Abstract

To what extent does the collapse of a commercial bank spread contagion across cryptocurrency markets? How do markets behave around bankruptcy if digital assets remain stuck within the bank and cannot be withdrawn? We use a BEKK model to examine contagion effects across major digital assets during the Silicon Valley Bank (SVB) collapse period in early March 2023. We find evidence of contagion across major stablecoins and Bitcoin. We also examine the price action when nearly all withdrawals at SVB were prohibited. We find substantial abnormal movements in stablecoin cumulative returns and traded volumes, indicating a “flight to safety” from less to more authoritative and trusted stablecoins. The implications for practitioners and policymakers are discussed.

Introduction

Cryptocurrencies are gradually gaining traction in trading and becoming a notable segment of the financial landscape, providing new investment avenues. However, their inherent volatility continues to hinder their broader acceptance. Among cryptocurrencies, a subset known as stablecoins, which is intended to maintain a “stable” peg to a reference currency, is crucial to the market as it allows traders to work around the volatility issue and keep money in a form equal to US dollars (De Blasis et al., 2023). Access to those digital assets is typically given by new financial technologies (also called Fintech) such as digital wallets or Centralized cryptocurrency Exchanges (CEXs), with the aim of decentralizing finance through innovative offers of financial services. However, due to the high costs associated with the trading of cryptocurrencies in US dollars, the impossibility of using US dollars on cryptocurrency exchanges,1 and the simplicity and speed of transferring those virtual assets between exchanges, stablecoins occasionally trade at a premium2 to the underlying asset they imitate (De Blasis et al., 2023). The recent bankruptcy of Silicon Valley Bank (SVB), the largest commercial bank for nearly half of all venture-backed tech startups in Silicon Valley,3 in March 2023 tied up $3.3 out of $40 billions of stablecoins in its financial statement, making this exogenous shock a significant event to study.

Thus, the aim of this research is to examine the impact of bank failure on the stability of stablecoin markets. Specifically, we examine the question of whether halting access to digital assets through the bankruptcy procedure of a traditional finance institution spills over contagion effects across those innovative instruments. Answering this question is important as the March 2023 collapse of SVB, together with the ensuing instability in a number of major stablecoins, demonstrated the importance of the interlinks between traditional and digital financial systems. The results of this research are therefore relevant to policymakers in both the traditional and modern financial systems interested in avoiding the risk of financial contagion, as well as investors willing to defend their savings and balance their portfolios against market uncertainty.

Conclusion

This study examined the extent to which the collapse of a commercial bank spread contagion across cryptocurrency markets and how those markets behave around the failure when asset withdrawals are halted. We use a BEKK-GARCH multivariate model to test for financial contagion around the Silicon Valley Bank collapse across multiple digital assets. Arguably, this is the second-largest bank run after the global financial crises experienced in history. We find evidence of volatility spillover effects across major stablecoins and Bitcoin, with a “flight to safety” from less to more stable virtual currencies.

Hence, this study provides important evidence of the ways the traditional and cryptocurrency financial systems are interlinked. Arguably, it links centralized and decentralized finance, showing the ways in which spillovers can occur between what seem to be separate realms of finance. Academics, practitioners, and policymakers alike interested in destabilizing risk in the digital finance ecosystem should pay attention to the connections with traditional finance and thus to the results of this research. A limitation of this study is that it focuses on the most liquid stablecoins and cryptocurrencies. Future studies may, therefore, further explore the links between centralized and decentralized finance, including stablecoins and other means of digital value exchange.

Abstract

Corporate bankruptcy and financial distress prediction is a topic of interest for a variety of stakeholders, including businesses, financial institutions, investors, regulatory bodies, auditors, and academics. Various statistical and artificial intelligence methodologies have been devised to produce more accurate predictions. As more researchers are now focusing on this growing field of interest, this paper provides an up-to-date comprehensive survey, classification, and critical analysis of the literature on corporate bankruptcy and financial distress predictions, including definitions of bankruptcy and financial distress, prediction methodologies and models, data pre-processing, feature selection, model implementation, performance criteria and their measures for assessing the performance of classifiers or prediction models, and methodologies for the performance evaluation of prediction models. Finally, a critical analysis of the surveyed literature is provided to inspire possible future research directions.

Introduction

Corporate bankruptcy prediction (BP) and financial distress prediction (FDP) have been a field of investigation by researchers around the world for nearly a half century. However, increasing attention has been paid to this evergreen subject since the Global Financial Crisis of 2008 (Alaminos et al., 2016), as predictions can have a significant impact on the decisions and returns of various stakeholders (Alam et al., 2021).

