Predicting Corporate Failure Using a Neural Network Approach

This paper investigates the performance of Artificial Neural Networks for the classification and subsequent prediction of business entities into failed and non-failed classes. Two techniques, back-propagation and Optimal Estimation Theory (OET), are used to train the neural networks to predict bankruptcy filings. The data are drawn from Compustat data tapes representing a cross-section of industries. The results obtained with the neural networks are compared with other well-known bankruptcy prediction techniques such as discriminant analysis, probit and logit, as well as against benchmarks provided by directly applying the bankruptcy prediction models developed by Altman (1968) and Ohlson (1980) to our data set. We control the degree of ‘disproportionate sampling’ by creating ‘training’ and ‘testing’ populations with proportions of bankrupt firms ranging from 1% to 50%. For each population, we apply each technique 50 times to determine stable accuracy rates in terms of Type I, Type II and Total Error. We show that the performance of various classification techniques, in terms of their classification errors, depends on the proportions of bankrupt firms in the training and testing data sets, the variables used in the models, and assumptions about the relative costs of Type I and Type II errors. The neural network solutions do not achieve the ‘magical’ results that literature in this field often promises, although there are notable 'pockets' of superior performance by the neural networks, depending on particular combinations of proportions of bankrupt firms in training and testing data sets and assumptions about the relative costs of Type I and Type II errors. However, since we tested only one architecture for the neural network, it will be necessary to investigate potential improvements in neural network performance through systematic changes in neural network architecture.     

Corporate Bankruptcy and Managers' Self-Serving Behavior

We investigate whether insiders of bankrupt firms hold less stock or reduce their stockholdings compared to what we observed for insiders of similar firms that do not go bankrupt. We find little evidence of such time-series and cross-sectional differences in spite of the fact that the stock value of bankrupt firms falls by more than ninety percent in the five years preceding bankruptcy. One implication of our results is that the amount of stock owned and the magnitude of the trades undertaken by corporate insiders of both bankrupt and nonbankrupt firms appear to provide no information about firm value.     

Corporate bankruptcy prediction in Japan

The purpose of this study is to highlight the financial characteristics of failed firms in Japan, and to construct corporate bankruptcy prediction models with greater prediction accuracy. Our principal component analysis indicated that failed firms in Japan could be classified into two groups: a group having negative financial structures and a group having a declining flow of funds. Additionally, they can be classified into two other different categories of groups: one whose financial position during three years before shows a ‘V’ shape and another group that shows a ‘XXX’ shape.Our discriminant analysis indicated that improved prediction accuracy could be obtained by using, as predictor variables, both ratios and absolute amounts based on cash base financial statement data three years before failure. This data was adjusted to properly reflect the exceptions, reservations, and qualifications appearing in the audit reports and those based on accrual base financial statement data.     

Evidence from impending bankrupt firms that long horizon institutional investors are informed about future firm value

A key assumption in many accounting and finance studies is that long horizon institutional investors are informed shareholders. Yet past empirical research finds no evidence that these institutions anticipate major corporate events, including earnings-based events. I find that long horizon institutions are better informed in that they sell more shares of impending bankrupt firms than of matched distress firms at least one quarter ahead of bankruptcy. Share sales are greater in impending bankrupt firms whose shareholders ultimately lose all of their equity. In additional analyses, I document greater share sales by long horizon institutions with supposedly superior information processing abilities and/or access to corporate management. Share sales are significantly less in the post Regulation FD era. Overall, my findings support the validity of the common assumption that long horizon institutions are informed. Regulation FD appears to mitigate (but not eliminate) their information advantage.     

Global financial crisis and emerging issues for public policy

We study the effect of relationship lending on small firms' failure probability using a uniquely rich data set comprised of information on individual loans of a large number of small firms in Colombia. We control for firm-specific variables and find that small firms involved in long-term liaisons with commercial banks have a significantly lower probability of becoming bankrupt than otherwise identical firms not involved in a long-term credit relationship. We also find that small firms with multiple banking relationships face a lower failure hazard than otherwise identical firms involved in a unique long-term relationship.     

Excess cash holdings and shareholder value

We examine the determinants of corporate cash holdings in Australia and the impact on shareholder wealth of holding excess cash. Our results show that a trade‐off model best explains the level of a firm’s cash holdings in Australia. We find that ‘transitory’ excess cash firms earn significantly higher risk‐adjusted returns compared to ‘persistent’ excess cash firms, suggesting that the market penalises firms that hoard cash. The marginal value of cash also declines with larger cash balances, and the longer firms hold on to excess cash. The results are consistent with agency costs associated with persistence in excess cash holdings.     

