Figure - uploaded by Marco Balzano
Content may be subject to copyright.
Source publication
SME default prediction is a long-standing issue in the finance and management literature. Proper estimates of the SME risk of failure can support policymakers in implementing restructuring policies, rating agencies and credit analytics firms in assessing creditworthiness, public and private investors in allocating funds, entrepreneurs in accessing...
Contexts in source publication
Context 1
... groups of variables were then added to assess whether these were important for improving prediction. For the selected variables, variable correlations were analyzed and, when the models selected highly correlated variables, those variables whose univariate AIC had a greater value were included (Table SM2). In addition, variance inflation factors (VIFs) were analyzed for all the independent variables in the logit model. ...Context 2
... Hyperparameters are tuned automatically. 14 Figure SM1 and Table SM2 show correlation checks and decisions on which variable to keep among the correlated pairs. All models are checked for the VIFs. ...Context 3
... 1 (column 5) shows that prediction is improved compared to the LASSO models when financial and previous payment behavior variables are included (column 4 vs 5). Table 2 shows the LASSO selected variables and their individual logit coefficients. Predictors that are negatively associated with SME default are surplus dummy, quick ratio, retained earnings over total assets, equity over total investments, and employee tenure. ...Context 4
... that are positively associated with SME default are days of debtor and client change, personnel costs over value added, previous payment default, change in management and firing ratio. An increase in the number of previous payment defaults, a higher ratio of fires, or a change in management is associated with a higher probability of SME default (compare, Table 2). 15 These results provide further support for our hypotheses. ...Citations
... Our preference towards adaptive LASSO over other regularization methods relies on the oracle property, which guarantees a consistent selection of independent variables in large samples. For example, standard LASSO regularization (Tibshirani, 1996), which has been recently employed to predict high growth firms (Chae, 2024;Coad & Srhoj, 2020) and bankruptcy risk (Altman et al., 2023), may lead to an inconsistent selection (Zou, 2006). The adaptive LASSO regularization is defined as: ...
Why do entrepreneurs leave their current entrepreneurial ecosystems to relocate elsewhere? While entrepreneurial ecosystem (EE) theory tends to assume that entrepreneurs remain embedded within their ecosystems, this study challenges that notion by examining ecosystem “leakage” and highlighting the permeability of EE boundaries. Drawing from an embeddedness perspective, we extend EE theory by developing hypotheses on relational, structural, family, and societal embeddedness. In doing so, we integrate both traditional and non-business-related inputs that influence entrepreneurial relocation decisions. Particularly, we provide a novel operationalization of EE inputs, also including those related to private well-being systems (e.g., cost of living, family support) and societal development (e.g., public safety, technological advancement). Using survey data from 522 entrepreneurs and applying adaptive LASSO and random forests, we find that non-business inputs significantly shape relocation decisions. Entrepreneurs weigh multiple dimensions of embeddedness when considering relocation. As well as business-related inputs contribute to retention, the presence of private well-being systems and advanced societal development significantly reduces the likelihood of leaving. When these non-business factors deteriorate, even ecosystems rich in traditional EE inputs may experience leakage. Overall, our findings suggest that ecosystem retention strategies must extend beyond economic incentives to address the broader social and institutional factors that sustain entrepreneurial embeddedness.
... Management structures cover variables on governance. Director turnover, membership of female or family directors, CEO duality, and concentration of ownership appear to be negatively related to failure (Altman et al., 2016(Altman et al., , 2022Ciampi, 2015;Freixanet et al., 2024;Laitinen & Gin Chong, 1999;Wilson & Altanlar, 2014). Exit of key members is positive related to business failure (Altman et al., 2022;Ciampi, 2015). ...
... Director turnover, membership of female or family directors, CEO duality, and concentration of ownership appear to be negatively related to failure (Altman et al., 2016(Altman et al., , 2022Ciampi, 2015;Freixanet et al., 2024;Laitinen & Gin Chong, 1999;Wilson & Altanlar, 2014). Exit of key members is positive related to business failure (Altman et al., 2022;Ciampi, 2015). For family businesses, concentration of ownership appears to have a mixed effect on survival. ...
... It includes the ability to acquire critical levels of working capital (Halabi & Lussier, 2014;Lussier, 1995) and staffing (Lussier, 1995). Altman et al. find that the employee firing ratio is a statistically significant predictor as well (Altman et al., 2022). ...