Corporate bankruptcy and financial distress events are not desirable (Gordon, 1971), as their legal and financial costs are prohibitive (Weiss, 1990). In addition, these events increase the expected costs for financial institutions, such as banks, to hedge against the risk of these events happening (Alnassar & Chin, 2015). To reduce exposure to risk and catch early warning signs, stakeholders including investors, bankers and governments are proactively looking for solutions to effectively analyze and predict corporate bankruptcy and financial distress events. The earliest study in terms of modern bankruptcy prediction can be traced back to 1932 when Fitzpatrick (1932) presented a successful way to distinguish between failed and healthy companies by analyzing 20 pairs of companies’ accounting ratios. Since the 1960s, several accounting-based statistical and probabilistic models have been proposed to predict corporate bankruptcy and financial distress (e.g., Beaver, 1966; Altman, 1968; Ohlson, 1980; Zmijewski, 1984; Zavgren, 1985). In addition, several studies focused on the identification of bankruptcy and financial distress drivers2 (e.g., Altman, 1968; Liang et al., 2016; Tobback et al., 2017; Tinoco et al., 2018; Mai et al., 2019), on one hand, and others focused on the design of methodologies for selecting such drivers (e.g., Tsai, 2009; Lin et al., 2014; Tian & Yu, 2017; Uthayakumar et al., 2020; Kou et al., 2021), on the other hand. Furthermore, a stream of literature investigated the causes of the under-performance of bankruptcy and financial distress prediction models such as data sample imbalance (e.g., Le et al., 2018; Veganzones & Séverin, 2018; Zoričák et al., 2020, Shen et al., 2020). In the era of the rise of machine learning and artificial intelligence technologies, the prediction of bankruptcy and financial distress has been lifted to a whole new level, as more and more new methodologies managed to improve prediction accuracy through the design features of new models (e.g., Ouenniche & Tone, 2017; Ouenniche et al., 2018a, 2018b, 2018c, 2019; Hosaka, 2019; Matin et al., 2019; Yuan et al., 2022), the type of implementation decisions such as over-sampling and under-sampling (e.g., Le et al., 2018; Sun et al., 2020; Shen et al., 2020; Du et al., 2020), and bagging and boosting (e.g., West et al., 2005; Zięba et al., 2016; Jones & Wang, 2019; Chen et al., 2020). On the other hand, another stream of research focused on reducing the computational requirements of prediction models or methods using parallel implementations that take advantage of today’s powerful computers, or to be more specific, GPUs (e.g., Huang & Yen, 2019; Le et al., 2019).

Conclusions and future research directions

The topic of corporate bankruptcy and financial distress prediction has attracted the attention of many researchers for several decades and continues to evolve with more and more advanced prediction methodologies and issue fixation solutions being proposed. This study contributes to this domain of research by providing an up-to-date state-of-the-art review, classification and critical analysis of the current BP and FDP literature, where six major research streams are identified and discussed, namely, the definition and coding of bankruptcy and financial distress; design of new prediction models/classifiers or new application of existing ones; design of new drivers or evaluation of existing ones; design or evaluation of feature selection methods; design of methodologies for the performance evaluation of prediction models; and issues affecting the performance of prediction models and related solutions. By painting the landscape of research and analyzing each research stream, this study would serve as a guide for researchers who intend to explore this field of study and/or contribute to its development.

Our analysis of the surveyed papers revealed a clear trend in terms of prediction methodologies of bankruptcy and financial distress, where more emphasis is put on advanced machine learning and artificial intelligence models such as ensemble learning models. Ensemble learning models and other methodological advances, such as DEA- and MCDA-based prediction methods, have contributed to further pushing the boundaries of research. Overall, it is fair to conclude that there is no single methodology that is better than the others, as each methodology has its strengths and weaknesses, some of which are design-related, and others are implementation decisions related, which have made the design and implementation of new ensemble models more popular.

1- Introduction

2- Literature review of machine-learning methods for bankruptcy prediction

3- Data

4- Preliminary statistical analysis

5- Analysis of feature importance

6- Bankruptcy prediction

7- Conclusions

References

Abstract

Bankruptcy prediction is still important topic receiving notable attention. Information about an imminent bankruptcy threat is a crucial aspect of the decision-making process of managers, financial institutions, and government agencies. In this paper, we utilize a newly acquired dataset comprising financial parameters derived from the annual reports of small- and medium-sized companies. The data, which reveal the true ratio between bankrupt and non-bankrupt companies, are severely imbalanced and only contain a small fraction of bankrupt companies. Our solution to overcome this challenging scenario of imbalanced learning was to adopt three oneclass classification methods: a least-squares approach to anomaly detection, an isolation forest, and one-class support vector machines for comparison with conventional support vector machines. We provide a comprehensive analysis of the financial attributes and identify those that are most relevant to bankruptcy prediction. The highest prediction performance in terms of the geometric mean score is 91%. The results are validated on two datasets from the manufacturing and construction industries.

