Search

Climate risk refers to the risks associated with climate change and has already started to impact various sectors of the economy. In this work, we focus on the impact of physical risk on the probability of default for a firm in the agribusiness sector. The probability of default is estimated based on the Merton model, where the firm defaults when its asset value falls below the threshold defined by its liabilities. We study the relationship between the stock value of the firm and global surface temperature anomalies, observing that an increase in temperature negatively affects the stock value and, consequently, the asset value of the firm. A decrease in the asset value of the firm translates into an increase in its probability of default. We also propose a model to assess the exposure of the firm to transition risk. [ABSTRACT FROM AUTHOR]

AbstractThe credit risk assessment process is necessary for maintaining financial stability, cost and time efficiency, model performance accuracy, comparability analysis and future business implications in the commercial banking sector. By accurately predicting credit risk, highly regulated banks can make informed lending decisions and minimize potential financial losses. The purpose of this paper is to assess the power of conventional predictive statistical models with and without transforming the features to gain better insights into customer’s creditworthiness. The findings of the predicted performance of the logistics regression model are compared to the performance results of machine learning models for credit risk assessment using commercial banking credit registry data. Each model has its strengths and weaknesses, and where one model lacks, another performs better. The article reveals that simpler credit risk assessment techniques delivered outstanding performance while consuming less processing power and have given insights into the most contributing feature categories. Improving a conventional predictive statistical model using some of the feature transformations reduces the overall model performance, specifically for credit registry data. The logistics regression model outperformed all models with the highest F1, accuracy, Jaccard Index and AUC values, respectively.

Nowadays, Russian banks are developing and updating models for assessing the probability of default (PD) for various risk segments of corporate borrowers. This is to enable the use of rating models when assessing regulatory capital, in compliance with the requirements of the Regulations of the Bank of Russia. The application of rating models allows credit institutions to more accurately distribute regulatory and economic capital among borrowers, calculate reserves according to RAS and IFRS, set interest rates for transactions with borrowers as part of the pricing procedure, and conduct stress testing of both borrower portfolios and individual borrowers. The results of stress testing can then be incorporated into strategic planning. One of the Regulator's requirements is to consider assessments of sustainable development and responsible financing ESG (environmental, social, and corporate governance) indicators as part of the credit risk management procedure, particularly in the development of models for assessing the probability of default. This research focuses on the development of an integral module (ESG rating) that includes ESG indicators. The aim is to improve the accuracy of existing models for assessing the probability of default (PD models) for corporate borrowers through the use of ESG factors. The findings of this research can be utilized by Russian banks to enhance the discriminatory and predictive capabilities of their own PD models, and by the regulator to understand the set of ESG indicators that impact the creditworthiness of corporate borrowers. [ABSTRACT FROM AUTHOR]

Once a scoring model has been developed for use in assessing a borrower's credit risk, under the internal ratings-based (IRB) approach, it must be calibrated to a real-world measure of default frequency. The conservativeness of the calibration is tightly controlled, if it is not violated in the allowed number of digits of the rating scale, only then the model is authorized for use. If violated, the calibration probability of default must be raised, placing additional strain on the bank's capital and reserves. In the presented paper, two new methods to improve the calibration accuracy are proposed. The methods have been tested in practice and provide significantly positive results in certain segments of the loan portfolio. [ABSTRACT FROM AUTHOR]

The interest rate hikes intended to combat inflation have not only significantly increased mortgage payments, but also significantly depressed house prices. The probability of default for mortgages is typically estimated through the debt servicing ability of the obligor. The alternative estimation based on the default option, although well studied in literature, is typically not used by practitioners. This option is normally 'deep out of money' and moves in the money if the loan to value (LTV) increases significantly, even creating negative equity. This is typically a remote probability due to the down payment requirements, but not as much under the current environment, where the house price depreciation has significantly increased LTVs, especially for newer mortgages underwritten when house prices were at their peak. This paper discusses the potential risk management oversights in this environment, illustrating that the increasing default risk is not adequately captured in probability of default (PD) models based on debt servicing ability alone. This is also true for the loss given default (LGD) risk if the contemporaneous LTV effects are not captured. Numerical examples are provided to demonstrate the material increase in PD, LGD and the combined expected loss risk with increasing LTV. It also discusses an alternative default option PD model, its calibration and its usage for stress testing along with LGD modelling, which together capture the contemporaneous LTV effects. [ABSTRACT FROM AUTHOR]

In this study we investigate alternative interpretable machine learning ("IML") models in the context of probability of default ("PD") modeling for the large corporate asset class. IML models have become increasingly prominent in highly regulated industries where there are concerns over the unintended consequences of deploying black box models that may be deemed conceptually unsound. In the context of banking and in wholesale portfolios, there are challenges around using models where the outcomes may not be explainable, both in terms of the business use case as well as meeting model validation standards. We compare various IML models (deep neural networks and explainable boosting machines), including standard approaches such as logistic regression, using a long and robust history of corporate borrowers. We find that there are material differences between the approaches in terms of dimensions such as model predictive performance and the importance or robustness of risk factors in driving outcomes, including conflicting conclusions depending upon the IML model and the benchmarking measure considered. These findings call into question the value of the modest pickup in performance with the IML models relative to a more traditional technique, especially if these models are to be applied in contexts that must meet supervisory and model validation standards.