... Different types of payment defaults have been frequently used in certain studies focusing on firm performance, namely the business failure prediction studies (Lukason & valgenberg 2021). Due to the instability and lack of timely financial reports in smaller firms (Altman et al., 2020) and the effect of behavioural consistency between entrepreneurs' personal and occupational spheres, variables like managers' individual payment histories (Laitinen, 1999) or that of the firms they have been associated with are more effective at predicting financial distress in smaller firms than traditional financial ratios (Altman et al., 2023). A study by Andresson and Lukason (2024) of all Estonian SMEs found that payment disturbances in managers' earlier entrepreneurial ventures were better predictors of current firm's corporate default than traditional financial ratios. ...
... This finding allows to accept hypothesis 3, and it is consistent with existing literature that suggests a link between entrepreneurs' individual deviant payment behaviour and its existence in their future firms (Back, 2005;Kallunki & pyykkö, 2013). It emphasises the importance of using payment behaviour to predict financial distress in the context of serial entrepreneurship (Altman et al., 2023). The scatter plot depicting the association of DEFAuLTOLD and DEFAuLTNEW has been presented in Appendix 5. ...
This study aims to determine whether financial performance is persistent in serial entrepreneurship and, if so, whether good performance is more persistent than bad. It is novel as studies focusing on financial performance in successive firms ran by serial entrepreneurs are scarce. The paper creates a multi-theoretical framework from which four hypotheses are drawn. These focus on the persistence of firm size, export intensity, financial risk level, and payment defaults in serial entrepreneurship, all of which have been rarely utilized in such research. The validation of hypotheses is conducted by using ordinary least squares regression with a sample of 1599 Estonian serial entrepreneurs, being further divided into those with earlier good or bad performance. The dependent and independent variables, supplemented by various controls, portray the four performance measures in serial entrepreneurs' new and old firms. The regression analyses show that financial performance persists in serial entrepreneurship in general. Good performance is more persistent than bad, indicating inter-firm transfer of financial success. In most cases, bad performance is not persistent at all, indicating that some serial entrepreneurs repeat their earlier poor results, while others improve their future performance.
... In contrast, we focus on improving credit assessments for all SMEs in an economy using all available bankruptcy predictors derived from both financial and non-financial information. Another study related to ours is that of Altman et al. (2023aAltman et al. ( , 2023b, who found that models based on financial variables show improved predictive power when payment behavior, management-related and employee-related variables are incorporated. Moreover, Ciampi (2015) demonstrated that including management-related variables in existing models based on financial variables enhances the prediction of default among small enterprises. ...
... Moreover, Ciampi (2015) found that default predictions among small enterprises improve when management-related variables are included in existing models based on financial variables. Altman et al. (2023aAltman et al. ( , 2023b developed an SME default predictor by considering financial variables in combination with payment behavior, management-related and employee-related variables and found that the predictive power of models based on financial variables improves when introducing these additional variables. Other hard non-financial variables used to assess SME credit risk are derived from the local banking market (Arcuri & Levratto, 2020), published legal judgments (Yin et al., 2020) and bag-of-words models applied to content scraped from corporate websites (Crosato et al., 2023). ...
... Indeed, these three classes of ensemble methods have been found to outperform other methods for bankruptcy prediction (Barboza et al., 2017). XGBoost (Chen & Guestrin, 2016), which was also used to test the robustness in Altman et al. (2023a) and AdaBoost (Freund & Schapire, 1997) build ensembles of decision trees sequentially, where each tree corrects the errors of the previous ones. Furthermore, bagging, or bootstrap aggregating (Breiman, 1996), trains multiple decision tree classifiers on subsets of the training data selected randomly with replacement and then aggregates their predictions to improve stability and accuracy. ...
We investigate non-financial variables for predicting bankruptcy in small and medium-sized enterprises (SMEs). The variables encompass management, board and ownership structures and are sourced from universally accessible information, rendering them available to all stakeholders and allowing for the analysis of all SMEs within a market. Using a large and recent sample of SMEs, we empirically examine the variables that predict bankruptcy over time horizons of one, two and three years. Our analysis incorporates state-of-the-art discrete hazard models, the least absolute shrinkage and selection operator (LASSO), extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), bagging and random forest. We also test robustness using balanced datasets generated using the synthetic minority oversampling technique (SMOTE). We find that including non-financial variables enhances bankruptcy predictions compared to using financial variables alone. Moreover, our results show that among our variables, the most significant non-financial predictors of bankruptcy are the age of chief executive officers (CEOs), chairpersons and board members, as well as ownership share and place of the board members’ residences.