Introduction

The recent financial crisis showed the increasing vulnerability of firms involved in complex business relations, relations with financial institutions, obligations toward tax agencies, etc. The threat of financial contagion is rising with the growing complexity of the economy. The latter experience brought evidence of the fragile financial stability of numerous firms. These companies are prone to turbulent financial shocks with their origins in the external environment. Even though many studies have been devoted to bankruptcy prediction, a general methodology that would enable a firm to identify business partners in financial distress has not yet been proposed. The uniqueness of the bankruptcy prediction problem can be found in the nature of the data that are the subject of analysis. The majority of studies are based on a variety of financial ratios that are derived from annual financial statements. The annual financial statements usually consist of two documents –- the balance sheet and income statement. The first contains information regarding the assets, liabilities, and owners’ equity, whereas the income statement considers the costs, revenues, and eventual profit or loss. Because the frequency of data is annual, the information in the financial ratios is condensed and may conceal important fluctuations between two reporting periods. The quality of data is usually determined by the type of companies included in the analysis. In general, larger firms or firms listed on the stock exchange are more likely to disclose more information (Firth, 1979), thereby allowing a more meaningful analysis of their current financial condition.

Debt-to-equity conversion is a commonly used legal tool for financial restructuring of ailing businesses, yet systematic evidence on the consequences of debt-to-equity conversion for firm outcomes is scarce. Drawing on a novel dataset of bankruptcy reorganizations from Slovenia and exploiting variation in the incidence of debt-to-equity conversion across firms, we provide the first empirical analysis of the role of debt-to-equity conversion in bankruptcy reorganization as a determinant of post-bankruptcy firm survival. To address endogeneity concerns, we use a plethora of controls and fixed effects, quantify the sensitivity of our estimates to selection on unobservables, and rely on instrumental variable methods. We find that debt-to-equity conversion is robustly negatively associated with prospects of post-bankruptcy firm failure. Our findings provide a novel input into ongoing debates about the appropriate design of corporate bankruptcy institutions.

Introduction

Debt-to-equity conversion (DEC, in short) is a frequently utilized legal tool for alleviation of financial distress of firms nearing or experiencing insolvency (see, e.g., Chatterji and Hedges, 2001; Gilson et al., 1990; Gilson, 1990; James, 1995). From the creditors’ standpoint, DEC is often an attractive mode of debtor’s financial restructuring. Without DEC, creditors can often hope to recover only a small fraction of their claims. By reducing the company’s overall debt burden, DEC increases the prospects of claim recovery, even for creditors who choose not to take part in DEC. DEC further allows creditors to obtain control over the ailing business. DEC thereby provides creditors with the upside that the business will recover and that the value of their equity will grow. Institutionally, DEC can take place either during out-of-court financial restructuring or, when enabled by the legislative framework, within in-court bankruptcy reorganization proceedings. Empirical evidence on the effect of DEC on the performance of financially restructured business, however, is scant. Especially little is known about the consequences of bankruptcy-based DEC for post-bankruptcy firm outcomes. The corresponding lack of empirical evidence is unsettling because DEC has been routinely considered, or even advocated, as a mechanism of corporate bankruptcy resolution (Claessens et al., 2001; Hart, 2006; Laryea, 2010; Eric´ and Stosiˇ c, ´ 2015). In theory, DEC is expected to aid post-bankruptcy firm performance for two primary reasons. By lessening the debtor’s debt burden without an accompanying decrease in the firm’s assets, DEC instantaneously strengthens the firm’s balance sheet. An improved debt-to-equity ratio in turn enhances the firm’s borrowing position as well as the firm’s status vis-à-vis the customers, suppliers, and other business partners. In addition, by changing the firm’s ownership structure, DEC holds potential to improve decision-making at a critical juncture for the struggling business. Yet at the same time, DEC is not a guarantee that the business will succeed ex post. The firm’s new owners (creditors who opted for DEC) may come in conflict with the old owners or management, or continue to repeat errors that possibly contributed to the firm’s pre-bankruptcy turmoil.

1- Introduction

2- A general approach for valuing firm’s financial flexibility under default risk and bankruptcy costs

