Over the last dozen years, the topic of small and medium enterprise (SME) default prediction has developed into a relevant research domain that has grown for important reasons exponentially across ...multiple disciplines, including finance, management, accounting, and statistics. Motivated by the enormous toll on SMEs caused by the 2007–2009 global financial crisis as well as the recent COVID-19 crisis and the consequent need to develop new SME default predictors, this paper provides a systematic literature review, based on a statistical, bibliometric analysis, of over 100 peer-reviewed articles published on SME default prediction modelling over a 34-year period, 1986 to 2019. We identified, analysed and reviewed five streams of research and suggest a set of future research avenues to help scholars and practitioners address the new challenges and emerging issues in a changing economic environment. The research agenda proposes some new innovative approaches to capture and exploit new data sources using modern analytical techniques, like artificial intelligence, machine learning, and macro-data inputs, with the aim of providing enhanced predictive results.
Considering the fundamental role played by small and medium sized enterprises (SMEs) in the economy of many countries and the considerable attention placed on SMEs in the new Basel Capital Accord, we ...develop a distress prediction model specifically for the SME sector and to analyse its effectiveness compared to a generic corporate model. The behaviour of financial measures for SMEs is analysed and the most significant variables in predicting the entities’ credit worthiness are selected in order to construct a default prediction model. Using a logit regression technique on panel data of over 2,000 U.S. firms (with sales less than $65 million) over the period 1994–2002, we develop a one‐year default prediction model. This model has an out‐of‐sample prediction power which is almost 30 per cent higher than a generic corporate model. An associated objective is to observe our model's ability to lower bank capital requirements considering the new Basel Capital Accord's rules for SMEs.
In response to the pandemic, the Federal Reserve and US Treasury aggressively supported the corporate debt markets in 2020, both through easy monetary policy and direct participation. As the economy ...recovered, the Fed maintained an easy monetary policy in pursuit of more robust employment, gross domestic product growth, and inflation targets. Many observers have asserted that these policy actions and other factors have driven a bubble in risk assets, such as equities, special purpose acquisition companies, real estate, commodities, cryptocurrencies, nonfungible tokens, and risky bonds. Others believe that the valuations for at least some of these assets reflect investors’ rational incorporation of the current interest rate environment and economic outlook into their underwriting assumptions. Here, the authors analyze this question with regard to one of the riskiest classes of debt securities—the CCC-rated portion of the corporate high-yield debt market—to draw broader conclusions about the leveraged credit markets in the United States. Using historical market metrics, this portion of the market at its recent peak offered almost no excess return to compensate investors for the risk taken relative to low-risk alternatives. From this, the authors conclude that investors are underwriting significantly more optimistic outcomes than those reflected in historical averages. Although the authors believe that this segment of the market falls short of a true bubble, as they define it, they warn that current conditions pose key risks to investors and to the broader economy. Key Findings ▪ The US high-yield bond market, like other asset classes in the wake of the current easy monetary environment, has reached peak levels. ▪ Consistent with past bubbles, investors have moved up the risk curve, disproportionately bidding up the riskiest portion of this market. ▪ Although this reflects excessive risk appetite on the part of investors, it falls short of a true bubble as we define it. ▪ That said, these conditions pose key risks to investors and to the broader economy.
Fifty years ago, I published the initial, classic version of the Z-score bankruptcy prediction models. This multivariate statistical model has remained perhaps the most well-known, and more ...importantly, most used technique for providing an early warning signal of firm financial distress by academics and practitioners on a global basis. It also has been used by scholars as a benchmark of credit risk measurement in countless empirical studies. Practical applications of the Altman Z-score model have also been numerous and can be divided into two main categories: (1) from an external analytical standpoint, and (2) from an internal to the distressed firm viewpoint. This paper discusses a number of applications from the former's standpoint and in doing so, we hope, also provides a roadmap for extensions beyond those already identified.
Few studies that have focused on developing credit risk models specifically for small and medium-sized enterprises (SMEs) have included non-financial information as a predictor of company ...creditworthiness. In this study we have available non-financial, regulatory compliance and "event" data to supplement the limited accounting data that is often available for non-listed firms. We employ a sample consisting of over 5.8 million sets of accounts of unlisted firms, of which over 66,000 failed during the period 2000-2007. We find that data relating to legal action by creditors to recover unpaid debts, company filing histories, comprehensive audit report/opinion data and firm-specific characteristics make a significant contribution to increasing the default prediction power of risk models built specifically for SMEs. PUBLICATION ABSTRACT
A Race for Long Horizon Bankruptcy Prediction Altman, Edward I.; Iwanicz-Drozdowska, Małgorzata; Laitinen, Erkki K. ...
