Three Essays on the Economic Impact of FinTech Lending

This dissertation consists of three independent papers with a related theme focusing on the economic and capital market impacts of FinTech lending. In specific, this dissertation explores the influence of FinTech penetration on traditional banks’ risk-taking, the influence of non-traditional information on lending discrimination in P2P lending, and whether digital inclusion is a potential determinant of P2P penetration.

In Chapter 1, based on a sample of U.S. community banks and FinTech loans data from LendingClub and Prosper, I find a positive association between banks' future change in risk-taking and their current exposure to FinTech penetration. Path analysis reveals that FinTech penetration influences bank risk-taking through the erosion of bank charter value. Cross-sectional analysis suggests that banks with the lowest ex-ante charter values increase risk-taking the most. Additionally, I document that the risk-increasing effect of FinTech penetration is less pronounced for banks with more conservative loan loss provisioning and less reliance on hard information. My findings suggest that regulators may need to pay more attention to smaller banks with lower or deteriorating charter values.

Chapter 2 hypothesizes that racial discrimination can exist in P2P lending even when race-related information is not directly observable, and that the degree of racial discrimination decreases in the precision of credit quality signals generated from both traditional and non-traditional information. I document that loan listings in counties with a greater proportion of minority populations are associated with higher lending rates and higher loan denial rates. Cross-sectional tests indicate that racial discrimination is less pronounced when the availability of both traditional and non-traditional information is greater.

Chapter 3 investigates the influence of digital inclusion on financial inclusion. I document that digital inclusion is positively associated with P2P lending penetration, with such relation more pronounced in county-years with more vulnerable/excluded populations. The results are robust to the use of instrumental variable (2SLS) approach, alternative measurements, weighted least squares regression, additional controls, and single-year analysis. In addition, I document that higher-risk borrowing is less likely to be denied in county-years with higher digital inclusion.