3- A simulation analysis

4- Concluding remarks

References

Abstract

Purpose – The purpose of this paper is to present a model to value leveraged firms in the presence of default risk and bankruptcy costs under a flexible firm’s debt structure.
Design/methodology/approach – The authors assume that the total debt of the firm is a combination of two debt components. The first component is an active debt component which is assumed to be proportional to the firm’s value. The second one is a passive predetermined risk-free debt component. The combination of the two debt categories makes the firm’s capital structure more realistic and allows us to include flexibility into the firm’s debt structure management. The firm’s valuation is performed using the discounted cash flow technique based on the weighted average cost of capital (WACC) method.
Findings – The model can be used to define active debt management strategies that can induce the firm to deviate from its capital structure target in order to preserve debt capacity for future funding needs. The firm’s valuation is performed by using the WACC method and a closed form valuation formula is provided. Such a formula can be used to value costs and benefits of financial flexibility.
Research limitations/implications – The proposed approach provides a good compromise between mathematical complexity and model capability of interpreting the various economic and financial aspects involved in the firm’s debt structure puzzle.
Practical implications – This model offers a realistic approach to practical applications where real financing decisions are characterized by a simultaneous use of these two debt categories. By comparing costs and benefits deriving from using unused debt capacity for future funding needs, the model provides a quantitative support to investigate if financial flexibility can add value to firms.
Originality/value – To the authors knowledge, the approach the authors propose is the first attempt to build a valuation scheme that accounts for firm’s financial flexibility under default risky debt and bankruptcy costs. Including financial flexibility, this model fills an important gap in the literature on this topic.

Introduction

In a seminal paper, Modigliani and Miller (1959) showed that the use of debt financing can increase the market value of a firm as a consequence of the debt tax shield. However, the presence of debt in the capital structure can lead the firm to incur into default and distress costs (Warner, 1977; Altman, 1984; Opler and Titman, 1994). The valuation of distress costs is therefore an important task in financing decisions making (Oded and Michel, 2007; Lahmann et al., 2017). The impact of default costs on the firm’s value has been deeply investigated in the literature. Andrade and Kaplan (1998) estimate losses on the order of 10–23 percent of the predistress firm value. Glover (2016) confirms that the average cost among defaulted firms is about 25 percent of the predistress firm value.

1- Data

2- Experimental design, materials, and methods

References

Abstract

Bankruptcy prediction is a long-standing issue that receives significant attention of academic researchers and industry practitioners. Most of the papers on bankruptcy prediction focus on companies that are listed on the stock market, and there are only limited data for the rest of the companies. These companies, not indexed at any stock market, represent a significant part of the economy. The presented dataset consists of financial ratios of Slovak companies. There are 21 distinctive financial ratios which are available for three consecutive years prior to evaluation year in which companies may have filed for bankruptcy or not. The companies come from four different industries – agriculture, construction, manufacture, retail. We provide data for four consecutive years 2013–2016 for each industry. All companies are categorized as small-medium enterprises according to EU classification. Prediction performance results on this dataset are published in the research paper “Bankruptcy prediction for small- and medium-sized companies using severely imbalanced datasets” (Zoričák et al., 2019).

Data

The dataset is accessible on Data Mendeley [2] and provides financial ratios of limited liability companies. There are three possible views on the data as depicted in Table 1: 1) Companies are divided into four different industries: agriculture, construction, manufacture, and retail, 2) Four different evaluation years are considered: 2013, 2014, 2015 and 2016 3) Two classes are defined (bankrupt (B) and non-bankrupt (NB)) for each evaluation year and industry Each company is characterized by 21 financial ratios listed in Table 2. These are provided for three consecutive years prior to the evaluation year. The distribution of missing values is visualized in a heatmap in the Fig. 1. Missing values for individual datasets are displayed one, two and three years prior to evaluation. The most missing values are for the financial variable Labor Productivity (LP). From the industry perspective, the most missing values are for retail with the year of evaluation 2016.