Climate risk refers to the risks associated with climate change and has already started to impact various sectors of the economy. In this work, we focus on the impact of physical risk on the probability of default for a firm in the agribusiness sector. The probability of default is estimated based on the Merton model, where the firm defaults when its asset value falls below the threshold defined by its liabilities. We study the relationship between the stock value of the firm and global surface temperature anomalies, observing that an increase in temperature negatively affects the stock value and, consequently, the asset value of the firm. A decrease in the asset value of the firm translates into an increase in its probability of default. We also propose a model to assess the exposure of the firm to transition risk.

We compare different models on estimating the probability of default over a time horizon, considering censored data. Models were fitted from a survival analysis perspective, considering both classic models (Cox Proportional Hazard) with penalized variations and ensemble machine learning methods (boosting and bagging). Using a dataset of credit card refinancing operations, we assess accuracy performance in both out-of-sample and out-of-time observations. We assess well-established metrics, such as the concordance index, and measures that indicate calibration power (integrated Brier score and time-dependent dynamic AUC). Results show that a boosting approach with component-wise regression as base learner outperform other models for short term operations (36 months), in contrast to longer term transactions (60 months), where Cox Proportional Hazard (with and without penalization) depicted better results. [ABSTRACT FROM AUTHOR]

We develop the first climate risk Stress Test for the Peruvian financial system following a topdown approach. Focusing on the microeconomic channel, we evaluate how heavy rainfall and droughts, under a scenario of pure physical risk, will marginally affect the probability of default (PD) of borrowers by 2050. Using information from the Credit Registry, the National Oceanic and Atmospheric Administration (NOAA), and CMIP6 precipitation projections (37 modeling groups), we calibrate the marginal impacts differentiating by economic sector and geographical location. We find that, on average, by December 2050, the probability of default of the Peruvian financial system would increase by 4.9% with respect to December 2020. By geographic area, borrowers located on the northern coast (Piura, Lambayeque) and the southern highlands (Ayacucho, Cusco) would be negatively affected by heavy rainfall, while the rainforest (Madre de Dios, Ucayali) would be negatively affected by droughts. Moreover, the economic sectors affected by heavy rainfall or droughts would be agriculture, commerce, and transportation & communications. [ABSTRACT FROM AUTHOR]

The credit risk assessment process is necessary for maintaining financial stability, cost and time efficiency, model performance accuracy, comparability analysis and future business implications in the commercial banking sector. By accurately predicting credit risk, highly regulated banks can make informed lending decisions and minimize potential financial losses. The purpose of this paper is to assess the power of conventional predictive statistical models with and without transforming the features to gain better insights into customer's creditworthiness. The findings of the predicted performance of the logistics regression model are compared to the performance results of machine learning models for credit risk assessment using commercial banking credit registry data. Each model has its strengths and weaknesses, and where one model lacks, another performs better. The article reveals that simpler credit risk assessment techniques delivered outstanding performance while consuming less processing power and have given insights into the most contributing feature categories. Improving a conventional predictive statistical model using some of the feature transformations reduces the overall model performance, specifically for credit registry data. The logistics regression model outperformed all models with the highest F1, accuracy, Jaccard Index and AUC values, respectively. Impact statement: Financial institutions, specifically banks have questioned whether transformations using Weights of Evidence (WoE) have been significant in quantifying the relationship between categorical independent variables for various types of credit data. This study provides insights when considering the usage of feature transformation for credit risk modelling in commercial banking. The transformation technique is particularly useful in situations where statistical predictive modelling techniques are employed. The results revealed that not only can the logistic regression models perform similarly to the machine learning models but can also outperform them. The best performance is attributed to the simplicity, interpretability, and access to understanding features of individual clients within a portfolio of credit products. The logistic regression model without transformation turned out to perform the best out of the five machine learning models. Considering the business impact, enhancing the logistic regression model by using a WoE transformation did not improve the model's performance for commercial banking data considered. However, the transformation did provide insights regarding each binned categorical independent variable. Therefore, our findings in this article contribute towards assisting banks in managing the impact and interpretability of each binned feature category on the discriminatory power of credit scoring. [ABSTRACT FROM AUTHOR]

