The perfect bail-in: Financing without banks using peer-to-peer lending

Polyzos, Efstathios; Samitas, Aristeidis; Rubbaniy, Ghulame

Abstract

We explore the potential outcomes for financial stability when using peer-to-peer lenders to finance economic activity. Combining Random Regression Forests, a machine-learning process, with an agent-based model, we perform simulations on artificial economies with various degrees of adoption of peer-to-peer lending. We find that as peer-to-peer lenders proliferate, there is increased financial instability, lower GDP and higher unemployment. On the other hand, peer-to-peer lending increases the total volume of loans given out but demonstrates a preference towards consumer loans (over corporate loans), which has a negative effect in the long run. Finally, introducing peer-to-peer lenders increases the access of the unbanked to services which conventional banking is not able to offer within the extant regulatory framework. Our results can help policymakers as they address the issue of regulation in the peer-to-peer finance industry.

• Journal Article Type: Article
• Acceptance Date: May 7, 2023
• Online Publication Date: May 29, 2023
• Publication Date: Jul 1, 2024
• Deposit Date: Jun 2, 2023
• Publicly Available Date: May 30, 2025
• Journal: International Journal of Finance and Economics
• Print ISSN: 1076-9307
• Electronic ISSN: 1099-1158
• Publisher: Wiley
• Peer Reviewed: Peer Reviewed
• Volume: 29
• Issue: 3
• DOI: https://doi.org/10.1002/ijfe.2838
• Keywords: Economics; Econometrics; Finance; Accounting; agent-based finance; financial stability; machine learning; p2p lending; peer-to-peer; random regression forests
• Public URL: https://uwe-repository.worktribe.com/output/10833324
• Publisher URL: https://onlinelibrary.wiley.com/doi/10.1002/ijfe.2838
• Additional Information: Received: 2022-05-15; Accepted: 2023-05-07; Published: 2023-05-29