Disequilibrium in the UK corporate loan market

In the last decade, a debate has resurfaced about whether financial constraints stemming from asymmetric information and incentive problems play an important role in propagating monetary policy shocks. This paper investigates the monetary transmission mechanism in the UK and its impact on the availability of bank credit to small and medium size firms.The empirical specification is based on a disequilibrium model that allows for the possibility of transitory credit rationing. Sample firms are classified endogenously into ‘borrowing constrained’ and ‘borrowing unconstrained’. The analysis of credit rationing takes into account not only firm specific variables, but also important macroeconomic factors such as the prevailing monetary conditions and the stage of the business cycle.We find that (i) firms’ assets play an important role as collateral in mitigating borrowing constraints; (ii) during periods of tight monetary conditions corporate demand for bank credit increases, whereas the supply of bank loans is reduced; (iii) to avoid bank credit rationing smaller companies increase their reliance on interfirm credit; (iv) the proportion of borrowing constrained firms is significantly higher during the recession years of the early 1990s than at other times.     

Performance evaluation of bankruptcy prediction models: An orientation-free super-efficiency DEA-based framework

Prediction of corporate failure is one of the major activities in auditing firms risks and uncertainties. The design of reliable models to predict bankruptcy is crucial for many decision making processes. Although a large number of models have been designed to predict bankruptcy, the relative performance evaluation of competing prediction models remains an exercise that is unidimensional in nature, which often leads to reporting conflicting results. In this research, we overcome this methodological issue by proposing an orientation-free super-efficiency data envelopment analysis model as a multi-criteria assessment framework. Furthermore, we perform an exhaustive comparative analysis of the most popular bankruptcy modeling frameworks for UK data including our own models. In addition, we address two important research questions; namely, do some modeling frameworks perform better than others by design? and to what extent the choice and/or the design of explanatory variables and their nature affect the performance of modeling frameworks?, and report on our findings.     

Cash-flow shortage as an endogenous bankruptcy reason

This paper develops a simple model for a leveraged firm and endogenizes the firm’s bankruptcy point by assuming that equity issuance is costly. Equity-issuance costs reflect the difficulties in issuing new equity for firms that are close to financial distress. The resulting model captures cash-flow shortage as a reason to go bankrupt, though the equity value is positive. I analyze the optimal bankruptcy point as well as corporate bond prices and yield spreads for various levels of equity-issuance costs in order to study the impact of different liquidity constraints. Finally, I discuss the consequences on optimal capital structure.     

Bankruptcy,absolute priority,and the pricing of risky debt claims

In practice, there are substantial deviations from the doctrine of ‘absolute priority’, which governs the rights of the firm's claimholders in the event of bankruptcy. To determine whether or not the possibility of such deviations is reflected in the prices of the firm's securities, this study examines the risk and return characteristics of financial claims against firms in court-supervised bankruptcy proceedings. Debt claims against bankrupt firms are indeed ‘risky’, exhibiting levels of systematic risk similar to that of common stocks in general. While some of the findings are anomalous, the data are generally consistent with the view that the capital market ‘properly’ prices risky debt claims to reflect both their risk characteristics and the possibility of departures from the doctrine of absolute priority.     

Stock market reactions to adverse ESG disclosure via media channels

This study examines the value relevance of corporate reputation risks (CRR) from adverse media coverage of environmental, social and governance (ESG) issues on stock performance at the firm level. Empirical results advance signalling theory and resource-based view by providing evidence that corporate reputation is considered a valuable intangible asset by investors and adverse ESG disclosure via media channels have a significant and negative impact on firm valuation. The research is extended using various factors and indicates that heightened CRR have a substantially negative corollary effect on stock price of smaller and less liquid firms that are typically not S&P500 constituents. Further analysis using industry classifications reveals that stock performance of companies in the ‘sin’ triumvirate (i.e., alcohol, tobacco, and gaming) is not significantly affected by negative ESG media coverage. Instead, firms in candy & soda, steel works, banking, and insurance industries are the most susceptible to investors' repercussion from undesirable media spotlight. These findings provide new insights and indicate that beyond the type and delivery method of ESG disclosures, firm characteristics, corporate reputation status and industry explain differences in investors' reaction to ‘bad’ news.     

