In a new paper, we investigate how initial public offerings affect peer-to-peer lending platforms and, more specifically, whether the platforms tend to alter their operational decisions in anticipation of going public.
Peer-to-peer lenders are essentially online services that match anonymous lenders with borrowers at a lower cost than traditional banking institutions do. Consequently, borrowers can pay lower interest rates, and lenders can potentially earn higher returns. Interest rates are usually set by an intermediary platform on the basis of the borrower’s creditworthiness (based on features such as credit score, employment status, annual income, debt-to-income ratio, and credit history). The intermediary platform generates revenue by collecting a one-time fee on funded loans (from borrowers) and by charging a loan servicing fee to investors.
The peer-to-peer lending industry has gone global. In the U.S., it started in February 2006 with the launch of Prosper, followed by LendingClub, and as of July 2018, there were at least six reputable peer-to-peer lending companies. Some of these platforms are reporting good financial performance, but others are struggling and some of gone bankrupt. This is true in most countries. For example, in China, in 2016 there were more than 4,000 platforms, but half of them had already suspended operations by 2018.
Several peer-to-peer lending platforms have decided to go public in the last few years. Examples include Funding Circle in the U.K., LendingClub in the U.S., and PPDAI Group in China. Several other platforms remain privately held (e.g., Zopa in the U.K., Prosper Marketplace in the U.S., and WeLab in China). Manipulative activities in anticipation of an IPO, though prohibited by securities regulations, are not unusual in practice. According to a 2005 article, 173 of 908 IPOs from 1998-2000 were targets of class action lawsuits. With this in mind, our paper investigates whether such activities are present in the peer-to-peer lending industry. Since the business model of such platforms relies heavily on commissions, the number of issued loans directly translates into revenue. Thus, in anticipation of its IPO, a platform may have an incentive to increase its number of issued loans to inflate its predicted market value. This strategy can pay off in the short term but may be detrimental in the long run by adversely affecting the platform’s reputation. Indeed, if the platform starts accepting low-quality loans, investors can lose trust and decide to stop investing in the platform.
In our paper, we use a comprehensive dataset from two large peer-to-peer lending platforms (denoted P1 and P2). These two platforms operate in the same country and are the two dominant platforms in terms of market share and volume. P1 went public by filing an IPO, while P2 remained privately held. We exploit this empirical environment to carefully infer the causal effect of IPOs on peer-to-peer lending platforms. Our main motivation is to answer the following questions:
- Shortly before filing its IPO, did P1 alter its lending standards by accepting different types of loans or by lowering its borrowers’ requirements?
- What was the impact of P1’s IPO on investors and on borrowers?
- Which time events (related to the IPO) had the biggest effect (e.g., IPO filing and actual IPO date)?
We estimate several econometric models and find that several performance metrics were indeed affected by the IPO, especially around the IPO filing date. Specifically, when comparing P1 loans before and after the IPO filing (while using P2 loans as a control), we observed the following:
- The loans’ performance (default rate and annual return) significantly decreased. Based on our analysis, the average default rate increased from 10.9 percent to 13.96 percent, and the average annual return decreased from 3.18 percent to 2.65 percent.
- Borrowers’ requirements (credit score and annual income) diminished. Our analysis suggests that the average credit score decreased from 702 to 696.3, and the average annual income decreased from 73,750 to 70,083.
- The acceptance rate of P1 rose from 12.70 percent to 13.18 percent.
These findings suggest that P1 may have altered some of its decisions in anticipation of its IPO by lowering the requirements for accepting loans. Namely, it is plausible that P1 issued loans based on lower credit scores and annual incomes right after its IPO filing. This strategy can potentially allow the platform to increase its short-term revenue and boost its market value. Interestingly, after the actual IPO date, all performance metrics returned to their initial levels.
This research has two main implications. First, it seems that P1 decided to prioritize short-term revenues over long-term reputation to boost its predicted market value. Second, when peer-to-peer lending platforms (or other peer-to-peer financial services) announce IPOs, it may have important implications for investors and borrowers. In light of the prospective IPOs of several unicorn sharing-economy platforms (e.g., Uber, Lyft, Airbnb), such implications will become increasingly important.
 Aggarwal RK, Purnanandam AK, Wu G (2005), Underwriter manipulation in initial public offerings, Working paper.
 We mask the platforms’ names for neutrality. We also mask the exact time of events and the currency involved in the loans.
This post comes to us from Professor Maxime Cohen at New York University and Kevin Jiao, a PhD candidate at the school. It is based on their recent paper, “The Impact of IPO on Peer-to-Peer Lending Platforms,” available here.