peer-to-peer (P2P) lending

Carlos Serrano-Cinca, Begontilde;a Gutieacute;rrez-Nieto*

Department of Accounting and Finance, University of Zaragoza, Zaragoza, Spain

Abstract

This study goes beyond peer-to-peer (P2P) lending credit scoring systems by proposing a profit scoring. Credit scoring systems estimate loan default probability. Although failed borrowers do not reimburse the entire loan, certain amounts may be recovered. Moreover, the riskiest types of loans possess a high probability of default, but they also pay high interest rates that can compensate for delinquent loans. Unlike prior studies, which generally seek to determine the probability of default, we focus on predicting the expected profitability of investing in P2P loans, measured by the internal rate of return. Overall, 40,901 P2P loans are examined in this study. Factors that determine loan profitability are analyzed, finding that these factors differ from factors that determine the probability of default. The results show that P2P lending is not currently a fully efficient market. This means that data mining techniques are able to identify the most profitable loans, or in financial jargon, “beat the market”. In the analyzed sample, it is found that a lender selecting loans by applying a profit scoring system using multivariate regression outperforms the results obtained by using a traditional credit scoring system, based on logistic regression.

Keywords: P2P lending, microcredit, crowdfunding, banking, interest rates, credit scoring,

profit scoring, decision trees, internal rate of return.

The use of profit scoring as an alternative to credit scoring systems in peer-to-peer (P2P) lending

1. Introduction

Credit scoring poses a classification problem in that the dependent variable is dichotomous and assigns “0” to failed loans and “1” to non-failed loans. Subsequently, techniques such as logistic regression or neural networks try to estimate the borrowerrsquo;s probability of default (PD). For lenders, not only does the PD matter but also the profit gain which the loan is likely to produce. This profit gain also depends on the loss given default (the share of a loan that is lost when a borrower defaults) and on the interest rate charged [1]. Factors explaining the PD may differ from those factors explaining profits. For example, the PD of startup business loans may be higher than the PD of wedding loans; however, if a startup business loanrsquo;s interest rate is high enough, the profits from lending to entrepreneurs may be even greater than the profits from lending for weddings. Factors explaining the PD are well known: Abdou and Pointon [2] and Lessmann et al.

[3] review recent studies. However, few studies analyze the factors explaining loan profitability. This is caused by the difficulty of calculating customer profitability and the lack of necessary data [4]. The goal of this study is to develop a profit scoring Decision Support System (DSS) for investing in P2P lending.

The P2P lending market is made up of individual lenders that provide loans to individual borrowers using an electronic platform. This platform puts lenders in contact with borrowers by charging a fee. Lenders bear the full risk of this operation. Recent studies develop P2P credit scoring [5, 6, 7], although none propose profit scoring. A profit scoring DSS allows for selection of the most profitable borrowers, which is related to customer lifetime value [8]. The calculation of customer profitability for a store selling products on credit requires data from the management accounting system, such as the margin of each product sold to each customer. For financial institutions, each customer may own different products, ranging from mortgages to credit cards, and may use different channels, ranging from bank branches to online banking. All of these combined factors make it difficult to obtain precise data on customer profitability, and researchers complain about the lack of enough data to investigate profit scoring [3]. However, P2P lending platforms provide sufficient data; this is because P2P lending suffers from a severe problem of information asymmetry –lenders know little of borrowers and normally would not lend to them [9], and P2P platforms try to cope with this lack of data by disclosing as much information on borrowers as they can provide, including loan payments. Furthermore, the P2P business model is considerably leaner than the bank business model. Hence, it is feasible to calculate relevant borrower profitability measures.

This study proposes utilizing the internal rate of return (IRR) of each loan as a profitability measure. IRR is a well-known financial formula that may be easily computed for investments that have an initial cash outflow (the loan amount) followed by several cash inflows (the payments), and may contain irregular repayment schedules [10]. In the loans market, the IRR is the lenderrsquo;s effective interest rate, which may differ from the borrowerrsquo;s effective interest rate, due to delinquent loans and fees. The use of IRR has two advantages. First, IRR is a continuous variable that allows more precise information when compared to a dichotomous variable. Take, for example, three borrowers obtaining a $100 loan at a 10% interest rate: the first borrower pays back $110, the second borrower pays $102 and the third borrower pays back $5. The first loan is fully paid, while the second and third loans are considered as charged off, although the second borrower has paid most of the payments. In fact, the first loanrsquo;s IRR is 10%, the second loanrsquo;s is 2% and the third loanrsquo;s is -95%. The second advantage is that IRR takes into account not only loan payments, but loan interest rates. The riskiest loans have a high PD but also offer lenders high interest rates to compensate them for this high PD. An example is micr