The determinants of auditor switching from the perspective of corporate governance in China

This paper reports the association between firms' internal corporate governance mechanisms and their auditor switch decisions in the Chinese context. We identify two types of auditor switch, namely switching to a larger auditor and switching to a smaller auditor. Three variables are used to proxy for firms' internal corporate governance mechanism, including the ownership concentration (shareholding by the largest owner), the effectiveness of supervisory board (SB), and the duality of chairman of board of directors (CBoD) and CEO. We regressed the internal corporate governance variables over firms' audit switching types during a specific period of 2001-2004 when a bear market continued in China. The empirical results demonstrate that firms with larger controlling owners or in which CBoD and CEO are held by the same person are more likely to switch to a smaller auditor rather than to a larger one. However, the effect of the SB variable does not have a significant impact on auditor switching decisions. In general, the study findings suggest that firms with weak internal corporate governance mechanism tend to switch to smaller or more pliable auditors in order to sustain the opaqueness gains derived from weak corporate governance. On the other hand, with the improvement of corporate government, firms should be more likely to choose large (high-quality) auditors in making auditor switching decisions.     

Corporate investment during the financial crisis: Evidence from China

China's growth model suggests that the 2008 financial crisis may have affected the Chinese economy differently from what one observes in mature market economies. In this paper, we examine how Chinese corporate investment responded to the financial crisis by using 1689 listed nonfinancial firms during Q12006–Q32010. We document that (1) the overall impact of the financial crisis on Chinese corporate investment is negative; (2) among three channels conveying the effect of the financial crisis, namely, the demand channel, the financial constraints channel, and the uncertainty channel, the demand channel dominates; (3) financial assets held by a nonfinancial firm are important in explaining the firm's fixed investment behaviour; (4) as compared to non-state firms, state-controlled firms are less affected by the financial crisis and more active in engaging in financial assets investment; and (5) foreign ownership can be seen as a buffer against the negative effect of the financial crisis and foreign-involved Chinese firms are less active in financial assets investment as compared to domestic firms.     

When is good news really good news?

The impression management literature suggests that managers often resort to biased disclosures. However, there is little systematic evidence on what types of strategies management uses to achieve this bias. Do managers simply lie? Or, do they use more subtle ways of introducing positive bias into corporate narratives, such as selecting specific information items which result in a more positive impression (‘selectivity’) or by keeping their narratives vague and general (‘vagueness’)? In order to differentiate between the two scenarios, I re-examine the positive forward-looking statements examined by Schleicher and Walker (2010) and compare, across firms with improving and deteriorating financial performance, the managerial choices made in relation to eight forecast attributes.

I make two observations. First, there are significant differences in the characteristics of good- and bad-news firms’ positive statements. In particular, bad-news firms’ positive statements involve more non-specific time horizons, more segmental forecasts, and more references to conditions and aims and objectives, but fewer directional forecasts, fewer numbers, and fewer reinforcing qualifiers. Second, the identified differences in good- and bad-news firms’ positive statements can be exploited for classification purposes: including into a classification model additional regressors that measure a positive forward-looking statement's level of selectivity and vagueness significantly increases the model's ability to separate firms with improving financial performance from firms with deteriorating financial performance. Overall, my results are consistent with (a) impression management operating predominantly through selectivity and vagueness and (b) selectivity and vagueness being an important signal for future financial performance.     

Is Chapter 11 Efficient?

The efficiency of the Chapter 11 bankruptcy process is examined by estimating the impact of Chapter 11 filings on the operating performance of bankrupt firms. We control for firm‐level heterogeneity in prefiling characteristics using matching methods to select benchmark firms comparable to filing firms. We compare bankrupt firms’ operating performances with those of matched nonbankrupt firms. Our results challenge the contention that Chapter 11 is an inefficient, debtor‐friendly mechanism that rehabilitates economically nonviable firms. We demonstrate that firms that file under Chapter 11 perform no worse and, if anything, better than comparable nonfiling firms.     

Association between independent non-executive directors, family control and financial disclosures in Hong Kong

Our study examines whether comprehensive financial disclosures, used as a proxy for corporate board’s responsiveness, are positively associated with the proportion of independent non-executive directors (INDs) on corporate boards, and whether family control of the firm has an impact on this association. The findings suggest that the ratio of INDs to the total number of directors on corporate boards is positively associated with the comprehensiveness of financial disclosures, and this association appears to be weaker for family controlled firms compared to non-family controlled firms.     

