Loan Default Risk Research Articles

Regional redistribution through SBA guaranteed loan programs

We document and quantify a new implicit transfer mechanism in the SBA 7(a) loan program that redistributes funds between US states. We use SBA 7(a) loan data in conjunction with Dealscan private corporate loan data to show that SBA 7(a) loan interest rates are much less responsive to predicted local loan default risk compared to private corporate loans. This redistributes funds from states with low default risk to states with high default risk. These transfers are positively correlated with the severity of local economic shocks during the Great Recession and the COVID-19 Recession. Therefore, even though it was unintended, the interest rates on SBA 7(a) loans acted as an automatic stabilizer mitigating regional economic shocks.

Artificial intelligence and loan default: The case of commercial banks in India

AbstractThe supervised machine learning algorithm‐based logistic regression predicts the loan default risk by correctly identifying banks with risky/safer loans in the Indian commercial banking sector. Higher levels of interest income and spread have been found negatively associated with the probability of loan default risk, while interest expended, operating expenses and credit deposit ratio are positively related to the level of loan risk. The study recommends that a high spread, low rate of interest expended and lower operating expenses, will effectively lower the probability of loan default risk in the Indian banking sector.

KENYAN SACCO SOCIETIES IN LIQUIDITY CRUNCH; EVIDENCE FROM CORONAVIRUS PANDEMIC SHOCK

<p>The coronavirus pandemic shock has attenuated economic activity decline in various sectors of the economy across the globe. Financial service providers such as; banks, insurance firms and Sacco societies among others have been impacted by the pandemic shock as livelihoods of the people have been disrupted by the first spreading virus leading to job losses as well as weak demand and supply forces in the market. The unique model of Sacco societies’ business puts them in an even more complex situation which threatens their going concern concept. The uniqueness of the Sacco societies is that they depend on their members’ cumulated savings deposits and shares for financing rather than external borrowings from other financial institutions to conduct their business. Therefore, the study investigated the impact of the coronavirus pandemic shock on the liquidity of Sacco societies in Kenya and specifically, the study examined the covid-19 pandemic shock causal factors, institutional organizations` gaps, exclusion of Sacco societies by the national government in its mitigation measures against covid-19 pandemic shock as well as determining the response strategic interventions by the Sacco societies to stay afloat. The study found out that the majority of the Sacco societies are facing liquidity challenges more especially community, farmer and private-based Sacco societies. The loan demand has declined with long-term loans significantly impacted on by the pandemic shock. In addition, loan default has increased and non-remittance of Sacco deductions as well as withdrawal notices have also spiked up in some Sacco societies. The study discovered that the government has categorically excluded Sacco societies in its mitigation strategies to protect businesses across different sectors and in particular exclusion of Sacco societies from the credit guarantee scheme and non-classification of Sacco societies as essential service providers were cited as some of the evidence towards the exclusion concern. Besides, the Tax Laws (Amendment) Act, 2020 and Finance Bill, 2020 had very little if any touching on the plight of the Sacco societies in the wake of covid-19 pandemic shock. The response strategies instituted by Sacco societies are restructuring of loans, loan rationing, developing of new Sacco packages and recapitalization of returns as well as restricted external borrowings. An exploratory study covering four months from March to June 2020 was conducted among 120 Sacco societies from 11 counties across the country with interview schedules guiding the collection of primary data via telephone calls to the Chief Executive Officers of the targeted Sacco societies. Secondary data was obtained through document analysis of the financial reports of those societies. Judgmental techniques were employed in picking on the Sacco societies which constituted a sample size of 120 Sacco societies although only 75 responded. Data was analysed descriptively with the study findings being presented by the use of tables, graphs and pie charts. The paper recommends that the vulnerability of the Sacco societies in the wake of covid-19 pandemic shock requires immediate external interventions such as the credit guarantee schemes to cushion Sacco societies from the risk of loan default and at the same time improve their liquidity position. The perennial non-remittance of Sacco deductions by institutional organizations needs to be addressed urgently by developing new regulations to deter government organizations and private employers from unnecessarily holding members’ savings and other dues. The community, private and farmer-based Sacco societies seemed to be adversely affected by the covid-19 pandemic shock compared to the teacher and government clusters of Sacco societies. This may be due to the differential livelihoods sectoral impact of covid-19 and the weak financial framework of those Sacco societies to withstand any shock hence consolidation of Sacco societies should be seriously discussed to form entities which are strong enough to withstand shocks.</p><p>JEL: G21; G22; G32</p><p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/edu_01/0537/a.php" alt="Hit counter" /></p>

