HUD Assistant Secretary for Fair Housing and Equal Opportunity (FHEO) Anna Maria Farías filed a Charge of Discrimination against Facebook with HUD’s Office of Administrative Law Judges on March 28. HUD alleges that Facebook discriminates by enabling advertisers to restrict which Facebook users receive housing-related ads based on race, color, national origin, sex, familial status, disability, and religion – the protected characteristics under the Fair Housing Act. HUD asserts Facebook mines user data and classifies users based on their protected characteristics. Facebook’s ad-targeting tools invite advertisers to express unlawful preferences by suggesting discriminatory options, and Facebook provides housing-related ads to certain users and not to others based on users’ actual or imputed protected characteristics.
Some of the ways HUD alleges Facebook’s ad targeting tools enable advertisers of housing and housing-related services to discriminate include:
- Drawing a red line around majority-minority ZIP codes and not showing housing or housing-related ads to users who live in those ZIP codes (discrimination on the basis of race and color).
- Not showing ads to users Facebook characterizes as interested in “assistance dog,” “mobility scooter,” or “deaf culture” (discrimination on the basis of disability).
- Not showing ads to users Facebook characterizes as interested in “child care” or “parenting,” or by showing ads only to users with children above a specific age (discrimination on the basis of familial status).
- Not showing ads to users Facebook characterizes as interested in “Latin America,” “Southeast Asia,” “China,” “Honduras,” “Somalia,” or the “Hispanic National Bar Association” (discrimination on the basis of national origin).
- Showing ads only to users Facebook characterizes as interested in “Bible,” “Jesus,” or “Christian Church” (discrimination on the basis of religion).
- Showing ads only to men or only to women (discrimination on the basis of sex).
The official Charge of Discrimination details the many steps in the process Facebook uses that lead to housing discrimination. The last step in the process, determining the actual audience for an ad, entails Facebook combining data it has about user attributes and behavior on its platforms (including Instagram among others) with data it obtains about user behavior on other websites and in the non-digital world. According to HUD, Facebook then uses machine learning and other prediction techniques to classify groups of users in order to protect each user’s likely response to a given ad. By grouping users who “like” similar pages unrelated to housing and presuming a shared interest or disinterest in housing-related advertisements, HUD asserts Facebook’s mechanisms function in a manner similar to an advertiser that intentionally targets or excludes users based on their protected characteristics.
This secretary-initiated complaint of housing discrimination is the first to be issued by HUD during Secretary Ben Carson’s administration.
HUD’s Charge of Discrimination is at: https://bit.ly/2OxXTCC
A HUD media release is at: https://bit.ly/2HL3uVq