Introduction

This paper analyzes whether firms in financial distress impose indirect costs to their direct competitors and to the real economy by affecting the investment decisions of other firms in the industry. The analysis builds on previous findings that show that when some firms in the industry have financial difficulties, the costs of external financing to rivals increase (Lang and Stulz, 1992; Jorion and Zhang, 2007; Benmelech and Bergman, 2011; Hertzel and Officer, 2012). In principle, the higher financing costs that follow a distress in the industry could reduce investment by affecting the competitors’ ability to obtain sufficient funds. 1 However, a competitor facing financial difficulties could facilitate predation by other firms in the industry, who could exploit their rivals’ weaknesses and increase investment to obtain a higher market share (Fudenberg and Tirole, 1986; Bolton and Scharfstein, 1990; Opler and Titman, 1994). 2 The main objective of this paper is to examine whether competitors of the distressed firms are able to exploit the opportunity to increase their market share, in spite of the potentially higher costs of obtaining finance, or whether the increase in financing costs more than offsets the potential benefits of increasing investment in market share. In addition, the paper seeks to identify the characteristics of the firms that benefit most from their rivals’ financial weaknesses. The main analysis in this paper explores whether the higher financing costs associated with the financial distress of a competitor affect the real investment decisions of other non-distressed firms in the industry. 3 The identification challenge of this analysis is that common economic factors, such as negative demand shocks, could simultaneously lead the weakest firms to miss their debt payment obligations or even file for bankruptcy, and the rest of the firms to reduce their capital investments to adjust to the new economic situation. To overcome this fundamental endogeneity problem, the main identification strategy exploits the cross-sectional heterogeneity of firms’ long-term debt maturity structures within a given industry and year. Specifically, estimations examine whether firms with large fractions of their long-term debt maturing right after a rivals’ bankruptcy filing or debt default (treated firms) had to cut their investment expenditures more than otherwise similar firms that did not have to refinance their long-term debt at that time (control firms). Specifications include industry*year fixed effects, which control for common shocks to the cash flows of all industry participants in a given year. Additionally, the dependent variable is measured in differences to account for unobservable, idiosyncratic firm effects that are fixed around the distress period. Further, the models account for observable firm characteristics that could simultaneously determine investment and debt maturity structures – i.e. size, profitability, investment opportunities, cash flows, and leverage ratios (Barclay and Smith, 1995; Guedes and Opler, 1996; Choi, Hackbarth, and Zechner, 2013) – both as controls in the regressions and through a matching approach. Results from this analysis show that, on average, treated firms cut their yearly investment ratios by significantly larger amounts than controls. Economically, the coefficients imply that the difference in the change in investment to capital ratios is approximately 4 percentage points higher for control firms relative to treated firms. This represents a level of investment that is around 10% lower than predistress levels. Overall, these findings suggest that the potential benefits of increasing investment are more than offset, on average, by the high costs of finance triggered by the distress.

Even as modern researchers and practitioners recognize the critical need for more accurate bankruptcy and distress prediction models, a lack of consensus remains regarding how various proposed models perform in different economic circumstances. In particular, available bankruptcy prediction models might not generalize across economic environments, such as those that mark different nations. By scrutinizing the prediction capability of models across countries, the current study seeks to extend prior literature that tends to investigate prediction models only in relation to developed economies (e.g., Agarwal & Taffler, 2007, 2008; Boritz, Kennedy, & Sun, 2007). But such studies necessarily reflect the unique traits of their samples, suggesting the powerful demand for cross-country analyses of extant models (Altman, Iwanicz-Drozdowska, Laitinen, & Suvas, 2017), across economies that represent diverse settings. Furthermore, some prediction models fail to establish a firm theoretical basis for their financial ratio selections (Charitou, Neophytou, & Charalambous, 2004; Gentry, Newbold, & Whitford, 1985a; Grice & Dugan, 2003; Oz & Yelkenci, 2017), which could imply even greater sample dependence. To explore existing bankruptcy prediction models’ generalizability, and in particular their applicability to emerging economies, this study focuses on five prominent models proposed by Altman (1968), Ohlson (1980), Taffler (1983), Zmijewski (1984), and Shumway (2001). All five of these models originally were derived with samples that came from developed economies, whereas their applicability to emerging economy samples has not been tested. Furthermore, the models originally applied to industrial firms, and the health of such firms is central to the efforts of emerging markets to participate in the global economy (Khanna & Palepu, 2006; Oz & Yelkenci, 2017). In this sense, confirming the generalizability of these models would provide pertinent insights for research but also hold promise for informing practitioners about which prediction models they should adopt. Some previous research already has established that these prediction models are generalizable in terms of their classification accuracy across different samples (e.g., Grice & Dugan, 2001, 2003; Grice & Ingram, 2001). That is, these studies show that the prediction models can detect company distress accurately, independent of the observation samples. But in addition to testing the generalizability of these prediction models across different samples, it also is necessary to test for re-estimations of the model coefficients (Grice & Dugan, 2003) and confirm the statistical significance of the prediction results (Grice & Dugan, 2001). Few tests of proposed prediction models include these research considerations though. Instead, most research tends to implement specific prediction models for individual country samples, to measure and compare their prediction performance (Kordlar & Nikbakht, 2011; Lifschutz & Jacobi, 2010; Oude, 2013; Pongsatat, Ramage, & Lawrence, 2004), or else apply original versions of the models without examining the statistical validity of their results (Almamy, Aston, & Ngwa, 2016; Chouhan, Chandra, & Gosvami, 2014; Hussain, Ali, Ullah, & Ali, 2014; Malik, Aftab, & Noreen, 2013; Mizan & Hossain, 2014). To extend beyond such considerations, the present study checks the effectiveness of the five models across an economically diverse, multicountry sample.