What should lenders be more concerned about? Developing a profit-driven loan default prediction model

Reliable and effective loan default risk prediction can help regulators and lenders effectively identify risky loan applicants and develop proactive and timely response measures to enhance the stability of the financial system. Traditional prediction models concentrate more on improving loan default prediction accuracy, while neglecting to take profit maximization as the goal and evaluation measure of model construction. In this study, a novel profit-driven prediction model is proposed, taking a profit indicator as the optimization objective of the Bayesian optimization to optimize the hyperparameters of the predictor-categorical boosting. The Shapley additive explanations (SHAP) value is then calculated to further interpret the relationship between the input variables and the predicted values. Based on two datasets from Renrendai and Lending Club, the experimental results and statistical test indicate that the proposed model achieves the highest profit-related evaluation metrics values, with the mean average extra profit rate values of 3.0872% and 2.1858% respectively, and the mean Profit values of 5168.8762 and 352.9787 in two datasets respectively. SHAP value further reveals the key factors that will impact predictive output, which provides more valuable information for platforms and lenders for identifying possible defaulters.

Prediction and Analysis of Financial Default Loan Behavior Based on Machine Learning Model.

In recent years, the increase of customer loan risk and the aggravation of the epidemic have led to the increase of customer default risk. Therefore, identifying high-risk customers has become an important research hotspot for banks. The customer's credit is the standard to evaluate the loan amount and interest rate, and the ability to quickly identify customer information has become a research hotspot. Based on the bank credit application scenario, this paper realizes function extraction and data processing for customer basic attribute data and download transaction data. Then, a linear regression model with penalty and a neural network prediction model are proposed to improve the accuracy of bankruptcy assessment and achieve local optimization. In this way, the implicit risk prediction and control of customer credit are improved, and the default risk of bank loans is significantly reduced. According to the characteristics of the collected sample data, the most appropriate penalty linear regression prediction algorithm is selected and the experimental analysis is carried out to improve the risk management level of banks. The experimental results show that the improved logistic regression and neural network model has obvious advantages in the prediction effect for four models.

Default prediction of online credit loans based on mobile application usage behaviors

Credit scoring is widely used by financial institutions for default prediction, however, a significant portion of online credit loan customers have inadequate or unverifiable credit histories, making it difficult for financial institutions to make effective credit decisions. Since the widespread use of smartphones and the popularity of mobile applications, it is worth investigating whether mobile application usage behaviors (App behaviors) of customers can effectively predict online loan defaults. This paper proposes a combined algorithm of CNN and LightGBM, and establishes credit scoring models with App behaviors to evaluate the default risk of online credit loans based on logistic regression, LightGBM, CNN and the combined algorithm, respectively. The experimental results suggest that App behaviors have an obvious effect on the default prediction of customers applying for online credit loans, and the combined model outperforms the other models in terms of the area under the curve (AUC). Furthermore, integrated credit scoring models are developed by combining App behaviors with traditional scoring features. A comparison of the integrated models and the traditional scoring model indicates that the integrated models have achieved a significant improvement in classification performance and App behaviors can be a powerful complement to the traditional credit scoring model.

Enterprise Financial Risk Analysis Based on Improved Model C-Means Clustering Algorithm

As a provider of loans to SMEs, banks should prudently examine loan risks while ensuring that they provide loans to SMEs from the perspective of cooperating with policy implementation and controlling their own risks. The existing loan risk measurement tools include multiple discriminant analysis models, multiple regression models, and machine learning methods. Most machine learning methods have higher prediction accuracy than traditional models when using historical data for calculation, but the existence of problems such as overfitting seriously affects the robustness of machine learning methods. A similar method is introduced into the loan default risk prediction of SMEs, and the mean clustering method is used to preset penalty items to reduce overfitting and high accuracy to help banks effectively identify the default probability of SMEs during the loan period. This study will use the mean clustering method to iteratively train 900,000 SME credit records published by the US Small and Medium Business Administration, with 27 dimensions of data provided by Small Business Administration (SBA) to provide partial guarantees. A regression tree evaluates the data, combining the scores of multiple regression trees to produce a final prediction of the probability of credit default on the input data. The research results show that the mean clustering method can effectively improve the prediction accuracy of traditional machine learning methods and multiple linear regression in the scenario of SME loan default prediction and reduce the overfitting and black-box properties. As a supplementary loan default risk measurement tool, it can strengthen the ability of commercial banks to control the risk of loan business and can also promote the development of small- and medium-sized enterprises and the market economy to a certain extent.

