Estimating Racial Disparities When Race is Not Observed

TL;DR

This paper introduces BIRDiE, a new scalable method that uses surnames as instrumental variables to estimate racial disparities without direct race data, improving accuracy over existing methods like BISG.

Contribution

The paper develops BIRDiE, a novel Bayesian instrumental regression approach that corrects bias in racial disparity estimates when race is unobserved, using surnames as instruments.

Findings

BIRDiE reduces estimation error by up to 84% in validation studies.

The method is scalable to large administrative datasets.

Application to IRS data reveals racial differences in mortgage benefit distribution.

Abstract

The estimation of racial disparities in various fields is often hampered by the lack of individual-level racial information. In many cases, the law prohibits the collection of such information to prevent direct racial discrimination. As a result, analysts have frequently adopted Bayesian Improved Surname Geocoding (BISG) and its variants, which combine individual names and addresses with Census data to predict race. Unfortunately, the residuals of BISG are often correlated with the outcomes of interest, generally attenuating estimates of racial disparities. To correct this bias, we propose an alternative identification strategy under the assumption that surname is conditionally independent of the outcome given (unobserved) race, residence location, and other observed characteristics. We introduce a new class of models, Bayesian Instrumental Regression for Disparity Estimation (BIRDiE), that take BISG probabilities as inputs and produce racial disparity estimates by using surnames as an instrumental variable for race. Our estimation method is scalable, making it possible to analyze large-scale administrative data. We also show how to address potential violations of the key identification assumptions. A validation study based on the North Carolina voter file shows that BISG+BIRDiE reduces error by up to 84% when estimating racial differences in party registration. Finally, we apply the proposed methodology to estimate racial differences in who benefits from the home mortgage interest deduction using individual-level tax data from the U.S. Internal Revenue Service. Open-source software is available which implements the proposed methodology.