This paper deals with a topic of common concern to China, Europe and the United States: the negative effects of bankruptcy stigma on the second chance (fresh start) policy encouraging restructuring of businesses as an alternative to their liquidation. In most Continental European civil law systems, for example, business restructurings are still only aspirations rather than reality. This is to a great extent due to the ubiquity of intense bankruptcy stigma as a consequence of what, for example, creditors as well as the directors and officers of the bankrupt debtor avoid participating in restructuring proceedings. The resulting dominance of liquidations is perceived as a competitive disadvantage both for China and Europe compared to the United States that possesses the top model enshrined in Chapter 11 of the US Bankruptcy Code. It was for these practical reasons that the second chance policy was given clear priority by the European Union as best expressed in the Commission Recommendation of 12 March 2014 on a New Approach to Business Failure and Insolvency. Similar policy shift characterizes the 2007 Enterprise Insolvency Law of the People’s Republic of China as visible from Chapter 8 on reorganisation and Chapter 9 on compositions (workouts). While bankruptcy stigma is present also in the United States, its effects are the least ‘‘biting’’ in this country and are an issue primarily in the context of consumer-bankruptcies. In light of the above, this article’s main claim is that without proper understanding and acknowledging the impact of bankruptcy stigma, hardly could lawmakers’ efforts aimed at forging a legal environment that would incentivize restructurings of financially distressed businesses yield success. Although some research on the topic is available, it tends to be focused on consumer bankruptcies only. A comprehensive, empirically based, interdisciplinary scrutiny of the impact of stigma on business reorganisations is still lacking just like a ‘‘handbook’’ for combating the bankruptcy stigma. This article attempts to open the doors to this new inter-disciplinary area of law with the tools of comparative law. Besides canvassing the pertaining scholarship’s hereinbefore achievements, the paper extends also to such so far neglected niches of the globe as China and the post-socialist countries of Central and Eastern Europe.

Why focus on bankruptcy stigma?

Both, theoretical and practical reasons justify paying heightened attention to, as well as researching and writing on bankruptcy stigma from a comparative perspective. These revolve around two main interlinked commonalities of all the jurisdictions within the purview of this paper. On the one hand, bankruptcy stigma is ubiquitous and thus it inevitably affects the functioning of all bankruptcy systems. On the other hand, each of these jurisdictions have posited the second chance (fresh start) policybased business restructurings as first priorities. However, although both China and much of Europe faces major challenges with shifting to such rescue-oriented systems, the stigma’s impact on the process has so far been largely neglected both by legal scholarship (especially comparative law) and by expert international bodies having been involved in assisting related law reforms. A foolproof toolbox for decreasing the intensity of stigma, or eliminating some problematic aspects of it, has not yet been forged either. The following paragraphs provide more detail on each of these points. Firstly, bankruptcy stigma is ubiquitous in modern societies, from China, through Europe to the United States (US). Although no exact, commonly accepted quantitative indicators exist,3 the available sources posit that bankruptcy stigma is considerably more intense in China4 and Europe—especially in the Continental European civil law jurisdictions—compared to the US. The intensity of stigma, understandably, varies from country to country (sometimes divergences subsist even on regional levels), due to historic reasons and the dissimilar societies, economies, as well as the non-uniform bankruptcy laws. The manifestations of stigma may also be at variance5 along these geographic lines. Notwithstanding the disparities, however, there is a common denominator of central importance to us here: the bankruptcy stigma affects the functioning of the bankruptcy system—both consumer and business bankruptcies—with varying intensity in all the jurisdictions within the purview of this article. What is known is that in general the intensity of bankruptcy stigma tends to be lower in those Anglo-Saxon legal systems that are at the same time also among the most advanced economies of the world.

Introduction

Since the work of Guiso et al. (2004), there has been a renewed interest in the differences at the local level of financial development affecting firm and entrepreneur strategies. While it appears well stated that local financial development and, in general, institutional features of the local context, shape the financial decisions of firms (Cariola et al. 2010; Deloof and La Rocca 2014; Deloof et al. 2016), with particular regard to small and medium-sized enterprises (SMEs), it is a first-order problem to investigate how the differences in the local market can affect the SMEs’ quality of life in the shortrun and their capability to access growth opportunities and operate for long-term success. In particular, local financial development can provide valuable support at the time firms are more fragile, as in their early stages. Small and young companies are the primary source of job creation in economies (Haltiwanger et al. 2013), contributing to economic dynamism by injecting competition into markets and spurring innovation (Wiens and Jackson 2015). At the same time, these firms are the most financially vulnerable in the market (Mueller and Stegmaier 2014). This weakness leads to questioning the role played by the financial system and, more precisely, by banks, in the local economic activity and as a driver of the performance of local firms. This research addresses this question focusing on young firms’survival, one among many proxies describing the robustness of the companies. The analysis is centered on the role of local financial markets as determinants of firm bankruptcy. Considering that small and young firms are mainly hit by strong difficulties in the take-off years (Ogane 2016), we consider that the postcreation period is the moment where the local financial context provides the more valuable support to SMEs. The objective of this paper is thus to investigate whether the local financial market influences SMEs bankruptcies to enhance our understanding of the drivers of this failure, to explore potential areas of interventions and to transform business failures into learning opportunities for future improvements in entrepreneurship. The novelty of this paper results from several features: the emphasis put on the relationship between insolvency and the organization of the local credit market, the sample of companies considered in the empirical analysis based on firms at the post creation period, and the estimation technique used. Examining the sources of the regional disparities in the probability of corporate bankruptcy, it rapidly appears that the structure of the local debt market matters. A strand of the literature shows that credit rationing and institutional features vary across regions (Bonnet et al. 2005; Andriani 2013 and 2015), suggesting that the regional dimension is more important when companies are small (Bonnet and Le Pape 2012).

