Loan Default Research Articles

Does tight monetary policy threaten the soundness of Korean banks?

This study analysed the relationship between the implementation of the Central Bank of Korea's tight monetary policy due to the recent deepening of inflation and the soundness of Korean banks. This study focused on the fact that the recently published research results of Jung (2023) were different from previous research results. The traditional result is that tight monetary policy reduces loan defaults by shrinking banks' lending capacity. However, Jung (2023), who analysed banks in the EU, claims that loan defaults do not increase despite easing monetary policy. This study empirically analysed the risk indicators that evaluate the financial soundness of banks by dividing them into the financial insolvency, financing costs, recapitalization, and risk holding. The research results are as follows. First, tight monetary policy reduces banks' loan defaults in line with Gambacorta and Mistrulli (2004) instead of Jung (2023). Second, banks' financing costs increased due to the tight monetary policy with same context to the claims of Borio and Hofmann (2017) and Jung (2023). Third, banks' recapitalization level actually decreased due to tight monetary policy as argument of Jung (2023). Fourth, tight monetary policy had no significant relationship with banks' risk taking behaviour in consistent with Diamond and Rajan (2011) and Jimenez et al. (2012). In conclusion, tight monetary policy reduces insolvency of loans, but increases potential risk factors such as increased financing costs and reduced recapitalization in the future. Therefore, this study suggests novel findings and policy challenges.

COMPARATIVE ANALYSIS OF RF, SVR WITH GAUSSIAN KERNEL AND LSTM FOR PREDICTING LOAN DEFAULTS

This investigation elucidates the paramount endeavour of predicting loan defaults, which is imperative for the efficacious management of financial risk and the overall stability of financial institutions. Conventional statistical methodologies frequently encounter challenges in effectively capturing the nonlinear and sequential dynamics inherent in financial data, thereby necessitating the examination of more sophisticated machine learning methodologies. This research reports an experimental-based comparative evaluation of three ML and DL models—Long Short-Term Memory (LSTM) networks, Random Forest (RF), and Support Vector Regression (SVR)—to assess their efficacy in forecasting loan defaults. The models are evaluated using metrics such as Mean Squared Error (MSE), F1 score, and Accuracy, and their proficiency in addressing imbalanced datasets and elucidating intricate data relationships is highlighted. The results indicate that while the Random Forest model surpasses its counterparts in terms of accuracy and MSE, the LSTM model exhibits considerable potential in managing imbalanced data, as evidenced by its stable F1 score. Although SVR reveals competitive precision, it exhibits deficiencies in addressing class imbalance. The ANOVA analyses substantiate that the disparities in model performance are statistically significant. The research acknowledges that both the LSTM and SVR models remain in the developmental stages, with ongoing initiatives aimed at refining these models through hyperparameter optimization and advanced architectural frameworks to enhance their predictive efficacy in practical applications.

Personal Loan Default Prediction Analysis based on TabNet and Logistic Regression

Abstract. With the continuous development of the global economy, the demand for loans from individuals and enterprises is growing. However, loan defaults have gradually become a major challenge facing the financial industry. Loan defaults not only directly affect the profitability of financial institutions, but may also trigger systemic risks and pose a potential threat to the entire economic system. Therefore, improving the accuracy of loan default prediction is crucial for financial institutions to effectively manage credit risks. This study built a system for personal loan default prediction by combining TabNet with the Logistic regression model. Through feature engineering, this study extracts potential credit risk features by utilizing datasets from personal and Internet loans. The accuracy of default prediction is improved by combining TabNet's feature learning capabilities with Logistic Regression's interpretability. An AUC value of 0.89 was achieved by the integrated model, which achieved a notable performance. The results indicate that a system that relies on machine learning to predict defaults can significantly enhance the quality of credit approval decisions and lower the likelihood of bad debts for financial institutions. Future studies could aim to optimize the feature selection process, experiment with more advanced machine learning algorithms, or apply the model to diverse loan datasets, thereby enhancing both its generalization and accuracy. In conclusion, this study offers a novel approach to loan default prediction, demonstrating significant practical value and providing substantial support for the risk management and decision-making processes of financial institutions.