Objective The probability of default is one factor that determines the cost of capital due to its role in credit risk. Dividend as a sign of cash flow or as a sign of ownership of wealth is one of the factors affecting the probability of default. There are two theories of dividend signaling theory and agency theory regarding the effect of dividends on credit risk. The cash profit signaling theory expresses the management's confidence in the company's future cash flows and causes market participants to calculate the company's assets as high and the company's asset volatility as low, thus reducing information asymmetry, it causes the probability of default calculated by market decreases and there is a negative relationship between dividend and estimated default probability. Conversely, agency theory posits that cash dividends decrease the company's assets and increase the debt ratio, and this increases the risk of debtholders and the acquisition of their wealth by shareholders, and the conflict of interests between shareholders and debtholders. Therefore, it causes a positive relationship between dividends and estimated default probability. This article aims to examine the relationship between dividends and the probability of default, exploring its alignment with both dividends signaling theory and agency theory. Methods In this article, Geske's method, which is an extension of Merton's method, is used to calculate the probability of default. Subsequently, an unbalanced nonlinear regression is applied to investigate the quadratic relationship between cash interest and the probability of default during the years 2001-2021. The statistical population comprises companies listed on the Tehran Stock Exchange and the Iran Farabourse, with the exclusion of financial companies, those with more than six months of consecutive trading suspension, and those that have experienced trading suspension for more than twenty percent of their market lifespan. Ultimately, 462 companies meet these criteria and are included in the study. Results The findings suggest a quadratic U-shaped nonlinear relationship between dividends and the probability of default. Initially, as dividends increase, the probability of default decreases with a diminishing slope. However, beyond a certain threshold of dividend increase, the probability of default starts to rise with an increasing slope. Conclusion According to the theory of dividend signaling, as dividends increase up to a certain threshold, the company's future earnings outlook is positively evaluated. The market, in turn, assesses the company's assets at a higher value with lower asset volatility. This positive assessment, attributed to the reduction of information asymmetry, leads to a decrease in the market's estimated probability of default. Once the dividend surpasses this threshold, it becomes advantageous for shareholders. This is due to a reduction in the company's assets, an elevation in the debt ratio, an increased risk for debtholders, and the potential transfer of the company's assets, ultimately leading to the appropriation of wealth from debtholders by shareholders. Therefore, the probability of default increases, which is in accordance with the empirical findings of the quadratic U-shaped relationship between dividend and the probability of default. [ABSTRACT FROM AUTHOR]

Purpose: Specific objectives are to establish the effect of effect of probability of default and financial performance of Commercial Banks in Kenya. The theoretical framework was based on asymmetric information. Methodology/Approach: The adopted mixed research comprising of causal and longitudinal research designs. The study used all commercial banks which are 42 it total. Positivism research philosophy was adopted. The study used secondary data from financial statements of banks. Data was analyzed by both descriptive and inferential statistics. Data was presented using tables. Findings: From the findings Probability of default negatively affects the Performance of Commercial Banks in Kenya as indicated by a p value of 0.00 in the fixed effect model. This can be explained as a result of the following possible factors: Loans with higher Probability of default are more likely to become non-performing assets or be classified as doubtful debts. Implications: Implementing a credit risk stress testing plan can effectively mitigate the likelihood of a bank encountering a financial crisis. In order to mitigate liquidity risk, it is advisable for banks to augment their liquid holdings.

Research Questions: In the present paper the author answers the following research questions: 1) What are the potential strengths of the credit risk model based on the cash flow principle? 2) What are the weaknesses of the accrual-based credit risk model? 3) What are the benefits of the combined use of both cash-based and accrual-based credit risk modeling methods when analyzing companies? Motivation: nowadays there are no researches comparing the accrual-based credit risk model to a cash-based credit risk model with the application to a production company trading its shares on Stock Exchange. However, for investors, auditors and financial institutions it is important to know if there is a difference between these two models in the interpretation of analysis results, and determination of prebankruptcy stage of the company and credit risk default. Idea: in this paper, the author has focused on the comparative analysis of the cash-based credit risk model and the accrual-based credit risk model. The author applies it to the case of a manufacturing company and compares the effectiveness of determining the probability of default using a cash-based credit risk model and an accrual-based credit risk model. Data: the data analysed is obtained from the annual reports, managerial reports and auditor’s reports of Linas Agro Group for the years 2016-2022. The company information is taken from Nasdaq Baltic where Linas Agro Group has its shares traded. Tools: mixed research methods were used, combining quantitative calculations with analysis based on qualitative information. The author elaborates on the cash-based credit risk model based on the improved Timothy Jury’s template. The accrual-based model chosen for comparison and analysis is Altman’s Z-Score model. Findings: The results of the study have shown that the cash-based model is more effective in determining credit risk and default probability. The cash-based model indicated a high-risk default for the manufacturing company in four years out of seven years, while Altman’s Z-Score showed the company to be in the moderate risk grey zone in five years out of seven, and the two last years the model indicated the company in the green zone. The author suggests to financial institutions, financial managers, and investors using a cash-based credit risk model or combination of it with the accrual-based model. Contribution: the paper contributes to the knowledge about the comparison of cash-based and accrual-based credit risk models and emphasizes their strengths and weaknesses. It helps investors, auditors, business owners, and finance professionals to make a decision about which credit risk model to use for the analysis to determine the pre-bankruptcy state of the company, avoid bad loans and improve investment decision-making. It also encourages the academic society for further research and comparison on the topics of accrual-based and cash-based credit risk models in the strive to develop the ultimate credit risk model capable to analyse the data as precisely as possible.