1- Introduction

2- Research design

3- Empirical results

4- Robustness checks

5- Concluding remarks

References

Abstract

Extant literature suggests that corporate social responsibility (CSR) accrues social capitals that buffers business risk. We extend this literature by documenting that firms with higher prior history of positive CSR engagement are less likely to file for bankruptcy when they are in deep financial distress and are more likely to experience accelerated recovery from distress. Furthermore, we decompose social capitals accrued from prior CSR engagement into moral capital and exchange capital. The results show that moral capital reduces bankruptcy likelihood when the firm grows larger. On the other hand, exchange capital mitigates bankruptcy likelihood when the firm relies on intangible assets to operate and when firms operates in more litigious business environment.

Introduction

This study aims to evaluate the value of corporate social responsibility (CSR, henceforth) in the context of corporate financial distress. At the crux of the study is a simple question: whether and how the bankruptcy likelihood of financially distressed firms can be reduced by their CSR engagement? CSR engagement is increasingly perceived as a risk management instrument that not only reduces cash flow volatility but also helps avoid costly lower-tail outcomes (Earnst & Young, 2017; Minor & Morgan, 2011; Peloza, 2006; Perez-Batres, Doh, Miller, & Pisani, 2012). Godfrey (2005) theorize that a firm’s prior CSR engagement provides an insurance-like protection when the firm is in crisis. For instance, firms with proactive CSR engagement in managerial practices like environmental assessment and stakeholder management (Wood, 1991) tend to anticipate and reduce potential sources of business risk, such as potential governmental regulation, labor unrest, or environmental damage (Orlitzky & Benjamin, 2001).1 We extend Godfrey (2005) by looking for evidence on the relationship between prior CSR engagement and bankruptcy likelihood. The empirical investigation is conducted based on a sample of financially distressed firms. Our research question is partly motivated by Smith and Stulz (1985), who postulate that value-maximizing corporations rationally acquire risk management instruments when the corporations expect future financial distress. This is because risk management instruments are most valuable to the corporations when the expected financial distress cost becomes so onerous that bankruptcy becomes imminent. As CSR engagement can be used as a form of risk management instrument, we expect that the prior CSR engagement can mitigate the expected distress cost and, consequently, lowers the bankruptcy likelihood.

The interest in the prediction of frms’ bankruptcy is increasing in the recent recession period 2008–2012, when, in Italy, the number of distressed manufacturing frms increased sharply. The most popular model applied by bankruptcy researchers is the logit model (logistic regression model). In the present paper we extend this classical model in two diferent ways, to take into account the spatial efects that can highly afect bankruptcy probability. We propose to apply the spatial Autologistic model and the Logit Regression Tree, with the aim to fnd evidence of spatial dependence and spatial heterogeneity in bankruptcy probability, of the manufacturing frms of Prato and Florence (Italy). Our application shows that spatial contagion efects are an important issue when modelling bankruptcy probability. Moreover, the application of the regression tree analysis shows the presence of three diferent clusters, with heterogeneous behaviours.

