inversely related to the amount of surprise in the announcements. Prior WSJ anouncements of possible bankruptcy filings increase the markets a priori assessment of firms' probability of bankruptcy, thereby potentially reducing the surprise in subsequent bankruptcy filings. We hypothesize smaller price reactions to bankruptcy filings for firms where the WSJ previously published an article indicating that the firm may file for bankruptcy. Our results are consistent with this hypothesis. Specifically, we find smaller price reactions to bankruptcy filings for firms with prior WSJ announcements of possible bankruptcy filings. Our results hold after controlling for firm size, probability of bankruptcy, exchange listing, leverage, and predisclosure information.     

A comparison of alternative bankruptcy prediction models

Early models of bankruptcy prediction employed financial ratios drawn from pre-bankruptcy financial statements and performed well both in-sample and out-of-sample. Since then there has been an ongoing effort in the literature to develop models with even greater predictive performance. A significant innovation in the literature was the introduction into bankruptcy prediction models of capital market data such as excess stock returns and stock return volatility, along with the application of the Black–Scholes–Merton option-pricing model. In this note, we test five key bankruptcy models from the literature using an up-to-date data set and find that they each contain unique information regarding the probability of bankruptcy but that their performance varies over time. We build a new model comprising key variables from each of the five models and add a new variable that proxies for the degree of diversification within the firm. The degree of diversification is shown to be negatively associated with the risk of bankruptcy. This more general model outperforms the existing models in a variety of in-sample and out-of-sample tests.     

Automatic bankruptcy auctions and fire-sales

We test for fire-sale tendencies in automatic bankruptcy auctions. We find evidence consistent with fire-sale discounts when the auction leads to piecemeal liquidation, but not when the bankrupt firm is acquired as a going concern. Neither industry-wide distress nor the industry affiliation of the buyer affect prices in going-concern sales. Bids are often structured as leveraged buyouts, which relaxes liquidity constraints and reduces bidder underinvestment incentives in the presence of debt overhang. Prices in “prepack” auctions (sales agreements negotiated prior to bankruptcy filing) are on average lower than for in-auction going-concern sales, suggesting that prepacks may help preempt excessive liquidation when the auction is expected to be illiquid. Prepack targets have a greater industry-adjusted probability of refiling for bankruptcy, indicating that liquidation preemption is a risky strategy.     

Variable selection and corporate bankruptcy forecasts

We investigate the relative importance of various bankruptcy predictors commonly used in the existing literature by applying a variable selection technique, the least absolute shrinkage and selection operator (LASSO), to a comprehensive bankruptcy database. Over the 1980–2009 period, LASSO admits the majority of Campbell et al. (2008) predictive variables into the bankruptcy forecast model. Interestingly, by contrast with recent studies, some financial ratios constructed from only accounting data also contain significant incremental information about future default risk, and their importance relative to that of market-based variables in bankruptcy forecasts increases with prediction horizons. Moreover, LASSO-selected variables have superior out-of-sample predictive power and outperform (1) those advocated by Campbell et al. (2008) and (2) the distance to default from Merton’s (1974) structural model.     

Employee well-being,firm leverage,and bankruptcy risk

Employees of liquidating firms are likely to lose income and non-pecuniary benefits of working for the firm, which makes bankruptcy costly for employees. This paper examines whether firms take these costs into account when deciding on the optimal amount of leverage. We find that firms with leading track records in employee well-being significantly reduce the probability of bankruptcy by operating with lower debt ratios. Moreover, we observe that firms with better employee track records have better credit ratings, even when we control for differences in firm leverage.     

Are the bankrupt skies the friendliest?

We use data from the US airline industry to investigate whether firms that are under bankruptcy protection, as well as these firm's product market rivals, change the quality of the products they offer. We measure the quality of the services offered by a carrier using flight cancelations and delays, and the age of the aircraft used by the carrier. We find that delays and cancelations are less frequent during bankruptcy filings but return to their pre-bankruptcy levels once the bankrupt firm emerges from bankruptcy. We also find that firms use Chapter 11 filings to permanently reduce the age of their fleet. We do not find evidence of statistically and economically significant changes by the airline's competitors along any of the dimensions above.     

