The Luxury of Rounding: Heuristics and Private Information
Publication Date
1-16-2026
Abstract
When the marginal value of precision is heterogeneous, deviations from decision heuristics can reveal private information. On a major FinTech lending platform, non-round loans default 1.76 percentage points more than round loans, with worse FICO, larger contractual losses, and lower prepayment, conditional on the platform’s full information set and stable under machine-learning benchmarks. A model, field heterogeneity around a safety-net contraction, and a preregistered randomized experiment support the mechanism: tighter constraints raise the marginal value of precision. This information is not priced in platform-set rates. Institutional screening is most effective within this segment; retail investors bear essentially all the loss.
Document Type
Article
Keywords
Behavioral Heuristics, Private Information, Consumer Credit, Financial Constraints
Disciplines
Finance
Source
SMU Cox: Finance (Topic)
Language
English