Loan Default Risk Assessment using Supervised Learning

Loan Default Risk Assessment using Supervised Learning

PENGARUH EVALUASI KREDIT DAN PENGAWASAN KREDIT TERHADAP RISIKO KREDIT MACET

The research background is risk of the loan credit which is the one problem which inseperable from the operations system in credit service unions in district Oebobo of Kupang City. In observation the loan default, the efforts of tackling was involved two important factors were credit evaluation and control credit. This research was conducted in 15 units of credit unions that were selected based on the criteria previously and the sample were 30 persons from credit management and supervisory elements. Technique in collecting data is observation interview, questionnnary, and documentation. The result proved that the Credit Evaluation and Credit Control had the significant partially influence to minimizing the Risk of Loan default on the credit union in district Oebobo of Kupang City through simple Linear Regression Analysis results in agreement line standard. This research also proved that the managements in credit union have to maintain and applying the evaluation and control credit system that usefull to minimize the level of loan default risk. Although it can be minimized with applying the balance credit.
 Keywords: Credit Control, Credit Evaluation, Risk of Loan

Adjustable-rate mortgages in the era of global reflation: How to model additional default risk?

We investigate the problem of interest rate risk transforming into default risk of adjustable-rate mortgage loans in the EU. Bank regulation is strikingly not neutral in this aspect, it explicitly favors short-duration adjustable-rate loans over long-duration fixed-rate loans in the framework of the gap management. This asymmetry in the regulation creates perverse incentives both for banks and households, which can lead to aggressive risk-taking, over-indebtedness of unhedged households, high procyclicality of mortgage markets, and increased systemic risks. We present a stress test model to quantify potential losses stemming from this specific risk from the perspective of lender institutions. We estimate the average extra capital that is needed to cover the additional risk of adjustable-rate mortgage loans in the EU to be 0.53% of the value of the total mortgage portfolio and 1.97% of the value of the adjustable-rate mortgage portfolio. We propose introducing a stress test model as a new mandatory element into banks’ risk management framework.

Research on Risk Prediction Model of Internet Finance Based on Cloud Computing

With the rapid development of the Internet, the traditional Internet financial risk prediction methods can no longer meet the needs of individuals and enterprises, so the concept of cloud computing arises at the historic moment. Cloud computing has subverted the traditional financial risk prediction method and has been widely studied and applied for its distributed, dynamic and autonomous characteristics. How to efficiently and reasonably schedule the resources of cloud data center and improve the accuracy of financial risk prediction is the focus of current research. How to quantify financial risk and financial risk early warning is one of the urgent problems to be solved. Under the framework of cloud computing, this paper combines the feature extraction and data weighting to study the user’s basic attribute data and a large number of downloaded APP types. After that, linear regression with penalty is used to construct the prediction model to improve insolvency. The accuracy of customer default judgment can realize local optimization, so as to improve the prediction and control of hidden risks of customer commercial bank loans and greatly reduce the default risk of bank loans.

Application of Continous Time Model in Prediction of Loss Reserves in Credit Insurance for Asset-Based Lending Companies

