Fairness Testing for Algorithmic Pricing

TL;DR

This paper reveals flaws in current fairness testing methods for algorithmic pricing, derives correct variance estimators, and demonstrates that all tested insurers fail fairness criteria.

Contribution

It introduces valid statistical tools for fairness testing in deterministic pricing algorithms, correcting previous invalid standard error assumptions.

Findings

All 34 insurers fail the conditional demographic parity test.

Standard proxy discrimination formulas fail to detect discrimination in this context.

Corrected formulas identify all insurers as statistically significant in discrimination tests.

Abstract

Algorithmic systems now set prices across auto insurance, credit, and lending markets, and regulators increasingly require firms to demonstrate that these systems do not discriminate against protected groups. The standard audit regresses pricing output on a protected attribute and legitimate rating factors, then tests the resulting coefficient using ordinary least squares standard errors. We show that this approach is structurally invalid. Pricing algorithms are usually deterministic, so residuals reflect approximation error rather than sampling variability, rendering classical standard errors invalid in both direction and magnitude. We derive correct asymptotic variance estimators for OLS and GLM audit regressions and the correct cross-covariance formula for proxy discrimination testing. Applied to quoted premiums from 34 Illinois auto insurers, every insurer fails the conditional demographic parity test, with minority zip codes paying $34-$158 more per year than comparable-risk white zip codes. The standard proxy discrimination formula flags zero insurers. However, our corrected formula identifies all 34 as statistically significant, of which 16 exceed the substantive threshold. Our framework provides statistically valid audit tools for any deterministic algorithmic system subject to regression-based fairness testing.