The significant share of loans in the asset portfolio of banks has turned credit risk into one of the most important risks in the banking industry. Considering the difference in the structure, profit motive and risk management in different banks, the purpose of this research is to compare the factors affecting Non-Performing Loans (NPLs) as an indicator of credit risk in different banking groups. For this purpose, the relationship between Non-Performing Loans and explanatory variables (macroeconomic factors and bank-specific factors) was estimated for different banking groups by using the generalized method of moments (GMM) of dynamic panel data for the annual data of 1385-1400. Based on the results, bank-specific factors have a more effective role in increasing the non-performing loans of banks (especially state-owned banks) than macroeconomic factors, which is due to mechanisms that lack efficiency and effectiveness in the processes of granting credit and collecting bank claims; In addition, the influence of these factors is different in different banking groups.

We examine the debt holders' wealth effect of Australian firms and the factors that determine firms' decision to issue unit rights. The sample consists of 638 offerings observations spanning from year 2000 to 2014. Probability of default has been used as the proxy for debt holders wealth. We also use probit model to gauge determinants of firms' choice in issuing unit rights. As the robustness test, logit model was also presented. Aligned with Sequential Financing Hypothesis by Schultz (1993) and Signaling Hypothesis as in Chemmanur and Fulghieri (1997), we find that firms with small size, low issuance proceeds, high risk, low managerial ownership and positive growth prospect tend to issue unit rights. We also obtain findings that support to Leverage Risk Reduction Hypothesis which suggest that debt holders' return is favorably affected by lower financial leverage.

Following recent financial, health, and geopolitical events, concerns are growing regarding the systemic risk of global energy companies. We assess the impact of environmental, social, and governance (ESG) performance on the systemic risk of leading energy firms. For 158 energy firms from 16 countries for 2010–2021, we evidence that while ESG shows positive association with default risk, squared values of ESG performance are negatively significant, consistent with an inverted U-shaped association. Moreover, better ESG disclosure and carbon performance also reduce default risk. We interpret our results as the beneficial impact of ESG on lowering default risk depends on the threshold level of ESG investment. Results, holding only for the post GFC period, for firms with more than one environmental controversy, and for countries with higher levels of education, are consistent with an evolving need for energy firms to employ genuine ESG commitment to establish reputational capital towards enhancing resiliency. [ABSTRACT FROM AUTHOR]

The inferred rate of default (IRD) was first introduced as an indicator of default risk computable from information publicly reported by the Bulgarian National Bank. We have provided a more detailed justification for the suggested methodology for forecasting the IRD on the bank-group- and bank-system-level based on macroeconomic factors. Furthermore, we supply additional empirical evidence in the time-series analysis. Additionally, we demonstrate that IRD provides a new perspective for comparing credit risk across bank groups. The estimation methods and model assumptions agree with current Bulgarian regulations and the IFRS 9 accounting standard. The suggested models could be used by practitioners in monthly forecasting the point-in-time probability of default in the context of accounting reporting and in monitoring and managing credit risk. [ABSTRACT FROM AUTHOR]

We estimate the evolution of credit risk linkages in an international banking network of 46 advanced and emerging market economies between January 2000 and December 2019. In our study, credit risk is proxied with the aggregate probability of default (PD) in the banking sector, and cross-country linkages are described by two types of networks: one representing predictive Granger relationships between the PD measures, and the other representing contemporaneous partial correlations. During the Global Financial Crisis and its aftermath, the density of the networks appears the highest, while experiencing a substantial decline in the post-crisis period triggered by massive deleveraging and the implementation of prudential measures. Throughout the whole observation period, the causal network exhibits the small-world pattern which tends to fuel risk propagation within the network, and the increasing degree of dissortativity which exerts a countervailing effect. On the country level, we document the centrality of the major Asian economies’ banking sectors and France, whereas the role of the USA, the UK and Germany appears moderate, which contrasts with the extant literature. Overall, from the policymaking perspective, our findings indicate that credit risk linkages in the international banking networks can notably differ from the relationships derived from cross-country bank claims. [ABSTRACT FROM AUTHOR]