剩余内容已隐藏，支付完成后下载完整资料

peer-to-peer (P2P) lending

Carlos Serrano-Cinca, Begontilde;a Gutieacute;rrez-Nieto*

Department of Accounting and Finance, University of Zaragoza, Zaragoza, Spain

Abstract

This study goes beyond peer-to-peer (P2P) lending credit scoring systems by proposing a profit scoring. Credit scoring systems estimate loan default probability. Although failed borrowers do not reimburse the entire loan, certain amounts may be recovered. Moreover, the riskiest types of loans possess a high probability of default, but they also pay high interest rates that can compensate for delinquent loans. Unlike prior studies, which generally seek to determine the probability of default, we focus on predicting the expected profitability of investing in P2P loans, measured by the internal rate of return. Overall, 40,901 P2P loans are examined in this study. Factors that determine loan profitability are analyzed, finding that these factors differ from factors that determine the probability of default. The results show that P2P lending is not currently a fully efficient market. This means that data mining techniques are able to identify the most profitable loans, or in financial jargon, “beat the market”. In the analyzed sample, it is found that a lender selecting loans by applying a profit scoring system using multivariate regression outperforms the results obtained by using a traditional credit scoring system, based on logistic regression.

Keywords: P2P lending, microcredit, crowdfunding, banking, interest rates, credit scoring,

profit scoring, decision trees, internal rate of return.

The use of profit scoring as an alternative to credit scoring systems in peer-to-peer (P2P) lending

1. Introduction

Credit scoring poses a classification problem in that the dependent variable is dichotomous and assigns “0” to failed loans and “1” to non-failed loans. Subsequently, techniques such as logistic regression or neural networks try to estimate the borrowerrsquo;s probability of default (PD). For lenders, not only does the PD matter but also the profit gain which the loan is likely to produce. This profit gain also depends on the loss given default (the share of a loan that is lost when a borrower defaults) and on the interest rate charged [1]. Factors explaining the PD may differ from those factors explaining profits. For example, the PD of startup business loans may be higher than the PD of wedding loans; however, if a startup business loanrsquo;s interest rate is high enough, the profits from lending to entrepreneurs may be even greater than the profits from lending for weddings. Factors explaining the PD are well known: Abdou and Pointon [2] and Lessmann et al.

[3] review recent studies. However, few studies analyze the factors explaining loan profitability. This is caused by the difficulty of calculating customer profitability and the lack of necessary data [4]. The goal of this study is to develop a profit scoring Decision Support System (DSS) for investing in P2P lending.

The P2P lending market is made up of individual lenders that provide loans to individual borrowers using an electronic platform. This platform puts lenders in contact with borrowers by charging a fee. Lenders bear the full risk of this operation. Recent studies develop P2P credit scoring [5, 6, 7], although none propose profit scoring. A profit scoring DSS allows for selection of the most profitable borrowers, which is related to customer lifetime value [8]. The calculation of customer profitability for a store selling products on credit requires data from the management accounting system, such as the margin of each product sold to each customer. For financial institutions, each customer may own different products, ranging from mortgages to credit cards, and may use different channels, ranging from bank branches to online banking. All of these combined factors make it difficult to obtain precise data on customer profitability, and researchers complain about the lack of enough data to investigate profit scoring [3]. However, P2P lending platforms provide sufficient data; this is because P2P lending suffers from a severe problem of information asymmetry –lenders know little of borrowers and normally would not lend to them [9], and P2P platforms try to cope with this lack of data by disclosing as much information on borrowers as they can provide, including loan payments. Furthermore, the P2P business model is considerably leaner than the bank business model. Hence, it is feasible to calculate relevant borrower profitability measures.

This study proposes utilizing the internal rate of return (IRR) of each loan as a profitability measure. IRR is a well-known financial formula that may be easily computed for investments that have an initial cash outflow (the loan amount) followed by several cash inflows (the payments), and may contain irregular repayment schedules [10]. In the loans market, the IRR is the lenderrsquo;s effective interest rate, which may differ from the borrowerrsquo;s effective interest rate, due to delinquent loans and fees. The use of IRR has two advantages. First, IRR is a continuous variable that allows more precise information when compared to a dichotomous variable. Take, for example, three borrowers obtaining a $100 loan at a 10% interest rate: the first borrower pays back $110, the second borrower pays $102 and the third borrower pays back $5. The first loan is fully paid, while the second and third loans are considered as charged off, although the second borrower has paid most of the payments. In fact, the first loanrsquo;s IRR is 10%, the second loanrsquo;s is 2% and the third loanrsquo;s is -95%. The second advantage is that IRR takes into account not only loan payments, but loan interest rates. The riskiest loans have a high PD but also offer lenders high interest rates to compensate them for this high PD. An example is micr

剩余内容已隐藏，支付完成后下载完整资料

资料编号：[20774]，资料为PDF文档或Word文档，PDF文档可免费转换为Word