Proxy Discrimination
Discrimination that happens when a neutral variable, such as zip code or credit data, correlates with a protected class and produces unfair outcomes.
Proxy discrimination occurs when a model uses a variable that appears neutral but is correlated with a protected class, producing unfair outcomes for that group. Common examples include zip code, credit score, occupation, and certain geographic or lifestyle variables.
New York’s Circular Letter No. 7 says insurers should evaluate the extent to which external consumer data is “correlated with (i.e., proxy for)” protected-class status where unfair or unlawful discrimination may result. An insurer should then consider, for any correlation it identifies, whether a legitimate business necessity requires using that data.
The American Academy of Actuaries, quoted in a footnote to the letter, noted in February 2024 that “algorithm[s] may learn to identify and rely upon seemingly facially neutral variables that have a close correlation to protected characteristics or traits.” See our guide to the proxy test and glossary entries on protected class and disparate impact.