• Journal of Family Resource Management and Policy Review
• /
• v.26 no.3
• /
• pp.19-34
• /
• 2022

There are growing concerns about debt insolvency among youth and low-income households. The deterioration in household debt quality among young people is due to a combination of sluggish employment, an increase in student loan burden and an increase in high-interest loans from the secondary financial sector. The purpose of this study was to explore the possibility of household debt default among young borrowers in Korea and to predict the factors affecting this possibility. This study utilized the 2021 Household Finance and Welfare Survey and used random forest algorithm to comprehensively analyze factors related to the possibility of default risk among young adults. This study presented the importance index and partial dependence charts of major determinants. This study found that the ratio of debt to assets(DTA), medical costs, household default risk index (HDRI), communication costs, and housing costs the focal independent variables.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.2
• /
• pp.119-131
• /
• 2024

In recent years, there has been a significant growth in loan comparson services offered by fintech platforms in South Korea. However, it has been reported that loan comparison platform users tend to have a higher risk of default compared to non-users. This paper investigates the difference in platform-specific credit risk factors using survival analysis models - Kaplan-Meier curves and Accelerated Failure Time (AFT) model. Our findings show that, relative to non-users, users of loan comparison platforms are characterized by elevated default rates, a greater propensity for home ownership, lower credit scores, and shorter loan durations. Furthermore, our AFT models elucidate the variance in default risk among the various loan comparison service platforms, highlighting the imperative for customized strategies that address the unique risk profiles of customers on each platform.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.419-437
• /
• 2023

As the number of thin filers in Korea surpasses 12 million, there is a growing interest in enhancing the accuracy of assessing their credit default risk to generate additional revenue. Specifically, researchers are actively pursuing the development of default prediction models using machine learning and deep learning algorithms, in contrast to traditional statistical default prediction methods, which struggle to capture nonlinearity. Among these efforts, Graph Neural Network (GNN) architecture is noteworthy for predicting default in situations with limited data on thin filers. This is due to their ability to incorporate network information between borrowers alongside conventional credit-related data. However, prior research employing graph neural networks has faced limitations in effectively handling diverse categorical variables present in credit information. In this study, we introduce the Transformer embedded Graph Convolutional Network (TeGCN), which aims to address these limitations and enable effective default prediction for thin filers. TeGCN combines the TabTransformer, capable of extracting contextual information from categorical variables, with the Graph Convolutional Network, which captures network information between borrowers. Our TeGCN model surpasses the baseline model's performance across both the general borrower dataset and the thin filer dataset. Specially, our model performs outstanding results in thin filer default prediction. This study achieves high default prediction accuracy by a model structure tailored to characteristics of credit information containing numerous categorical variables, especially in the context of thin filers with limited data. Our study can contribute to resolving the financial exclusion issues faced by thin filers and facilitate additional revenue within the financial industry.

• The Journal of Society for e-Business Studies
• /
• v.23 no.3
• /
• pp.207-224
• /
• 2018

In this article, an empirical study was conducted by using public dataset from Lending Club Corporation, the largest online peer-to-peer (P2P) lending in the world. We explore significant predictor variables related to P2P lending default that housing situation, length of employment, average current balance, debt-to-income ratio, loan amount, loan purpose, interest rate, public records, number of finance trades, total credit/credit limit, number of delinquent accounts, number of mortgage accounts, and number of bank card accounts are significant factors to loan funded successful on Lending Club platform. We developed online P2P lending default prediction models using discriminant analysis, logistic regression, neural networks, and decision trees (i.e., CART and C5.0) in order to predict P2P loan default. To verify the feasibility and effectiveness of P2P lending default prediction models, borrower loan data and credit data used in this study. Empirical results indicated that neural networks outperforms other classifiers such as discriminant analysis, logistic regression, CART, and C5.0. Neural networks always outperforms other classifiers in P2P loan default prediction.

• Journal of Families and Better Life
• /
• v.26 no.6
• /
• pp.35-50
• /
• 2008

The purpose of this study was to analyze the characteristics and financial status of credit delinquents utilizing the debt management program of the Credit Counseling and Recovery Service between January-June in 2007. Total sample of 41,355 cases was analyzed using the statistical program SPSS(Version 12.0). For analysis, descriptive statistics, F-test, Scheffe test, t-test, logit analysis and regression analysis were employed. People in the age range of 30-40s, males, high-school graduates, married couples, part-time employees, costfree residents and residents in other regions were relatively high users of the debt management program. Reasons of credit delinquency were diverse and was combined to credit default. However, increases in expenses and income reductions were found to be the most frequent reasons. Financial conditions of delinquents were worse than those of average persons shown on the national statistics. It was also found that age, sex, educational level, occupation, region of residence, home-ownership, reason of delinquency, income and total outstandings of debt were significant determinants of short-term debt burden which was measured by the ratio of monthly payment to income and long-term debt burden which was measured by repayment period.

• The Journal of Society for e-Business Studies
• /
• v.15 no.4
• /
• pp.285-301
• /
• 2010

Peer-to-peer (P2P) lending is an open marketplace for loans not from bank but from individuals online. Financial transactions are facilitated directly between individuals ("peers") without any intermediation of a traditional financial institution. A market study by renowned research company forecasts that P2P lending will grow very fast and a couple of P2P lending sites in Korea also are getting attentions by providing the alternative financial services. In P2P lending market, Lender will enjoy higher income generated from the loans in the form of interest than interest that can be earned by financial products provided by official financial institutions. Furthermore, lenders are able to decide who they would lend the money for themselves. Meanwhile, borrowers with low credit scores are able to finance their liquidity requirement with low cost and convenient access to the Internet. The objective of this paper is to introduce P2P lending and its issues of information asymmetry. We provide the insights from the case study of one of P2P lending sites in Korea and review the issues in P2P lending market as research topics. Specifically, information asymmetry issues in both traditional financial institutions and P2P lending are discussed.

• KDI Journal of Economic Policy
• /
• v.14 no.1
• /
• pp.191-212
• /
• 1992

This paper discusses the role of money in the process of capital accumulation where financial markets are impeded by contract enforcement problems in the context of overlapping generations framework. In particular, in less developed countries (LDCs) creditors may know little about the repayment capability of potential debtors due to incomplete information so that financial instruments other than money may not acceptable to them. In this paper the impediments to the operation of the private finanical markets are explicitly modelled. We argue that creditors cannot observe actual investment decisions made by the potential borrowers, and as a result, loan contracts may not be fully enforceable. Therefore, a laissez-faire regime may fail to provide the economy with the appropriate financial instruments. Under these circumstances, we introduce a government operated discount window (DW) that acts as an open market buyer of private debt. This theoretical structure represents the practice of governments of many LDCs to provide loans (typically at subsidized interest rates) to preferred borrowers either directly or indirectly through the commercial banking system. It is shown that the DW can substantially overcome impediments to trade which are caused by the credit market failure. An appropriate supply of the DW loan enables producers to purchase the resources they cannot obtain through direct transactions in the credit market. This result obtains even if the DW is subject to the same enforcement constraint that is responsible for the market failure. Thus, the DW intervention implies higher investment and output. However, the operation of the DW may cause inflation. Furthermore, the provision of cheap loans through the DW results in a worse income distribution. Therefore, there is room for welfare enhancing schemes that utilize the higher output to develop. We demonstrate that adequate lump sum taxes-cum-transfers along with the operation of the DW can support an allocation that is Pareto superior to the laissez-faire equilibrium allocation.