Can Listing Information Indicate Borrower Credit Risk in Online Peer-to-Peer Lending?

ABSTRACT

Effective assessment of borrower credit risk is the greatest challenge for peer-to-peer (P2P) lenders, especially in the Chinese market, where borrowers lack widely recognized credit scores. In this study, based on credit data from 2012 to 2015 from the website Renrendai.com, a logit model was used to assess borrower credit risk and predict the probability of default in every out-of-sample listing. The predicted probability of default was then compared with the actual default observation of default. The empirical results show that the logit model can evaluate the credit risk of P2P borrowers, and the model reduces the default rate to 9.5%, compared with the total sample default rate of 16.5%.

KEY WORDS:

• credit-risk identification
• credit score
• peer-to-peer lending

Acknowledgments

The authors are most grateful to the Editor, Ali Kutan, and the anonymous referees for their comments and suggestions.

Notes

^1. For the traditional P2P business, P2P platforms do not involve in lending activities, and only charge intermediary fee for providing information which consist with fixed administration fee and service fee for different level of platform users. The rate of fee depends on credit level of borrowers. This kind of business is the research subject of this article. During the development of P2P business in China, some P2P platforms once provided part of guarantee for the borrowers to attractive investors. But in August, 2016, Chinese government largely reformed the industry, requiring that P2P lending platforms conduct business only as information intermediary and not be involved in lending activities. The subject of this study was only the information intermediary business, because the regulation does not affect this research.

^2. But in 2014 and 2015, Lending Club and Prosper stopped to provide public data, because the data has regard as intellectual poverty and competitive advantages.

^3. “National P2P Lending Industry Bulletin” issued by Lending House, a well-known third-party online loan consulting platform in China.

^4. Individual credit registries established by Chinese officials cover only a quarter of the population and without credit score. Coverage of personal credit score established by third-party companies is still insufficient, such as the Alibaba credit score, which is now applied by some commercial businesses, but the main basis of scores is consumer-to-consumer transactions and the payment data of a company. Consequently, third-party credit scores are not applied in P2P lending.

^5. To establish their identity, borrowers are required to upload a photo ID. To show their academic bona fides, borrowers are required to provide an academic certificate number that can be checked on the official website. Borrowers who own property must also provide an official certificate of property. To provide their income borrowers must provide documents such as employer certificate or kinking financial records.

^6. Rate of Type II errors = listings of Type II errors/predicted listings of non-defaulted loans. Rate of Type I errors = listings of Type I errors/predicted listings of defaulted loans.

Additional information

Funding

The authors greatly appreciate the financial support from the National Natural Science Foundation of China (NSFC) and Economic and Social Research Council of UK (ESRC) [71661137006].