Invisible Primes: Fintech Lending with Alternative Data
Abstract
We study the impact of alternative data on credit access and borrower outcomes using anonymized data from a major U.S. financial technology (fintech) platform that incorporates education, employment, and other nontraditional variables into its underwriting algorithm. The platform approves 15%–30% of low-credit-score applicants rejected by traditional models and offers them lower rates. It particularly benefits “invisible primes”—borrowers with thin credit files and low credit scores but low default risk. We show that gains for invisible primes are primarily driven by alternative data, while model sophistication yields additional improvements in segments where traditional credit report information is more extensive. Using exogenous variation, we find that expanded credit access improves borrowers’ subsequent financial outcomes.
This paper has been accepted by Kay Giesecke for the Virtual Special Issue on Digital Finance.
Supplemental Material: The internet appendix and data files are available at https://doi.org/10.1287/mnsc.2024.07854.

