Federated learning improves income estimates for digital lending
FedIncome: Federated Learning for Income Estimation in Digital Lending Under Data Sovereignty Constraints
Machine Learning
Summary
Verifying how much money people earn is hard in online loan applications, causing banks to make mistakes in lending. The authors created FedIncome, a way for many lenders to work together on a shared income prediction model without sharing private data. Their method works almost as well as if all data was combined and helps smaller lenders get better estimates. Using FedIncome can increase loan approvals without significantly raising the risk of defaults.
What this means in practice
- •For financial institutions: Improve income predictions for loan applicants without sharing raw data between institutions, enhancing credit decisions under privacy constraints.
- •For credit risk teams: Use federated income estimates combined with specific debt-to-income rules to increase loan approval rates while maintaining default risk.
Authors
Sultan Amed, Tanmay Sen, Sayantan Banerjee
Abstract
Verified income is often unavailable in digital loan applications, forcing lenders to rely on reported income and potentially leading to over-lending, overly conservative offers, or rejection of creditworthy applicants. Cross-institutional data-sharing constraints make this problem especially difficult for smaller lenders with limited training data. We introduce FedIncome, a federated learning framework for income estimation that enables institutions to train a shared model without pooling raw borrower records. Using more than one million LendingClub loans partitioned into $50$ state-level clients, we simulate a heterogeneous lending consortium. The best federated model achieves out-of-time $R^2=0.608$, compared with $0.619$ for a pooled centralised benchmark. Small-sample clients obtain an average out-of-time $R^2$ improvement of $3.8$ percentage points relative to the pooled centralised benchmark, while the fitted client-level relationship places the empirical crossover at approximately $4,790$ training observations in this setting. When pooling is infeasible and the relevant alternative is local-only training, federation improves out-of-time performance across all sample-size groups, with the largest gains for data-scarce clients. We also combine federated income estimates with state- and income-specific debt-to-income thresholds. In a retrospective decision analysis, replacing reported income with the federated estimate increases simulated approval rates with only modest changes in observed default rates. FedIncome supports collaborative learning under data-locality constraints with little aggregate loss relative to pooled training and larger gains relative to local-only estimation.