Evolving Transparent Credit Risk Models: A Symbolic Regression Approach Using Genetic Programming

Credit scoring is a cornerstone of financial risk management, enabling financial institutions to assess the likelihood of loan default. However, widely recognized contemporary credit risk metrics, like FICO (Fair Isaac Corporation) or Vantage scores, remain proprietary and inaccessible to the public. This study aims to devise an alternative credit scoring metric that mirrors the FICO score, using an extensive dataset from Lending Club. The challenge lies in the limited available insights into both the precise analytical formula and the comprehensive suite of credit-specific attributes integral to the FICO score’s calculation. Our proposed metric leverages basic information provided by potential borrowers, eliminating the need for extensive historical credit data. We aim to articulate this credit risk metric in a closed analytical form with variable complexity. To achieve this, we employ a symbolic regression method anchored in genetic programming (GP). Here, the Occam’s razor principle guides evolutionary bias toward simpler, more interpretable models. To ascertain our method’s efficacy, we juxtapose the approximation capabilities of GP-based symbolic regression with established machine learning regression models, such as Gaussian Support Vector Machines (GSVMs), Multilayer Perceptrons (MLPs), Regression Trees, and Radial Basis Function Networks (RBFNs). Our experiments indicate that GP-based symbolic regression offers accuracy comparable to these benchmark methodologies. Moreover, the resultant analytical model offers invaluable insights into credit risk evaluation mechanisms, enabling stakeholders to make informed credit risk assessments. This study contributes to the growing demand for transparent machine learning models by demonstrating the value of interpretable, data-driven credit scoring models.

Risk Typologies and Their Mitigation in Banking: A Study of Azerbaijani Banks

Abstract This paper explores the typologies of risks in the banking sector and examines the strategies employed to mitigate them, with a focus on Azerbaijani banks. The study identifies key risk categories, including credit risk, market risk, liquidity risk, and operational risk, each of which plays a critical role in the financial health and profitability of banks. The research highlights that credit risk, driven by loan defaults, remains the most significant concern for Azerbaijani banks, while market risks, particularly from currency fluctuations, also pose major challenges. Effective risk mitigation strategies, including collateralization and diversified lending practices, are essential to maintaining stability in this emerging market. Additionally, the study emphasizes the importance of integrating modern risk management technologies and adhering to international regulatory standards. Azerbaijani banks that adopt advanced monitoring systems and proactive risk management frameworks demonstrate improved financial performance and resilience in the face of economic volatility. The findings underscore that well-implemented risk management practices are crucial for enhancing profitability and ensuring long-term sustainability in the banking sector.

Credit Anomaly Detection Method based on Bayesian Networks

Due to the virtual nature of online transactions and the underdeveloped personal credit risk assessment system of Internet finance lending platforms in China, these platforms often struggle to verify the creditworthiness of applicants prone to fraud and default, leading to loan losses. As a result, risk control has become the core of Internet lending platforms. Modern risk control relies on statistical and data mining techniques to analyze and model data, uncover patterns that indicate loan defaults, and identify anomalies. This paper proposes a Bayesian network-based credit anomaly detection method, which detects anomalies by calculating and ranking the joint anomaly probabilities of sample instances. The technique operates under unsupervised learning and can effectively handle missing and imbalanced data. This method analyzes a loan credit dataset with 32,581 samples and 12 features of the customers. The result of the analysis demonstrates the feasibility and effectiveness of this method in detecting anomalies in online loans.