The impact of board quality and nomination committee on corporate bankruptcy

This paper examines the effects of board quality on the relationship between corporate bankruptcy and nomination committee effectiveness. We argue that the proportion of outside directors, a proxy of board quality, arguably captures the extent of board control and resources. Based on dataset from 1835 firm-year observations for 98 bankrupt and 269 non-bankrupt UK listed non-financial firms between 1994 and 2011 and using the agency and resource dependence theories, we predict and find that nomination committee effectiveness negatively affects corporate bankruptcy and that board quality mitigates the negative effects. The results lend support to the notion that firms benefit from board quality in terms of outside directors' ability to monitor CEO on behalf of shareholders and also provide advice, counsel and legitimacy to the firm. This study extends the present research on corporate bankruptcy by providing evidence on the impact of board quality and nomination committee effectiveness on UK corporate bankruptcy.     

The impact of multi-dimensional corporate transparency on us firms’ credit ratings and cost of capital

This study examines corporate transparency in the US market for a sample of 319 S&P 500 firms. We examine whether a number of disparate measures of corporate transparency used by other researchers are distinct, cohere as measures of a single factor of corporate transparency, or capture multiple different dimensions. Next, we begin to examine the impact of corporate transparency, conceived in the broadest sense, and not limited to financial reporting, on US firms. We develop a model of corporate transparency based on a broad definition and framework proposed by Bushman, Piotrowski and Smith, which we extend in several ways, and then study the effect of corporate transparency on cost of debt, credit rating, and cost of equity. First, we find that corporate transparency is neither a unitary concept nor merely an ambiguous term for multiple distinct concepts: factor analysis of ten corporate transparency variables identifies four independent underlying dimensions: public disclosure information, intermediary information, earnings quality information and insider information. Second, we find that corporate transparency has significant power to explain cross-sectional variation in credit rating and cost of capital. More specifically, (i) credit rating, cost of debt, and beta are significantly associated with disclosure information transparency; (ii) credit rating, cost of equity, and beta are significantly associated with intermediary information transparency; and (iii) cost of equity and beta are significantly associated with insider information transparency. Our findings offer a more comprehensive evaluation of corporate transparency than prior studies, and we demonstrate direct economic implications for both US firms and markets.     

Bankruptcy,overlapping directors,and bank loan pricing

Using a sample of loan facilities borrowed by firms that share directors with bankrupt firms, this study investigates whether the overlapping directors are a transmission channel of the bankruptcy contagion effect in the bank loan market and, if so, what the underlying mechanism is. We find that firms are charged higher loan spreads in the period following the bankruptcy filing of a firm with a common director and that overlapping directors are a relevant channel for the bankruptcy contagion effect, in addition to other channels identified in literature. We also find that the negative contagion effect on loan pricing is most likely driven by the overlapping directors' reputation loss due to their involvement in bankruptcy events, and not by competing hypotheses, such as director distraction and director career concern/experience. Further analyses reveal that the adverse contagion impact on loan spreads is more pronounced when overlapping directors have greater influence over corporate policies or when their reputation is more seriously damaged. Meanwhile, the contagion effect is mitigated when interlocked firms have a higher-quality board. These results further support our evidence of the director reputation loss hypothesis. We strengthen the identification strategy to establish causality. In sum, our study identifies common directors as a channel of bankruptcy contagion effects on loan pricing and director reputation loss as an underlying mechanism.     

Security Analysts’ Site Visits and Stock Price Synchronicity

This paper investigates how security analysts’ corporate site visits impact listed firms’ stock-price informativeness. Examining a sample of security analysts’ visits to Chinese listed firms from 2010 to 2019, we find that security analysts incorporate firm-specific information into share prices through site visits, significantly reducing the visited firms’ stock price synchronicity. This finding is robust to an alternative measure of stock price informativeness and a two-stage least-squares approach using the introduction of high-speed rail as the instrumental variable. We also find that the impact of analysts’ site visits on firms’ stock price synchronicity is more pronounced for firms with lower information disclosure quality and poor corporate governance than for other firms. Further analysis on firm characteristics documents that this effect is stronger for large-size firms, firms in the manufacturing industry, and state-owned enterprises.