... Such studies compare healthy to defaulted firms (i.e. Altman et al., 2023a;2023b;. Instead, our study contributes to a scarce literature that tries to empirically identify variables that can explain why some firms in financial distress recover and others do not. ...
... Instead, our study contributes to a scarce literature that tries to empirically identify variables that can explain why some firms in financial distress recover and others do not. Notably, while there has been extensive research on factors leading to financial distress (e.g., Altman et al., 2023a;2023b;Cheraghali & Molnár, 2023), our understanding remains much more limited regarding the aspects that enhance the recovery pathway for distressed firms. Consistently, scholars called for a deeper understanding on the post-distress key predictors for activating the recovery pathways (Arora, 2018;Cepec & Grajzl, 2021), also highlighting the relevance of analyzing the failing and troubled firms (Jiang et al., 2021;Harrigan & Wing, 2021). ...
... First, existing studies underscore how SME financial indicators, including liquidity ratios, profitability metrics, and leverage ratios, serve as essential tools in understanding their operational health (e.g., Altman et al., 2023a). These indicators capture the immediate financial outcomes of the operational decisions and strategies that SMEs undertake. ...
... He and his colleagues also developed other research on SMEs and financial ratios in Refs. [56,57]. The work of Altma emphasised the importance of financial ratios for SMEs and their success and failure; hence, financial ratios should be addressed. ...
Micro and small enterprises (MSMEs) play a positive and significant role in developing economies by creating job opportunities and mitigating poverty; this necessitates their attention and focus, mainly during challenging times. Accordingly, this study explored the key factors contributing to enhancing entrepreneurial competency (EC) and the intention to maintain business continuity among MSMEs in Yemen during adverse times. A sample of 280 responses was collected from MSMEs operating diverse types of businesses in the capital of Yemen, Sanaa. The collected data were analysed using partial least squares–structural equation modelling (PLS-SEM), which is considered suitable for this purpose. The data collection process included an online questionnaire and a physical visit to the business locations of the business owners. The findings of the study reported that entrepreneurs’ network ties (ENT), as well as recovering capability (RC), positively and significantly influence the EC of MSMEs in the context of the study. Additionally, EC positively and significantly influences business continuity intention (BCI). Finally, EC partially mediates the relationship between ENT, RC, and BCI. The study concludes by providing recommendations and implications for policymakers.
... In this segment, timely annual reports may be one of the few opportunities to understand firm performance. In recent years, numerous non-financial variables have been introduced in prediction studies to address the unavailability or inaccuracy of financial information (Altman et al., 2023;Iwanicz-Drozdowska et al., 2016). Despite this, few studies have investigated the determinants of timely annual report submissions in the MSME segment. ...
This paper investigates the factors determining the forced closures of firms, which occur when entrepreneurs fail to submit annual reports for their businesses. Drawing on various theoretical perspectives, including a novel focus on the past reporting misconducts of the same entrepreneurs, the study sheds light on this phenomenon. Analysis of Estonian micro‐firms run by serial entrepreneurs reveals that recent reporting misconducts, particularly those of a severe nature, significantly determine a firm's forced closure. Additionally, factors such as firm size, age, and certain aspects of corporate governance play significant roles in this regard. In turn, the financial performance of a firm largely fails to signal future forced closure, potentially indicating that when entrepreneurs submit annual reports showing normal performance, they might be hiding bad performance. The paper also delineates different types of violators based on the severity of their past misconducts, noting that a particular type characterized by a large number of severe violations is especially prone to forced closures. Finally, the paper develops high‐accuracy prediction models for forced closures, identifying the number of most severe violations and firm size as the most important variables.
... In contrast, firms classified as KZ index firms are characterized by their big size, tendency to overinvest, heavy analyst coverage, and significantly higher occurrence of bond ratings compared to the whole population of enterprises. However, Altman [54], [55] have recently presented a model for predicting default in small and medium-sized enterprises (SMEs) known as Omega Score. They utilized variables relating to management and employees to determine whether a SME is likely to default. ...