Elucidating the Non-Performing Assets in Indian Banking Sector: Does Macroeconomic Factor Matter?*

Using the system generalized methods of moments on panel data set of 45 selected Indian commercial banks from 2005–2006 to 2019–2020, the study investigates the determinants of non-performing assets (NPAs), taking into account macroeconomic and bank-specific factors. The study finds evidence of a persistent effect in stressed assets. The profitability ratio, operational capacity, cost efficiency and aggressive lending negatively and significantly affect the NPAs. The negative impact of the profitability ratio implies that less efficient banks are more exposed to bad loans. While cost efficiency discourages the NPAs, on the other hand, higher credit deposit ratio (operational capacity) suggests that a well-diversified portfolio leads to lesser probability of loan defaults. The probability of loan default is high when more advances are released to the economy’s weaker and vulnerable sectors (high-priority sector lending). Small-sized banks are more efficient in NPAs management than large-sized banks. Various macroeconomic variables, such as lending rates and credit growth in the banking industry, significantly affect the NPAs. The higher lending rate adversely affects borrowers’ repayment capacity, leading to more stressed assets. Therefore, to maintain bad assets within a bearable limit, policymakers should take macroeconomic determinants into consideration.

Sunshine and bank lending

This study uses a unique proprietary loan-level dataset from a state-owned Chinese commercial bank to examine how sunshine-induced emotions influence loan pricing decisions made by high-level loan officers who would be punished for loan defaults. Our findings reveal that an increased intensity of sunshine at loan officers' workplace can increase loan pricing for borrowers. The result can be explained by the mood channel: sunshine induces a positive mood among loan officers leading to greater risk aversion in their loan pricing decisions, known as the mood maintenance effect. This effect is magnified for firms with high risk, high bank–firm information asymmetry, and in locations characterized by less bank competition. Our main results are still valid after taking into account various robustness issues and endogeneity concerns.

The impact of climate change on banks loan portfolios and strategies for effective climate risk management

As extreme weather events continue to impact every continent and the world moves towards establishing a lower-carbon economy, the banking industry is expected to incorporate climate risk into their risk management practices. Climate change poses significant risks to bank loan portfolios through increased physical and transition risks. This study systematically analysed the literature to identify effective strategies for managing these risks. Our findings reveal that climate-related events can lead to substantial loan defaults and credit losses. To mitigate these impacts, banks can integrate climate risk into their risk management frameworks, adopt sustainable lending practices, and diversify their portfolios. Some banks have already implemented measures to mitigate climate risk through insurance policies, while others are incorporating sustainability criteria into their lending practices, such as financing green projects. By proactively addressing climate risks, banks can protect their portfolios, enhance financial resilience and contribute to a low-carbon economy.

Evaluating the impact of demographic characteristics on residential mortgage default risk: Evidence from Lebanon

Purpose: This study aims to examine the relationship between borrower's demographic characteristics and default risk in mortgage loans to help financial institutions develop more effective lending policies. Design/Methodology/Approach: Cross-sectional data were elicited from randomly selected 6743 individual accounts from Lebanese housing banks. This study applied the binary logistic and stepwise regression models to analyze the dataset using the Stata statistical software. Model diagnosis is performed using the Hosmer-Lemeshow goodness of fit test, likelihood ratio test, model accuracy classification table and statistically significant test-ROC curve. Findings: The findings revealed that there is a significant relationship between residential loan default risk and borrower’s marital status, nature of job occupation, job economic sector, job location and loan purpose. The performance of the binary logistic regression analysis demonstrates the overall percentage who is correctly classified is 91.61%. Conclusion: The log odds of default risk for widowed borrowers are about 90 percent higher than those of divorced borrowers and that of self-employed borrowers is about 54 percent higher than that of employed borrowers. Borrowers working in the banking and real estate sectors have lower default rates than borrowers working in other economic sectors. In addition, loans granted for renovation purposes have the lowest default rates compared to loans provided for purchase, under-construction and construction purposes. Practical Implications: The empirical results help financial institutions to have early warning signals in detecting financial distress and to differentiate between a high- and low-risk group of borrowers, helping in the development of tailored risk mitigation strategies and adjusting the lending criteria.