Applied economics,
08/2020, Volume:
52, Issue:
37
Journal Article
Peer reviewed
Open access
This study compares the accuracy and efficiency of five different estimation methods for predicting financial distress of small and medium-sized enterprises. We apply different methods for a large ...set of financial and non-financial variables, using filter and wrapper selection, to predict bankruptcy up to 10 years before the event in an open, European economy. Our findings show that logistic regression and neural networks are superior to other approaches. We document how the cost-return ratio considerably affects the location of optimal cut-off points and attainable profit in credit decisions. Once a loan provider selects a particular prediction model, an effort should be made to find the optimal cut-off score to maximize the efficiency of the technique. Indeed, this often involves determining several cut-off levels where the portfolio of products and services exhibits different cost-return characteristics.
Surveys on the use of agency credit ratings reveal that some investors believe that rating agencies are relatively slow in adjusting their ratings. A well-accepted explanation for this perception on ...the timeliness of ratings is the through-the-cycle methodology that agencies use. According to Moody’s, through-the-cycle ratings are stable because they are intended to measure default risk over long investment horizons, and because they are changed only when agencies are confident that observed changes in a company’s risk profile are likely to be permanent. To verify this explanation, we quantify the impact of the long-term default horizon and the prudent migration policy on rating stability from the perspective of an investor – with no desire for rating stability. This is done by benchmarking agency ratings with a financial ratio-based (credit-scoring) agency-rating prediction model and (credit-scoring) default-prediction models of various time horizons. We also examine rating-migration practices. The final result is a better quantitative understanding of the through-the-cycle methodology.
By varying the time horizon in the estimation of default-prediction models, we search for a best match with the agency-rating prediction model. Consistent with the agencies’ stated objectives, we conclude that agency ratings are focused on the long term. In contrast to one-year default prediction models, agency ratings place less weight on short-term indicators of credit quality.
We also demonstrate that the focus of agencies on long investment horizons explains only part of the relative stability of agency ratings. The other aspect of through-the-cycle methodology – agency-rating migration policy – is an even more important factor underlying the stability of agency ratings. We find that rating migrations are triggered when the difference between the actual agency rating and the model predicted rating exceeds a certain threshold level. When rating migrations are triggered, agencies adjust their ratings only partially, consistent with the known serial dependency of agency-rating migrations.
We explore the quality of risk assessment for entrepreneurs/small business borrowers as compared to consumers, when the same information on previous credit history is used for both segments in ...marketplace lending. By building several cross-sectional logistic regression and machine-learning models and applying them separately to small business loans (SBL) and consumers we can measure models’ predictive accuracy for different segments, and thus, make observations about the value of the information used for screening. We find the differences in profiles between SBL and consumers, hence they should be assessed by separate models. Yet separate SBL models do not perform well when applied to a future time period. We attribute this to the relatively low predictive value of personal credit history for entrepreneurs as compared to the consumers. We advocate the use of additional information for risk assessment of entrepreneurs, in order to improve the quality of credit screening. This should lead to improved access of small business borrowers to credit in situations when they have to compete with consumers for funding.
This paper analyzes the association between default and recovery rates on credit assets and seeks to empirically explain this critical relationship. We examine recovery rates on corporate bond ...defaults over the period 1982–2002. Our econometric univariate and multivariate models explain a significant portion of the variance in bond recovery rates aggregated across seniority and collateral levels. We find that recovery rates are a function of supply and demand for the securities, with default rates playing a pivotal role. Our results have important implications for credit risk models and for the procyclicality effects of the New Basel Capital Accord.
This paper uses a new data set of daily secondary market prices of loans to analyze the specialness of banks as monitors. Consistent with a monitoring advantage of loans over bonds, we find the ...secondary loan market to be informationally more efficient than the secondary bond market prior to a loan default. Specifically, we find that secondary market loan returns Granger cause secondary market bond returns prior to a loan default. In contrast, secondary market bond returns do not Granger cause secondary market loan returns prior to a loan default.