This paper considers a hypothetical case in which a bank wants to build a routine climate stress test exercise on residential mortgage loans. The bank has regularly updated the probability of default (PD) and loss given default (LGD) on each residential mortgage loan under the internal-rating-based (IRB) approach of Basel II/III. Additionally, the bank estimates the stressed PD and stressed LGD associated with a predetermined extreme weather event. Using simulation techniques, this paper shows that the loss of the bank's residential mortgage portfolio can reach a median of around 36% of the portfolio value. This remarkable loss comes from the effects of default correlation and property damage. Banks should pay more attention to such impacts of extreme weather events. [ABSTRACT FROM AUTHOR]

Research Questions: 1) Can the newly developed by the author methodology based on Timothy Jury’s credit risk analysis provide sufficient visibility of the company’s financial situation and creditworthiness? 2) Can such methodology based on Timothy Jury’s credit risk analysis be a trustworthy indicator of production company bankruptcy and the likelihood of default? Motivation: nowadays financial institutions use an accrual-based credit risk model for the analysis of the creditworthiness of the companies, however, such models are not always trustworthy due to the data manipulations in the accrual financial statements. Since the cash flow financial statement is more reliable for the determination of the probability of default, the authors developed the cash-based credit risk model and applied it to the production company for the analysis of its likelihood of default. Idea: This paper looks at Timothy Jury’s little-known methodology of credit risk analysis and its underlying template, and proposes a version that the authors improved in accordance with the requirements of International Financial Reporting Standards (IFRSs). The authors used improved methodology for the analysis of a manufacturing company to find out the company’s pre-bankruptcy situation and reveal the likelihood of credit risk default. Data: the company analyzed is a publicly listed production company Linas Agro Group with its shared traded at Nasdaq Baltic. The data for the research was taken from the annual reports, managerial reports, and the auditor’s reports issued for seven years 2016–2022. Tools: Timothy Jury’s cash-based analysis template was modified by the authors into the credit risk model. Findings: the results show that the modified cash-based credit risk model provides sufficient visibility of the company’s likelihood of default and specifies the actual source used to cover the debt. The analysis has shown that the production company paid significant dividends with negative cash available to satisfy capital providers. Contribution: the present article contributes to the knowledge base about cash flow and credit risk. The article encourages an academic society to further research the topics related to cash flow and cash flow theory, which is currently less researched than accrual-based theory. The article is also beneficial for business owners, investors, and finance professionals to improve investment decision-making, for bank managers to diminish the number of bad loans and for the auditors to determine the pre-bankruptcy state of the company more precisely using the cash-based credit risk model presented in this article.

Credit risk models, particularly those seeking to understand signals given by companies close to bankruptcy, are under constant discussion and updating. Statistical models are traditional and easy to understand but machine learning have been tested in the financial context. Here, we developed bankruptcy forecast models for non-financial US companies, using a dataset comprising 1980–2014. We examined static, growth and also growth variation as inputs for predicting which firms would go bankrupt one year later. Seven techniques were tested, including Discriminant Analysis, Logistic Regression, and machine learning ones, such as Neural Networks, Support Vector Machines, AdaBoost, Bagging, and Random Forest. Given the number of models evaluated, the analyses are rich and varied, with emphasis on the finding that traditional techniques with the appropriate variables are more capable of achieving superior performance than machine learning models. These results open a debate with the current literature, which has been declaring precisely the opposite. Nevertheless, this article contributes to the literature in several ways, as it uses a comprehensive sample and a performance comparison, allowing potential application by both practitioners and researchers. Besides, this is the first work that simultaneously tests growth measures and growth variation rates, which have relevant rationale and impacts. [ABSTRACT FROM AUTHOR]

We use a variety of organization-level datasets to examine the effectiveness and efficiency of the nations for the coronavirus epidemic. COVID-19 subsidies appear to have saved a significant number of jobs and maintained economic activity during the first wave of the epidemic, according to conclusions drawn from the experiences of EU member countries. General allocation rules may yield near-optimal outcomes in favor of allocation, as firms with high ecological footprints or zombie firms have lower access to government financing than more favorable, commercially owned, and export-inclination firms. Our assumptions show that the pandemic has a considerable negative impact on firm earnings and the percentage of illiquid and non-profitable businesses. Although they are statistically significant, government wage subsidies have a modest impact on corporate losses compared to the magnitude of the economic shock. Larger enterprises, which receive a lesser proportion of the aid, have more room to increase their trade liabilities or liabilities to linked entities. In contrast, according to our estimations, SMEs stand a greater danger of insolvency. [ABSTRACT FROM AUTHOR]