Risk Analysis of Mortgage Loan Default for Bank Customers and AI Machine Learning

Abstract Risk Analysis of Mortgage Loan Default for Banks is a crucial issue. On one hand, it concerns the quality of the bank's credit decisions, and on the other hand, it affects the rights of homebuyers to obtain financial support. In 2008, the U.S. subprime mortgage crisis sparked a global financial meltdown. What began with the collapse of the housing market quickly spread throughout the global financial system, resulting in the failure of numerous banks and a widespread economic recession. Recently, the banking sector has increasingly leveraged Artificial Intelligence (AI) and Machine Learning (ML) to enhance decision-making processes, particularly in assessing the risk of mortgage loan defaults. This paper aims to explore the application of ML techniques to predict and analyze the risk of default among bank customers, thereby enabling financial institutions to make more accurate and informed lending decisions. Keywords: Mortgage Loan Default Risk Analysis AI ML K-MANS LTV.

Technology-Driven Financial Risk Management: Exploring the Benefits of Machine Learning for Non-Profit Organizations

This study explores how machine learning can optimize financial risk management for non-profit organizations by evaluating various algorithms aimed at mitigating loan default risks. The findings indicate that ensemble learning models, such as random forest and LightGBM, significantly improve prediction accuracy, thereby enabling non-profits to better manage financial risk. In the context of the 2008 subprime mortgage crisis, which underscored the volatility of financial markets, this research assesses a range of risks—credit, operational, liquidity, and market risks—while exploring both traditional machine learning and advanced ensemble techniques, with a particular focus on stacking fusion to enhance model performance. Emphasizing the importance of privacy and adaptive methods, this study advocates for interdisciplinary approaches to overcome limitations such as stress testing, data analysis rule formulation, and regulatory collaboration. The research underscores machine learning’s crucial role in financial risk control and calls on regulatory authorities to reassess existing frameworks to accommodate evolving risks. Additionally, it highlights the need for accurate data type identification and the potential for machine learning to strengthen financial risk management amid uncertainty, promoting interdisciplinary efforts that address broader issues like environmental sustainability and economic development.

Using machine learning to investigate the determinants of loan default in P2P lending: Are there differences between before and during COVID-19?

The coronavirus disease (COVID-19) has led to a persistent increase in the volatility of the credit market and triggered a series of financial distress and bankruptcy. To investigate whether there are differences in loan default determinants before and during COVID-19 and to identify the most effective predictors of loan default during COVID-19, this study employs machine learning methods to establish a comprehensive loan default prediction model for Peer-to-peer (P2P) lending based on four perspectives: loan characteristics, credit transaction history, personal information, and macroeconomic environment. The results show that the EXtreme Gradient Boosting (XGBoost) outperforms the other models and that credit transaction history plays a vital role in forecasting loan default over the two periods. We also find discrepancies between the effects of consumer price index, purchasing manager’ index, and the number of bidders on loan default before and during the pandemic. Our study contributes to related research fields on loan default prediction by identifying loan default determinants that are more applicable to unstable periods and investigating the impact of COVID-19 on default predictions. Meanwhile, our findings can provide P2P lending investors, platforms, and policymakers with practical implications to reduce uncertainty and losses that result from similar black swan events.

Character Counts: Psychometric-Based Credit Scoring for Underbanked Consumers

Psychometric-based credit scores measure important personality traits that are characteristic of good borrowers’ behaviors. While such data can potentially improve credit models for underbanked consumers, the utility of psychometric data in consumer lending is still largely understudied. The present study contributes to the literature in this respect, as it is one of the first studies to evaluate the efficacy of psychometric-based credit scores for predicting future loan defaults among underbanked consumers. The results from two culturally diverse samples of loan applicants (Sub-Saharan Africa, n = 1113; Western Europe, n = 1033) found that psychometric scores correlated significantly with future loan defaults (Gini = 0.28–0.31) and were incrementally valid above and beyond the banks’ own credit scorecards. These results highlight the theoretical basis for personality in financial behaviors, as well as the practical utility that psychometric scores can have for credit decisioning in general and the facilitation of financial inclusion for underbanked consumer groups in particular.

