Investor churn analysis in a P2P lending market

ABSTRACT

This study aims to determine why investors leave a peer-to-peer (P2P) lending market, and the pathways through which several factors affect investor churn. To this end, this study performed Cox proportional hazard and Kaplan-Meier survival analyses, as well as a linear regression analysis of data from Moneyauction, Korea’s first, but recently failed, P2P platform. The results show that investors’ investment patterns (i.e., investment frequency, amount, and method) affect churn directly or indirectly via investment profitability. This result supplements the findings in the stock market that the higher the trading frequency and volume are, the lower the returns are in the P2P lending market. Furthermore, this study is the first to cover the issue of investor churn in P2P lending markets and to reveal the detailed pathways of investors’ investment patterns in both the P2P and traditional financial markets in terms of profitability and churn.

KEYWORDS:

• Investment pattern
• churn
• survival analysis
• repayment performance
• P2P lending

JEL CLASSIFICATION:

• D81
• D90
• M31
• G41

Disclosure statement

The authors declare no potential conflicts of interest with respect to the authorship and/or publication of this article.

Notes

¹ Customer churn is generally defined as the propensity of customers to discontinue their business relationships with a company in a given period (Dyché 2002); however the detailed definition varies by study.

² In the current study, churn was defined as actual cessation of investment activity, not the propensity to do so. More specifically, a churned investor refers to an investor who had no single investment activity in the last six months as of the end of September 2016, and the time of churn was set as the most recent investment date.

³ Transforms to dichotomous values were made to limit the extreme value effect and the misreading effect of naïve continuous return figures. Some cases of positive returns were found to have extremely high returns due to the extended overdue period, while some cases of negative returns had negative extremes due to small repayment. Furthermore, without knowing the actual loan period due to the limitations of available data, the original continuous return rate showed a tendency to mislead the repayment performance in that the return rate was relatively low in the case of prepayment because it was repaid earlier than the planned maturity, while the return rate tended to be high in cases of overdue repayment.

⁴ https://www.scmp.com/business/banking-finance/article/2180731/chinas-p2p-lending-crisis-worsens-second-firm-runs-trouble.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.