February 14, 2025
Linda Blaser
Study uncovers unexpected relationship between banks and peer-to-peer lending
Lake Forest College professors and students challenge conventional wisdom with groundbreaking research.
For years, the assumption was simple: in areas where traditional banks disappear, peer-to-peer (P2P) lending platforms like Prosper would step in to fill the void. But a recent research study led by Professor of Economics Rob Lemke and Associate Professor of Finance Nancy Tao with student researchers Belen Cuadros ’25 and Zofia Czarnik ’25, has turned that narrative on its head.
Their co-authored study, published in Managerial Finance, reveals that P2P lending doesn’t replace banks—it complements them. In fact, both systems thrive best when they coexist.
The research took place last summer, supported by funding from the Rosemary Hale Fund for Teaching and Research. That financial support made it possible for the team to use four critical datasets:
- Census data
- Bank data
- Credit union data
- Peer-to-peer loan data from Prosper
Cuadros, an economics and data science double major, and Czarnik, an economics major who is double-minoring in politics and legal studies, worked closely with their professors, applying the knowledge they gained from the Richter Scholar Program and the James Rocco Data Research Scholarship to analyze the data and interpret the findings. What they discovered came as a surprise.
The core questions
“Does the presence of more peer-to-peer lending lead to a decrease in the presence of banks or vice-versa? Does this affect areas with people who have higher or lower incomes?” —Belen Cuadros ’25
The research team set out to determine whether P2P lending serves as a substitute for traditional brick-and-mortar banks, especially in underserved “banking deserts,” or if the two systems are interdependent. Their extensive analysis spanned data from over 3,000 US cities, covering banks, credit unions, and over a million loans issued through Prosper.
“Banking deserts” typically occur in rural, minority, and low-income areas. In wealthier communities, individuals often can choose from several banking and loan options.
To apply for a peer-to-peer loan, “individuals go on the app, submit information about their credit scores, and provide some demographic information—basically creating a profile for themselves for a lender to view,” Czarnik said. “And then other individuals go on the app and, if they want to, can choose to lend their money with interest to borrowers, so they’re making a profit on those ‘investments’.”
Key findings
- Complementary, not competitive: Contrary to previous studies, peer-to-peer lending does not replace traditional banks. Instead, both systems coexist and thrive in the same areas. P2P borrowers, in fact, need to have a bank account to get a loan.
- Middle-class focus: While it was anticipated that data would show P2P lending aids lower-income communities, the study found that peer-to-peer lending primarily serves middle-income borrowers. Wealthier individuals rely on traditional banks and other options, while the poorest often remain underserved and, oftentimes, only have predatory payday loans and pawn shops as options.
- Bank closures hurt communities: The decline of brick-and-mortar banks still negatively impacts individuals’ access to financial services. P2P platforms haven’t effectively filled this gap.
The power of data (and AI)
“We had about a million data points. Cleaning that data manually would’ve taken weeks. Using AI and Python saved us countless hours.” — Zofia Czarnik ’25
Handling such massive datasets presented challenges, especially with inconsistent entries caused by individuals incorrectly entering one of the most critical datapoints, the name of their city or town. To overcome this obstacle, the Lake Forest team wrote Python script and used ChatGPT to clean the data efficiently—an innovative approach that impressed peer reviewers.
“If it wasn’t for that, we would have spent 10 hours a day for two weeks just fixing the datasets,” Cuadros noted.
The results? “I think our results are more solid, authentic, and convincing because we had massive datasets,” Tao said. “We had data on about 200,000 Prosper loans per year. So over five years, that was about a million pieces of data. We were using a better dataset—a superior sample—that gave us more accurate results.”
Real-world impact
“Our major contribution is our conclusion that both banks and peer-to-peer lending need to be present in order to provide access to affordable financial services for borrowers.” —Associate Professor Nancy Tao
“Students at Lake Forest College have the opportunity to engage with faculty researchers and publish in academic journals. These experiences benefit students but they also invigorate faculty encourage us to remain active in research.” —Professor Rob Lemke
This project not only contributed to economic research but also provided invaluable experience for the student researchers. Their hands-on work with complex datasets and real-world financial issues has prepared them for future academic and professional success. As Belen reflects, “Research isn’t just about data. It’s about collaboration, persistence, and the willingness to question what we think we know.”
Cuadros already has a job lined up for after graduation, thanks, in large part, to the research opportunities she’s had at the College.
Czarnik has her sights set on law school and also credits the research work she’s done at Lake Forest with preparing her for that next step.
“For law school, one of the most sought-after majors is economics because it requires qualitative and quantitative skills. To have a published, co-authored paper that showcases those skills is incredible,” she said. “I’m incredibly grateful.”