Neyman-Pearson Multi-class Classification via Cost-sensitive Learning

Most existing classification methods aim to minimize the overall misclassification error rate. However, in applications such as loan default prediction, different types of errors can have varying consequences. To address this asymmetry issue, two popular paradigms have been developed: the Neyman-Pearson (NP) paradigm and the cost-sensitive (CS) paradigm. Previous studies on the NP paradigm have primarily focused on the binary case, while the multi-class NP problem poses a greater challenge due to its unknown feasibility. In this work, we tackle the multi-class NP problem by establishing a connection with the CS problem via strong duality and propose two algorithms. We extend the concept of NP oracle inequalities, crucial in binary classifications, to NP oracle properties in the multi-class context. Our algorithms satisfy these NP oracle properties under certain conditions. Furthermore, we develop practical algorithms to assess the feasibility and strong duality in multi-class NP problems, which can offer practitioners the landscape of a multi-class NP problem with various target error levels. Simulations and real data studies validate the effectiveness of our algorithms. To our knowledge, this is the first study to address the multi-class NP problem with theoretical guarantees. The proposed algorithms have been implemented in the R package npcs, which is available on CRAN.

EFFECT OF SAVING LITERACY ON INVESTMENT DECISION AMONG SECONDARY SCHOOL TEACHERS IN MERU COUNTY

The education sector in Kenya receive the largest share of the country's budget, with a significant portion allocated to paying teachers' salaries, as they constituted a large portion of the employed workforce. This study aimed to assess the effect of saving literacy on investment decisions among public secondary school teachers in Meru County, Kenya. It investigated why teachers in Meru County struggled to invest for their future and explored the role of their financial literacy in contributing to this issue, as indicated in the Solution SACCO annual report for the year 2022. The study was anchored under Financial Literacy Theory. Empirical analysis was conducted based on the study's objectives and conceptual framework. The target population for the study consisted of 1,825 teachers from public secondary schools in Meru County. A descriptive research design was employed, gathering data from primary and secondary sources. Purposive sampling was used to select accessible schools in the nine sub-counties within Meru County, while a simple random sampling technique was applied to select 328 respondents. A self-administered questionnaire was tested on 30 respondents from Tharaka Nithi County, constituting 10% of the sample size, and feedback from the university supervisors was used to address any ambiguities or irrelevant questions. The researcher distributed and collected questionnaires using the drop-and-pick method. After collecting the data from the field, it was reviewed to identify errors such as spelling mistakes and unanswered questions. The data was then coded and entered into the Statistical Package for Social Sciences (SPSS) for analysis. The findings were presented through tables, and regression analysis, ANOVA tests, and coefficients of determination were conducted to examine correlations and establish the model equation. Descriptive statistics were used to calculate mean values and standard deviations, and Pearson correlation analysis was employed for hypothesis testing. Subsequently, the findings were compiled into summaries, reports, and frequency distribution tables. The analysis of multiple regression indicated that the R^2 value stood at is 0.691, suggesting that the variables related to financial literacy accounts for 69.1% of change for variation in investment decisions made by TSC teachers in public secondary schools in Meru County. The findings concluded that employees generally exhibit prudent financial practices, as seen in their efforts to avoid loan defaults, read credit terms carefully, repay borrowed money promptly, and use unsecured loans judiciously. The study recommended for adoption of a culture that emphasizes saving and investing among teachers in the Teaching Service Commission (TSC) and suggest for an attempt to explore non-economic elements that might affect teachers' investment decision-making processes. KEY WORDS; saving, financial literacy,

Implementation of Indonesian Banking Legal Policy in Consumer and Creditor Protection