When predicting the defaults of a large number of samples in a region, this will be affected by industry default heterogeneity. To build a credit risk model that is more suitable for Chinese-listed firms, which have highly industry-specific default heterogeneity, we extend the forward intensity model to predict the defaults of Chinese-listed firms with information about the default heterogeneity of industries. Compared with the original model, we combine the Bayes approach with the forward intensity model to generate time-varying industry-specific default heterogeneity indicators. Our model can capture co-movements of different industries that cannot be observed based on the original forward intensity model so that the model can flexibly adjust the firm's PD according to the industry. In addition, we also consider the impact of default heterogeneity in other industries by studying the influence of the level and trends of other industries' default heterogeneity on a firm's credit risk. Finally, we compute PDs for 4476 firms from January 2001 to December 2019 for 36 prediction horizons. The extended model improves the prediction accuracy ratios both for the in-sample and out-of-sample firm's PDs for all 36 horizons. Almost all the accuracy ratios of the prediction horizons' PDs are increased by more than 6%. In addition, our model also reduces the gap between the aggregated PDs and the realized number of defaults. Our industry-specific default heterogeneity indicator is helpful to improve the model's performance, especially for predicting defaults in a large portfolio, which is of significance for credit risk management in China and other regions. [ABSTRACT FROM AUTHOR]

The International Financial Reporting Standard (IFRS) 9 relates to the recognition of an entity's financial asset/liability in its financial statement, and includes an expected credit loss (ECL) framework for recognising impairment. The quantification of ECL is often broken down into its three components, namely, the probability of default (PD), loss given default (LGD), and exposure at default (EAD). The IFRS 9 standard requires that the ECL model accommodates the influence of the current and the forecasted macroeconomic conditions on credit loss. This enables a determination of forward-looking estimates on impairments. This paper proposes a methodology based on principal component regression (PCR) to adjust IFRS 9 PD term structures for macroeconomic forecasts. We propose that a credit risk index (CRI) is derived from historic defaults to approximate the default behaviour of the portfolio. PCR is used to model the CRI with the macroeconomic variables as the set of explanatory variables. A novice all-subset variable selection is proposed, incorporating business decisions. We demonstrate the method's advantages on a real-world banking data set, and compare it to several other techniques. The proposed methodology is on portfolio-level with the recommendation to derive a macroeconomic scalar for each different risk segment of the portfolio. The proposed scalar is intended to adjust loan-level PDs for forward-looking information. [ABSTRACT FROM AUTHOR]

This study extends the literature on the capital structure by examining the effect of firm specific and macroeconomic factors on probability of default using 2385 firm year observations of Non-financial firms listed on Pakistan Stock Exchange (PSX) for the period 1998 to 2021. This is the first study that used a large dataset to analyze the default risk of Pakistani listed non-financial companies. This study follows Bharat and Shumway (2008) methodology to calculate expected default probability, which is a simplified version of the Merton (1974) structural default model. Fixed effect model has been used for data analysis. The empirical results of firm specific variables show that growth, operating cash flow ratio, liquidity, and performance is negatively related with the probability of default while leverage and tangibility of assets are positively related with the probability of default. Size of the company has no relationship with the probability of default. Macroeconomic variables economic conditions measured by GDP growth rate and index return have negative while short term interest rate have positive impact on probability of default. This study may be beneficial to the managers of non-financial companies since it may help them become more aware of the consequences of default risk and may also help them build effective policies linked to managing default risk. The board of directors of non-financial companies can extract valuable information from this study which is required to conduct control measures related to default risk management. The study excluded financial firm because those have different capital structure as compared to non-financial firms. Further studies can also investigate the effect of firm-specific factors and macroeconomic factors in different sectors to check is there any difference in results on sector basis. [ABSTRACT FROM AUTHOR]

After reaching the completion point of the Heavily Indebted Poor Countries (HIPC) initiative and the subsequent debt relief, Côte d'Ivoire decided to take the necessary measures to avoid a new public debt crisis. Ten years later, the country is at a crossroads with a rapidly growing debt. Given this situation, it is important to question the effectiveness of the methods of financing the budget deficit in Côte d'Ivoire. This article aims to develop a credit scoring model in order to analyse the efficiency of budget deficit financing modes using a sample of 3222 loan lines from the database of the Department of Public Debt and Grants of Ivory Coast. The results show that loans from bilateral and commercial banks can be considered as "risky" financing. This mode of financing has a less important impact on the probability of default. On the other hand, the results obtained with the estimation of the Logit model show that the probability of default is strongly reduced when the government is financed by bondholders and other debt instruments and multilateral institutions. [ABSTRACT FROM AUTHOR]