Asset-based lending companies and other loan providers are exposed to risk of loan defaults by borrowers. To reduce this risk, these companies acquire credit insurance. Thus when the borrower defaults in payment, the insurance company covers a percentage of the outstanding balance which generates a way to lessen and spread credit risk that the lender incurs. Therefore there are a number of methods put in place such as frequency-severity and hazard rate models used to value credit insurance. Valuing of credit insurance for asset-based lending companies is a challenging task especially in Kenyan market, where in the case of a borrower’s default, the process for recovering of the collateral will last a longer period of more than a year and where data on the borrower’s behavior of payment is of poor quality or generally unavailable. The existing methods do not consider the time to repossession of the collateral in case of loan default. Our proposed model takes into account time to repossession of the collateral and can be used in emerging market economies where other available methods may be either unsuitable or are too complex to implement due to lack of enough data. Therefore, this paper incorporates a continuous time model to forecast loss reserves in credit insurance for asset-based lending companies. First, we establish a discrete-time model to describe delinquency of credits in loan insurance product. Martingale properties, Replicating of asset portfolio strategy and Itô’s calculus are used to obtain results on expected values of future losses of credit insurance products. Secondly, we used the Black-Scholes model to develop a continuous-time model to forecast future losses in credit insurances. This is constructed by linking the latter from the discrete-time (Binomial) model using the methods of stochastic calculus. We estimated the loss reserves by applying the Geometric Brownian Motion on data of outstanding balances of 30% of 100 loanees from car loan business with assets valued at between 1 million and 10 million, to predict the probability of default of the borrower.

ВЗАЄМОДІЯ СТРАХОВИХ КОМПАНІЙ ТА БАНКІВ У ЦІЛЯХ МІНІМІЗАЦІЇ КРЕДИТНИХ РИЗИКІВ

The restoration of lending by banks in Ukraine, taking into account riskiness and minimizing credit risk, is the main strategic task of today. This article is devoted to the analysis of the use of insurance to minimize the bank's credit risk. The article examines the forms of loan insurance: loan default risk insurance; insurance by the borrower of liability for non-repayment of the loan; insurance of the subject of pledge. Actions to strengthen the financial position and status of insurance companies in Ukraine and Ukrainian society are characterized. Mechanisms for increasing responsibility for evasion of compensation payment in case of risk insurance - failure to repay the loan by the client are described. The necessity of introducing legislation regarding the regulation of the relationship between banks and insurance companies by the regulator (NBU) and the introduction of clear legislation regarding the insurance of the risk of non-repayment of the loan with the full distribution of the bank's loan portfolio, taking into account the different degrees of riskiness of borrowers, has been studied. The problem of intentional insurance of the loan portfolio by the bank, exclusively of low-quality creditors, as something that does not fall under the normative requirements of the NBU as a minimization of the bank's credit risk, is studied. In view of Ukraine's obtaining the status of a candidate for EU membership, changes to the legislation regulating the financial market according to EU standards are proposed.

COVID-19 pandemic risk and probability of loan default: evidence from marketplace lending market

As the COVID-19 pandemic adversely affects the financial markets, a better understanding of the lending dynamics of a successful marketplace is necessary under the conditions of financial distress. Using the loan book database of Mintos (Latvia) and employing logit regression method, we provide evidence of the pandemic-induced exposure to default risk in the marketplace lending market. Our analysis indicates that the probability of default increases from 0.056 in the pre-pandemic period to 0.079 in the post-pandemic period. COVID-19 pandemic has a significant impact on default risk during May and June of 2020. We also find that the magnitude of the impact of COVID-19 risk is higher for borrowers with lower credit ratings and in countries with low levels of FinTech adoption. Our main findings are robust to sample selection bias allowing for a better understanding of and quantifying risks related to FinTech loans during the pandemic and periods of overall economic distress.

MICRO FACTORS AFFECTING PERSONAL LOAN DEFAULT RISK IN COMMERCIAL BANKS IN DANANG, VIETNAM

This paper examines the micro factors affecting the risk of personal loan default using 6,287 loans to 4,023 individual customers from a commercial bank in Danang for the period between 2016 and 2020. Variables related to the characteristics of the loan, the customer, and bank’s credit officer are included in the logistic regression model. The result shows that the risk of personal loan default is related to the size and the purpose of the loan, customer’s gender and borrowing experience, and credit officer’s age. On the other hand, no relation is found between loan’s maturity duration, customer’s age, and credit officer’s gender and experience.

Evaluating Default Risk and Loan Performance in UK Peer-to-Peer Lending: Evidence from Funding Circle

The United Kingdom is the third-largest peer-to-peer (P2P) lending market in the world, which is surpassed only by the two dominant forces in P2P investing, China and the United States of America. As an innovative financial market in the UK, P2P lending brings not only many opportunities but also many risks, especially the loan default risk. In this context, this paper uses binary logistic regression and survival analysis to evaluate default risk and loan performance in UK P2P lending. The empirical results indicate that credit group, loan purpose for capital needs, sector type, loan amount, interest rate, loan term, and the age of the company all have a significant impact on the probability of loan default. Among them, the interest rate, loan term, and loan purpose for capital needs are the three most important determinants of the probability of loan defaults and survival time of loans.