Factors Affecting the Probability of Bankruptcy: A Managerial Decision Based Approach

The majority of classification models developed have used a pool of financial ratios combined with statistical variable selection techniques to maximize the accuracy of the classifier constructed. Rather than follow this approach, this article seeks to provide an explicit economic basis for the selection of variables for inclusion in bankruptcy models. This search to develop an economic theory of bankruptcy augments the existing bankruptcy prediction literature. Variables which occur in bankruptcy probability expressions derived from the solution of a stochastic optimizing model of firm behaviour are 'proxied' by variables constructed from financial statement data. The random nature of the lifetime of a single firm provides the rationale for the use of duration or hazard-based statistical methods in the validation of the derived bankruptcy probability expressions. Results of the validation exercise confirm that the majority of variables included in the empirical hazard formulation behave in a way that is consistent with the model of the firm. The results highlight the need for developments in the measurement of earnings dispersion.     

Do corporate payouts signal going-concern risk for auditors? Evidence from audit reports for companies in financial distress

We examine the association between payout policy changes and going-concern decisions for financially distressed clients. Extant auditing standards indicate that payout reductions, which offer a prospect of short-term cash relief, can potentially mitigate going-concern uncertainty, whereas economic theory suggests payout decreases (increases) convey mixed but mostly negative (positive) signals about a company’s future financial status. We find that, compared with a bankruptcy prediction model over short (not to exceed 1 year) and long (2–3 years) horizons, auditors seem to significantly underreact to payout decreases (i.e., negative signals) but react appropriately to payout increases (i.e., positive signals) in their going-concern decisions. Moreover, auditors are three times more likely to make Type II misclassification errors in payout-decreasing firms than in payout-increasing and no-change firms. We also find that auditors take longer to determine the appropriate opinion for clients with payout changes, especially for those who cut their payouts. Overall, our findings suggest that auditors respond differently to positive and negative signals about companies’ future prospects, reflecting the mixed nature of payout decreases relative to payout increases and the professional standards’ emphasis on the prospect of short-term cash relief from payout reductions.     

Mean Reversion in Profitability for Non‐listed Firms

The presence of mean reversion in profitability at the firm level is important for valuation and prediction of growth and earnings. We investigate the mean reversion in accounting profitability for Norwegian non‐listed firms for the period 1988–2006. We find a mean reversion rate of about 0.44 per year. This is higher than found in other studies. We also find that small firms have a higher mean reversion rate than large firms. Our results should have important practical implications for the difficult task of valuing non‐listed firms. Previously, price‐to‐book ratios have been used to investigate changes in profitability over time for listed firms. We examine bankruptcy risk as an alternative variable for unlisted firms. We find that bankruptcy risk may help explain changes in profitability, but the results are not as strong as found in previous work.     

Bankruptcy risk induced by career concerns of regulators

We introduce a model in which a regulator employs mechanism design to embed her human capital beta signal(s) in a firm’s capital structure. This can enhance her compensation at the firm, and the value of her contract with the firm in the form of an executive stock option. We prove that the agency cost of this revolving door behavior increases the firm’s financial leverage, bankruptcy risk, and affects estimation of firm value at risk (VaR).     

Effect of corporate tax avoidance activities on firm bankruptcy risk

Corporate tax avoidance has been shown to raise the cost of bank debt and lower credit and bond ratings. However, it is unclear whether tax avoidance actually increases a firm’s bankruptcy risk or whether it is just viewed negatively by banks and rating agencies. We find that firms engaging in tax avoidance and firms that are thinly capitalized face higher bankruptcy risk. To account for endogeneity and functional form misspecification, we verify our results using instrumental variable and propensity score matching methods. Our findings are consistent with the view that tax avoidance is a risk-enhancing activity.     