This paper investigates the impact of financial conditions, for example, cross‐border capital flows, interest rates and foreign exchange rates, on the well‐being of the real estate developers in five Association of Southeast Asian Nations economies. The study uses a Bottom‐up Default Analysis model to stress test their creditworthiness by reproducing the financial shocks during the global financial crisis, taper tantrum, and the U.S.‐China trade war and COVID‐19 pandemic. The median developers remain sound under the prescribed adversities. The performance is underpinned by their strong fundamentals and a conducive mix of monetary, foreign exchange, and open capital account policies by national authorities. [ABSTRACT FROM AUTHOR]

Predicting the creditworthiness of bank customers is a major concern for banking institutions, as modeling the probability of default is a key focus of the Basel regulations. Practitioners propose different default modeling techniques such as linear discriminant analysis, logistic regression, Bayesian approach, and artificial intelligence techniques. The performance of the default prediction is evaluated by the Receiver Operating Characteristic (ROC) curve using three types of kernels, namely, the polynomial kernel, the linear kernel and the Gaussian kernel. To justify the performance of the model, the study compares the prediction of default by the support vector with the logistic regression using data from a portfolio of particular bank customers. The results of this study showed that the model based on the Support Vector Machine approach with the Radial Basis Function kernel, performs better in prediction, compared to the logistic regression model, with a value of the ROC curve equal to 98%, against 71.7% for the logistic regression model. Also, this paper presents the conception of a support vector machine-based rating tool designed to classify bank customers and determine their probability of default. This probability has been computed empirically and represents the proportion of defaulting customers in each class.

In this study we consider the construction of through-the-cycle ("TTC") probability-of-default ("PD") models designed for credit underwriting uses and point-in-time ("PIT") PD models suitable for early warning uses, considering which validation elements should be emphasized in each case. We build PD models using a long history of large corporate firms sourced from Moody's, with a large number of financial, equity market and macroeconomic candidate explanatory variables. We construct a Merton model-style distance-to-default ("DTD") measure and build hybrid structural and reduced-form models to compare with the financial ratio and macroeconomic variable-only models. In the hybrid models, the financial and macroeconomic explanatory variables still enter significantly and improve the predictive accuracy of the TTC models, which generally lag behind the PIT models in that performance measure. While all classes of models have high discriminatory power by most measures, on an out-of-sample basis the TTC models perform better than the PIT models. We measure the model risk attributable to various model assumptions according to the principle of relative entropy and observe that omitted variable bias with respect to the DTD risk factor, neglect of interaction effects and incorrect link function specification has the greatest, intermediate and least impacts, respectively. We conclude that care must be taken to judiciously choose how we validate TTC vs. PIT models, as criteria may be rather different and apart from standards such as discriminatory power. This study contributes to the literature by providing expert guidance to credit risk modeling, model validation and supervisory practitioners in managing model risk.

Predicting the firm's default over time is marked persistently with certain flaws for firms seeking to avoid their exposure to credit risk and losses induced by defaults. From this perspective, a Gamma–lindley cure model, relying on a Gamma–lindley distribution and the survival analysis framework, was introduced to model the default probability and time of corporate credits referring to their inference ability once the firm is likely to default. In addition, it extends the standard mixture cure model allowing more flexibility for modeling lifetime data through relaxing the independence assumption of the probability and the time of default. Furthermore, it handles the missing defaulting labels as a latent variable and identifies three classes of default probability distribution, namely short, medium, and long-term through its potentially accurate classification method. A relevant parameter-learning procedure was described resting on the Maximum Likelihood approach. The results of an empirical analysis on a firms dataset reveal that the advanced Gamma–lindley cure model has a rather effective performance in terms of default time prediction compared to Exponential, Gamma, Weibull, and Cox proportional hazards regression modeling specifications, which yield a meaningful solution for timely credit risk management and cluster firms defaulter in three classes, enabling a deeper analysis of firm's defaults situation. The numerical experiments were conducted with Python language. • The Generalized Gamma–Lindley cure model was introduced to model corporate credits default time. • The proposed model extends the standard mixture cure model to address classification problem. • Empirical evaluation on a real-world dataset. • The elaborated model is a meaningful solution for timely credit risk management. [ABSTRACT FROM AUTHOR]

The inferred rate of default (IRD) was first introduced as an indicator of default risk computable from information publicly reported by the Bulgarian National Bank. We have provided a more detailed justification for the suggested methodology for forecasting the IRD on the bank-group- and bank-system-level based on macroeconomic factors. Furthermore, we supply additional empirical evidence in the time-series analysis. Additionally, we demonstrate that IRD provides a new perspective for comparing credit risk across bank groups. The estimation methods and model assumptions agree with current Bulgarian regulations and the IFRS 9 accounting standard. The suggested models could be used by practitioners in monthly forecasting the point-in-time probability of default in the context of accounting reporting and in monitoring and managing credit risk.