SMEs Intervention Programmes in Nigeria: Evaluating Challenges Facing Implementation

To ensure price and economic stability, the central bank of Nigeria has adopted several unconventional monetary policy measure such as MSMEs credit intervention with the aim of boosting credit availability in specific sector of the economy. The intuition is that rise in productive activities/investment will indirectly promotes price stability the core mandate of the bank. Therefore, this study investigated the challenges facing implementation of real sector (MSMEs) intervention programmes of the CBN since year 2000 to 2020. The study employed mixed method using descriptive survey approach to sample 62 intervention programme implementers and 400 Micro, Small and Medium Sized Enterprises (MSMEs). The findings reveal among others that high loan default risks, politicization of programmes, and inadequate infrastructural development are the leading challenges facing programme implementers in Nigeria. Applicants' non-eligibility in programmes applied for, poor business plan or inadequate knowledge in proposed business topped the reasons for failures among applicant MSMEs. Consequently, a need for more public-private partnerships in programme design, monitoring, and evaluation to forestall political interference is advised.

THE ECONOMIC EXPLAINABILITY OF MACHINE LEARNING AND STANDARD ECONOMETRIC MODELS-AN APPLICATION TO THE U.S. MORTGAGE DEFAULT RISK

This study aims to bridge the gap between two perspectives of explainability−machine learning and engineering, and economics and standard econometrics−by applying three marginal measurements. The existing real estate literature has primarily used econometric models to analyze the factors that affect the default risk of mortgage loans. However, in this study, we estimate a default risk model using a machine learning-based approach with the help of a U.S. securitized mortgage loan database. Moreover, we compare the economic explainability of the models by calculating the marginal effect and marginal importance of individual risk factors using both econometric and machine learning approaches. Machine learning-based models are quite effective in terms of predictive power; however, the general perception is that they do not efficiently explain the causal relationships within them. This study utilizes the concepts of marginal effects and marginal importance to compare the explanatory power of individual input variables in various models. This can simultaneously help improve the explainability of machine learning techniques and enhance the performance of standard econometric methods.

Herd Behavior of Investors and Default Risks of P2P Online Lending

We examine the impact of investors’ herding on the default risk of P2P online loans. More specifically, we first decompose the number of investors in each loan into two components: the component caused by public information and the component reflecting factors other than public information. Then we investigate the effect of each component on the default risk of loans, using an ordered probit analysis. We use the data on 3,720 loans that were traded through 8percent, a Korean price-posted P2P platform from February 2015 to December 2017. We find the results as follows: First, the number of investors is determined by information that are provided by the platform and macro-economic variables (hereafter public information). Second, the number of investors explained by public information decreases the default risk of the loans. However, the number of investors reflecting factors other than public information increases the default risk of the loans. These results are interpreted as an evidence supporting ‘herding’ hypothesis: Investors follow intentionally other investors’ investment decision. These results suggest that the quantity increasement and quality improvement of public information provided by a P2P platform can improve the efficiency of P2P lending market reducing herding caused by factors other than public information.

Analyzing the Relationship between Consumer Spending Habitsto Loan Default Risk: A Case of Istanbul Turkey

The lending industry makes use of credit scoring models to classify consumers as potentially good or bad loan candidates. These credit scoring models often make use of a previous consumer's financial performance and demographics and not a consumer's money spending habits which can lead to loopholes in identifying new loan applicants as potentially having a high risk of loan default. This correlation study made use of four predictive variables, which were purchase of cigarettes, outdoor dining, entertainment, anxiety and distrust towards money and previous loan default to impart sufficient information in classifying consumers as good or bad applicants. The survey data was obtained from online surveys, completed by 166 participants and analyzed using Pearson’s correlation, logistic regression, linear regression analysis, Cronbach’s alpha, analysis of variance and covariance with the IBM SPSS Statistics 28 software. The findings of this research revealed that there was no significant correlation between the four predictive variables and loan default risk. This study showed that there is no significant relationship between the four predict variables and loan default risk, therefore money lending institutions may not include these variables in their loan applications.