Introduction

The interest of empirical literature on the frm’s performance in terms of survival or exit from the market has recently increased, due to the worldwide negative efect of the recent recession on frms’ crises. In Italy, where the enterprise system is characterized by family running small-sized frms, the economic crisis of 2008 signifcantly worsened the fnancial situation of the frms (Ferretti et al. 2016), and the number of distressed manufacturing frms increased sharply during this period. Financial structure, like leverage, plays a key role in the solvency of frms and during the crises it is deteriorated, together with the accumulation of the trade debt of the Italian frms. Bonaccorsi di Patti et al. (2015) and Fort et al. (2013) are few references on this topic. Early attempts to predict corporate failure in a univariate context used the ratio analysis (Beaver 1966), which was later extended by Altman (1968) to the multivariate case. More recently, the most popular model applied by bankruptcy researchers is the logit model (logistic regression model). Since in this case the outcomes are between two discrete alternatives, fail and non-fail, bankruptcy classifcation is an appropriate application for a binary choice model and logit model is commonly used in such qualitative response studies. For a review on the models applied in the bankruptcy analysis, see Gissel et al. (2007). In the time domain, an alternative way to model bankruptcy is the use of duration models, that allow to estimate the length of the time until failure (Manjon-Antolin and Arauzo-Carod 2008). The survival of frms depends on several factors, that can be distinguished between internal factors (i.e. frm-specifc) and external factors. The latter are related to the environment in which the frm operates and can be summarised in industry, spatial and business-cycle factors. Empirical analysis in diferent nations [see, e.g., Giovannetti et al. (2011) and Mariani et al. (2013) for Italy, Bernard and Jensen (2007) for the US, Box (2008) for Sweden, Bellone et al. (2006) for France, Disney et al. (2003) for the UK] has produced interesting fndings, not always with according results that may bring to stylized facts. The role of frmsspecifc factors in determining frm failure, like age and size, is central in theoretical frm-survival models, however empirical evidence is not unanimous. Although Esteve-Pérez et al. (2004), Fackler et al. (2013) and Strotmann (2007) show that small frms exhibit a shorter life expectation, Varum and Rocha (2012) assert conversely that smaller enterprises may be more fexible in adjusting to downturns, being more able to exploit market niches and activities characterized by agglomeration economies, and being less reliant on formal credits compared with larger frms (Tan and See 2004). Moreover, Agarwal and Gort (2002) found that the age efect has an inverted-U shape (Esteve-Pérez and Mañez-Castillejo 2008) and size-age infuences may also difer across industries (Giovannetti et al. 2011; Lopez-Garcia et al. 2007): large frms that operate in high-tech sectors have higher probability of survival than small frms in traditional sectors.

In this paper, we compare the performance of two non-parametric methods of classification and regression trees (CART) and the newly multivariate adaptive regression splines (MARS) models, in forecasting bankruptcy. Models are tested on a large universe of US banks over a complete market cycle and run under a K-fold cross validation. Then, a hybrid model which combines K-means clustering and MARS is tested as well. Our findings highlight that (i) Either in training or testing sample, MARS provides, in average, better correct classification rate than CART model (ii) Hybrid approach significantly increases the classification accuracy rate in the training sample (iii) MARS prediction underperforms when the misclassification of the bankrupt banks rate is adopted as a criteria (iv) Finally, results prove that non-parametric models are more suitable for bank failure prediction than the corresponding Logit model.

Introduction

The financial crisis, which started in 2007, has dramatically affected banks sector throughout the world. The shock wave epicenter was in the US and it took long time for regulators to stop the default chain and save big banks. Therefore, the prevention against the systemic risk – failure of the banking system- becomes an ineluctable concern and the need of new forecasting tools is of major importance to not only regulators but also academics. In this sense, federal banking supervisors [the Federal Reserve, the Federal Deposit Insurance Corporation (FDIC), and the Office of the Comptroller of the Currency (OCC)] and other financial supervisory agencies provide a supervisory rating (convenient summary of bank conditions at the time of an exam). This helps investors to detect banks that have a great default probability ratio. A key outcome of such an on-site exam is a CAMELS rating. The acronym “CAMEL” refers to the five components of a bank’s condition that are evaluated: Capital adequacy, Asset quality, Management, Earnings, and Liquidity. A sixth component, a bank’s Sensitivity to market risk, was added in 1997; hence the acronym was changed to CAMELS. The FDIC developed also a Statistical CAMELS Off-site Rating system (SCOR) to perform the bank’s stability evaluation. Collier et al. (2003) examine the performance of this model over the period 1986–2002 and point out the limitations of this model despite the usefulness of SCOR which is based only on financial ratios. Cole and Gunther (1995) prove the same results and report that the CAMELS ratings decay rapidly. Predicting bank bankruptcy has reached a specific interest in financial literature. Thus, numerous models have been developed since the early 70s. All proposed models are based on classification methods in a multidimensional space defined by a set of specific variables. The literature is rich of non-parametric and parametric models. With regard to the later, Beaver (1966) was one of the first researchers who focused on univariate analysis to study bankruptcy prediction. He tested the power of financial ratios to classify and predict bankrupt firms. Cash flow and debt ratios appeared to be the important predictors of bankruptcy. Altman (1968) used Multivariate Discriminant Analysis (MDA) to develop a five-factor model to calculate the well-known “Z-score” and predict bankruptcy of manufacturing firms. As an examples of well-known statistical methods to predict failure, the logistic regression (logit) (Ohlson 1980; Demirgüç-Kunt and Detragiache 1997; Affes and Hentati-Kaffel 2017), Probit analysis (Zmijewski 1984; Hanweck et al. 1977) and factor analysis. West et al. (1985) demonstrated that that the combination of factor analysis and Logit estimation was promising in evaluating banks’ condition. The factors identified by the Logit model as important descriptive variables for the banks’ operations are similar to those used for CAMELS ratings.