Invisible Primes: Fintech Lending with Alternative Data
62 Pages Posted: 11 Oct 2021
Date Written: October 6, 2021
We exploit anonymized administrative data provided by a major fintech platform to investigate whether using alternative data to assess borrowers' creditworthiness results in broader credit access. Comparing actual outcomes of the fintech platform’s model to counterfactual outcomes based on a “traditional model” used for regulatory reporting purposes, we find that the latter would result in up to 60% higher probability of being rejected and higher interest rates for those approved. The borrowers most positively affected are the “invisible primes”--borrowers with low credit scores and short credit histories, but also a low propensity to default. We show that funding loans to these borrowers leads to better economic outcomes for the borrowers and higher returns for the fintech platform.
Keywords: Fintech Lending, Alternative Data, Machine Learning, Algorithm Bias
JEL Classification: D14, H52, H81, J24, I23
Suggested Citation: Suggested Citation