Automated Credit Scoring (ACS) is the process of predicting user credit based on historical data. It involves analyzing and predicting the association between the data and particular credit values based on similar data. Recently, ACS has been handled as a machine learning problem, and numerous models were developed to address it. In this paper, we address ACS issues concerning credit scoring in a batch of machine learning problems, namely, feature irregularities due to empty features in many records, class imbalance due to non-uniform statistical distributions of the records between classes, and concept drift due to changing statistical characteristics concerning certain classes and features with time. Considering the limited credit scoring data volume, we propose to address the challenge using the Transfer Learning with Lag (TLL) algorithm based on embedded shallow neural networks that enable knowledge transfer when the number of active features changes. Knowledge transfer is based on lags having an adaptive length that is changed based on performance change feedback. Furthermore, the framework proposes classifier aggregation and the chunk balancing mechanism for handling class imbalance. An evaluation was conducted using the Lending club, German, Default, and PPDai datasets. The results show the superiority of the proposed algorithm over the benchmarks in terms of the majority of classification metrics concerning both time series and overall results.

Many enterprises across the European Union (EU) have been hampered by the massive spread of COVID-19. It has severely impacted revenues and financial flows, potentially leading to an increase in corporate insolvency. This study investigates the influence of this new coronavirus on the solvency status of businesses in EU Member States. Several stress scenarios were constructed for non-financial listed enterprises. The results reveal a gradual surge in the possibility of default, a rise in loan repayment, and coverage being refused. According to our findings, the solvency profiles of all firms are deteriorating. Industries, such as mining, mass production, and retail, are the most susceptible to a drop in sales income and market capitalization. Before COVID-19, previous research had looked at policy options for maintaining solvency. Our data imply that a tax delay is adequate if there is a slight deterioration in the economic outlook. There should be hybrid assistance through loans and equity for even a slight deterioration in the state of an economy. This research will benefit policymakers, corporate executives, and creditors. [ABSTRACT FROM AUTHOR]

This work investigates the impact of COVID-19 on the Italian manufacturing industry, testing whether the recovery measures introduced by the government were effective in alleviating the economic consequences of the virus in 2020. In particular, this work aims to address the impact of COVID-19 on the Italian manufacturing industry, and evaluation of the adopted recovery measures. Considering the current situation, with the war in Ukraine and the related gas crisis across the European Union, such investigation on policy relief is even more relevant, contributing to the current debate. Adopting RE models, and considering the latest economic and financial information available, we analyzed active private limited firms in the Italian manufacturing industry between 2019 and 2020, investigating the impact of layoff on their productivity (i.e., Total Factor Productivity) and profitability (i.e., Return On Assets), as well as their expected probability of default. According to the results of these regression models, and assuming 8 weeks of layoff, we observed an increase in productivity (between 1.20% and 1.59%), a decrease in profitability (1.47%) and an increase in bankruptcy risk (2.27%). Hence, the relief policy was not able to alleviate the economic consequences of COVID-19 for these firms, even though the layoffs were able to support their productivity. Practical implications concern the necessary improvements for the above relief policy, i.e., interventions to support the demand of manufactured products, interventions to support the digitalization of services, interventions to support the remote working, and interventions to support the introduction of innovative products on the market. [ABSTRACT FROM AUTHOR]

Conventional credit scoring models evaluated by predictive accuracy or profitability typically serve the financial institutions and can hardly reflect their contribution on financial stability. To remedy this, we develop a novel regulatory scoring framework to quantify and compare the corresponding regulatory capital charge errors of credit scoring models. As an application of RegTech, the proposed framework considers the characteristic of example-dependence and costsensitivity in credit scoring, which is expected to enhance the ability of risk absorption of financial institutions and thus benefit the regulators. Validated on two real-world credit datasets, empirical results reveal that credit scoring models with good predictive accuracy or profitability do not necessarily provide low capital charge requirement error, which further highlights the importance of regulatory scoring framework. The family of gradient boosting decision tree (GBDT) provides significantly better average performance than industry benchmarks and deep multilayer perceptron network, especially when financial stability is the primary focus. To further examine the robustness of the proposed regulatory scoring, sampling techniques, cut-off value modification, and probability calibration are employed within the framework and the main conclusions hold in most cases. Furthermore, the analysis on the interpretability via TreeSHAP algorithm alleviates the concerns on transparency of GBDT-based models, and confirms the important roles of loan characteristics, borrowers' solvency and creditworthiness as powerful predictors in credit scoring. Finally, the managerial implications for both financial institutions and regulators are discussed. [ABSTRACT FROM AUTHOR]