Purpose – The purpose of this study is to present a comprehensive summary of the current state of research and to critically analyse the accomplishments that have been made in the field of financing constraints. Hence, the objective of this study is to analyse the research patterns in this domain by consolidating the knowledge framework.Design/methodology/approach – A bibliographical study was conducted on 190 Scopus-indexed articles filtered from 1081 articles to analyse the literature. This dataset was generated through a systematic search executed on the Scopus database, encompassing the time frame from 2000 to 2023.Findings – By employing the bibliographic coupling methodology, a total of eleven distinct research clusters were found. The author performed a comprehensive analysis of these clusters and gleaned potential research ideas from them. Our work strengthens the current empirical data. The study's findings indicate that enterprises need to overcome credit constraints in order to improve their performance. Afterwards, we constructed a comprehensive framework to serve as the foundation for suggesting significant areas for future research.Research limitations/implications – The present study has certain limitations. The methodology operates under the assumption that citations serve as the sole indicator of article quality. Consequently, it is possible that the study has overlooked publications of exceptional quality that have not yet garnered any citations.Originality/value – For this investigation, the author relied on the reliable information provided by the Scopus database. Bibliographic coupling methods was employed to examine the dynamic landscape of research encompassing this nascent knowledge domain.
... An important step forward has recently been made by Altman et al. (2023a), who proposed the Omega Score as a novel, more accurate predictor of SME default. By testing several categories of financial and non-financial information (i.e., (i) financial ratios, (ii) previous creditor payment behaviour, (iii) management-related variables, (iv) employee-related variables, and (v) macroeconomics data), they demonstrated that the incorporation of different types of predictors: (i) employee-fires ratio, (ii) mean employee tenure, (iii) history of creditor payment defaults, and (iv) change in top management, significantly improved model prediction accuracy compared to the model with solely financial ratios. ...
... Within this context, this note addresses the general scarcity of studies on SME default prediction in the tourism sector. As the first study, this note contributes to the literature by applying the Omega Score model (Altman et al., 2023a) to understand the mechanisms underlying tourism SME default better. Notably, none of the studies focusing on tourism firms use payment behaviour data or employee characteristics to predict default, which is a significant gap this note intends to fill. ...
... Note that a firm's default does not imply nonsurvival. For example, Altman et al. (2023a) use an early default event, such as 60 days of bank account blockage, to identify early warning signals of SMEs' financial difficulties, which might continue to do business later. Studies on firm survival investigate the survival determinants for specific hospitality activities, including hotels, resorts, exhibitions, restaurants, peer-to-peer accommodations (Fan et al., 2023) and multi-business hospitality firms (Li et al., 2019). ...
... Policies regarding the sector's financial sustainability, can be to create credit channels that will complementarily assist hotels limit liabilities, as for example the partial coverage of employment costs and renovation schemes. One positive development is that models offering early warning insights and link firms in financial difficulties to rescue plans, tend to be created from governments across Europe (Altman et al. 2022(Altman et al. , 2023Cultrera 2020). ...
The aim of this study is to research the financial determinants of bankruptcies for firms belonging exclusively to the economic activity 'hotels and similar accommodation'. The study adopts the stance, that estimations and models focused on specific sectors can be a more precise assessment of firm failure in contrast to generic models that sample firms from broader sectors. The dataset covers the years 2010-2020 and uses available financial and bankruptcy information from two credible data sources for over 5,500 hotels which includes 13 bankruptcy events. A multi-period logistic regression model with clustered robust standard errors is applied, a very competent econometric methodology not applied before in hotel failure studies in this exact form. This type of setting is considered a close-to-reality econometric experiment examining the whole economic activity. A stepwise procedure led to three influential financial ratios with the main results revealing that the likelihood of hotel bankruptcy is positively related to leverage and size in terms of the natural logarithm of total assets and negatively related to EBITDA to total liabilities. By extending the bankruptcy horizon further by one and two years, information content appears fairly robust and all variables retain their sign, as leverage remains stable across all horizons, EBITDA to liabilities remain significant for the one additional period, while sizes' statistical significance becomes intermittent. Comparisons are performed also in terms of sensitivity and variable relevance with two earlier Greek studies for the whole economy and the hotel sector, and in both cases sensitivity is reduced and some variables are rendered not relevant. Implications stemming from the empirical results, suggest that leverage is the most influential factor increasing the probability of bankruptcy, so liabilities should be continuously controlled. Size in terms of asset acquisition should be mitigated too, as the 'too big to fail' principle seems not to hold. Also, a balance of liabilities coverage is recommended. The comparisons in terms of sensitivity, variable relevance and the opposite sign for size implies that bankruptcy modeling per economic activity is more appropriate, stressing the need for an ad hoc variable selection that includes size also, as a predefined set of variables may not always be the most optimal set.
