Using people's text messages to decide their credit scores is just making the gap between the rich and the poor even bigger.
March 26, 2026
Original Paper
Signals from SMS: Alternative Data and Distributional Effects in Credit Scoring
SSRN · 6345399
The Takeaway
While Fintech apps claim that using alternative data (like SMS transaction records) helps the 'unbanked,' this study of 23 million applications found the benefits are heavily skewed toward the wealthy. Financially disadvantaged groups are less willing to share their data and have 'quieter' digital footprints, meaning the algorithm finds fewer reasons to approve them compared to wealthier peers.
From the abstract
Using data from 23 million loan applications from a FinTech lender in India, we demonstrate how transaction records and credit histories can be extracted from text messages and used to assess applicants' creditworthiness. We show that an alternative credit scoring system built on these data and machine-learning algorithms predicts consumer delinquency more accurately than traditional credit scores, with gains attributable to both algorithmic improvements and the use of alternative data. We find