Business environment and the incorporation decision

Using firm-level data from 52 countries we investigate how a country’s institutions and business environment affect firm’s organizational choices and what impact the organizational form has on access to finance and growth. We find that businesses are more likely to choose the corporate form in countries with developed financial sectors and efficient legal systems, strong shareholder and creditor rights, low regulatory burdens and corporate taxes and efficient bankruptcy processes. Corporations report fewer financing, legal and regulatory obstacles than unincorporated firms and this advantage is greater in countries with more developed institutions and favourable business environments. We do find some evidence of higher growth of incorporated businesses in countries with good financial and legal institutions.     

Human capital costs,firm leverage,and unemployment rates

Because bankruptcy is costly for employees, theoretical studies argue that firms with higher leverage have to pay their employees higher wages. In this paper we empirically test this prediction. We find that firm leverage is positively related to the wages of employees, both in the United States and in the Netherlands. In the United States, the positive relation between wages and leverage is strongest in the 21st century, which is a period that also shows a positive relation between wages and unemployment rates. We conclude that the human capital costs of bankruptcy are an important disadvantage of debt.     

Corporate Bankruptcy in Korea: Only the Strong Survive?

We analyze whether the build‐up of financial vulnerabilities led listed Korean companies to bankruptcy. We find that pre‐crisis leverage is systematically high for both poor performing/slow growing firms and for profitable/fast‐growing firms. Pre‐crisis leverage raises the probability of bankruptcy, which is lower for firms: (1) relying more on (renegotiable) bank credit; (2) with less inter‐firm debt; and (3) having higher interest coverage ratios. Finally, none of these liquidity variables help predict bankruptcies for chaebol‐firms, suggesting that liquidity constraints are more stringent for non‐chaebol. Thus, in a systemic crisis it is not only the strong/healthy that survive.     

The effectiveness of currency-hedging techniques over multiple return horizons for foreign-denominated debt issuers

In this study we use a unique dataset to examine the effectiveness of hedging techniques in reducing currency risk across different time horizons. Our primary results indicate that issuers of foreign-denominated debt (FDD) effectively use foreign exchange derivatives (FXD) and geographic sales-to-asset alignment to reduce currency exposure. We measure currency exposure, over multiple return horizons, as the sensitivity of firm’s stock returns to changes in the exchange rate while controlling for market’s return. In particular, we find that the use of foreign exchange derivatives is effective in the short-term horizon and the geographic sales-to-asset alignment is effective in the intermediate and longer-term horizons. When we control for foreign sales and for firm size, the primary results are robust. Our study contributes to the literature by providing evidence of the time period when the use foreign exchange derivatives and the geographic sales-to-asset alignment is most effective for issuers of foreign-denominated debt.     

Bankruptcy Litigation and Relationship Banking

This paper analyzes how bankruptcy litigation affects the value of relationship banking. In our model, bankruptcy courts may make type 1 errors, i.e., they may declare that an insolvent firm is solvent; and they may make type 2 errors, i.e., they may declare that a solvent firm is insolvent. Our model provides four results. First, the cost of bank debt decreases when the probability that bankruptcy courts make type 2 errors increases. Second, the value of relationship banking increases when the probability that bankruptcy courts make type 1 errors increases. Third, the cost of credit intermediation decreases when the probability that bankruptcy courts make type 2 errors increases. Fourth, the diversification mechanism does not fully solve the delegated monitoring problem.     

Who gets credit after bankruptcy and why? An information channel

Conventional wisdom holds that individuals find it difficult to obtain new credit post-bankruptcy. Using credit bureau data, we test this hypothesis and show that more than 90% of bankrupt individuals receive credit shortly after filing. Individuals with good credit history prior to filing have reduced credit availability after bankruptcy while those with ex-ante low credit quality receive more credit. We show that credit supplied to low quality individuals is severely curtailed during the financial crisis. We also find that the default probability on new debt increases after bankruptcy, especially among individuals with high ex-ante credit score. These findings are consistent with an information channel, in which bankruptcy reveals new information about a borrower’s credit quality.