Screening Peers Softly: Inferring the Quality of Small Borrowers

Abstract

This paper examines the performance of new online lending markets that rely on non-expert individuals to screen their peers' creditworthiness. We find that these peer lenders predict an individual's likelihood of defaulting on a loan with 45% greater accuracy than the borrower's exact credit score (unobserved by the lenders). Moreover, peer lenders achieve 87% of the predictive power of an econometrician who observes all standard financial information about borrowers. Screening through soft or nonstandard information is relatively more important when evaluating lower quality borrowers. Our results highlight how aggregating over the views of peers and leveraging nonstandard information can enhance lending efficiency.JEL codes: D53, D8, G21, L81

Crowdfunding models: Keep‐It‐All vs. All‐Or‐Nothing

“…Our study further complements the one by Iyer et al. () that studies the contributions of individuals in the form of loan‐based crowdfunding, evidencing that peer lenders are good at predicting default risk and are able to assess the risk of crowdfunded projects.…”

“…These platforms enable clear mechanisms through which individuals can provide money for, or even invest in, early‐stage entrepreneurial firms (Belleflamme, Lambert, & Schwienbacher, ; Mollick, ; Vismara, ). Understanding how crowdfunding works has attracted increasing interest from research scholars (Burtch, Ghose, & Wattal, ; Chemla and Tinn, ; Iyer, Khwaja, Luttmer, & Shue, ; Kuppuswamy & Bayus, ; Roma, Petruzzelli, & Perrone, ; Stanko and Henard, ; Yu, Johnson, Lai, Cricelli, & Fleming, ). However, most of this research abstracts from the fact that platforms differ significantly from each other.…”

Equity crowdfunding, shareholder structures, and firm performance

“…Moreover, recently, there is also evidence of wisdom of crowd effects in reward‐based crowdfunding (Mollick & Nanda, ) and lending‐based crowdfunding (Iyer, Khwaja, Luttmer, & Shue, ) contexts. Mollick and Nanda (), for instance, find large agreement between the funding decisions of crowds and experts related to theatre projects.…”

Will FinTech make regional banks superfluous for small firm finance? Observations from soft information‐based lending in Germany

“…In the context of this shift to the active screening of peer‐to‐peer lending platforms, researchers are debating the extent to which alternative and soft information is considered in lending decisions, that is scoring algorithms (Iyer, Khwaja, Luttmer, & Shue, ). Jagtiani and Lemieux () document that LendingClub's scoring system uses alternative information and machine learning that have caused a gradual increase of prediction validity in recent years.…”

“…Studies show that the combination of hard and soft respectively alternative information enhances default prediction (compared to credit bureau scores such as the FICO), especially when lending to more risky borrowers (Iyer et al, ; Berg et al, ; Jagtiani & Lemieux, ). Kreditech Holding SSL GmbH, a FinTech start‐up specialised in subprime private loans, demonstrates the possibilities of the digital footprints for Big‐Data‐Scoring.…”