This article examines the implementation of Indonesian banking legal policy in the protection of consumers and creditors. The study employs a normative juridical approach to evaluate existing banking regulations and their effectiveness in practice. The primary focus is on the legal protections provided to banking consumers and creditors in cases of loan defaults and dispute resolution. The findings reveal that although Indonesia’s banking regulations have undergone several updates to enhance consumer and creditor protection, there remains a significant gap between the policies and their practical application. The article identifies several factors contributing to this gap, including insufficient regulatory oversight, the complexity of legal procedures, and a lack of consumer awareness regarding their rights. To improve protection, the article recommends strengthening the regulatory framework, enhancing the capacity of supervisory institutions, and intensifying public education on consumer and creditor rights in the banking sector.

Loan Defaults Prediction Based on Stacked Models Trained by Personalized Features

Training one machine learning model with features that all clients have will result in a waste of features, which is likely to adversely affect the model’s performance. To solve the problem, the study attempts a new method, which it to train an individual stacked model for each loan client based on personalised features. Data used contains information of about fifteen million loan applicants, their default status, and 468 features in all. 41 of the features that can be quantitively analysed are selected according to the feature importance output by a Random Forest model. Default prediction of every client is made by a stacked model trained with all selected features he/she has. The stacked model consists of two layers, in which a Light Gradient-Boosting Machine (LGBM) classifier is the base learner, and a Logistic Regression model is the meta learner. As the defaulters account for only 3.14%, which is significantly unbalanced, Area Under the Curve (AUC) and F1 scores are employed to evaluate the method, instead of accuracy. Test results show that models trained by personalised features perform better than the ones trained by shared features. Additionally, the stacked model outperforms individual Logistic Regression model, but performs nearly the same as individual LGBM Classifier. In detailed, the stacked models trained with personalised features result in AUC=0.772 and F1=0.188. Due to data unbalance, although the method’s F1 score is relatively low, it’s considered to be passable. In the future, stacked models combining different models will be attempted.

Attention-based dynamic multilayer graph neural networks for loan default prediction

Whereas traditional credit scoring tends to employ only individual borrower- or loan-level predictors, it has been acknowledged for some time that connections between borrowers may result in default risk propagating over a network. In this paper, we present a model for credit risk assessment leveraging a dynamic multilayer network built from a Graph Neural Network and a Recurrent Neural Network, each layer reflecting a different source of network connection. We test our methodology in a behavioural credit scoring context using a dataset provided by U.S. mortgage financier Freddie Mac, in which different types of connections arise from the geographical location of the borrower and their choice of mortgage provider. The proposed model considers both types of connections and the evolution of these connections over time. We enhance the model by using a custom attention mechanism that weights the different time snapshots according to their importance. After testing multiple configurations, a model with GAT, LSTM, and the attention mechanism provides the best results. Empirical results demonstrate that, when it comes to predicting probability of default for the borrowers, our proposed model brings both better results and novel insights for the analysis of the importance of connections and timestamps, compared to traditional methods.

설명 가능한 인공지능을 활용한 은행대출연체 예측 모델 연구

[Purpose] This study aims to create a loan default prediction model using machine learning, a core concept of AI technology, and explainable AI (XAI) techniques, and to compare and evaluate it against highly interpretable models such as logistic regression. [Methodology] This study utilizes the data of household loan customers of Local Bank A in South Korea from December 2020 to June 2022 to generate and compare the logistic regression model and the XGBoost model. We used SHAP, one of the XAI techniques, to explain the XGBoost model. [Findings] The XGBoost model shows better prediction performance than the logistic regression model. By applying SHAP to the XGBoost model, we identify key predictors： credit rating, average bank deposit balance over the past six months, card loan risk rating, K rating agency’s multiple debt risk index, average core deposit balance over the past six months, and the current interest rate of the loan account. [Implications] The machine learning model (XGBoost) combined with XAI shows better prediction performance than the logistic regression model, and the key predictors identified by the two models are similar. Therefore, the combined model could serve as an alternative to the logistic regression model commonly used in business research.