Yesterday, Secretary Donovan announced the findings from HUD's latest paired-testing study of discrimination against minority homeseekers. An Urban Institute team conducted the study, the third national paired-testing study we've completed for HUD (see video).
What did we find?
When well-qualified minority homeseekers contact housing providers to inquire about recently advertised housing units, they generally are just as likely as equally qualified whites to get an appointment and learn about at least one available housing unit. However, when differences in treatment occur, white homeseekers are more likely to be favored than are minorities. Most important, minority homeseekers are told about and shown fewer homes and apartments than whites.
Here are the numbers:
Although the most blatant forms of housing discrimination (refusing to meet with a minority homeseeker or provide information about any available units) have declined since HUD's first national paired-testing study in 1977, the forms that persist (providing information about fewer units) raise the costs of housing search for minorities and restrict their housing options.
How did we produce these results?
In a paired test, two people, one white and the other minority, pose as equally qualified homeseekers and inquire about available homes or apartments. Paired testing originated to support the enforcement of federal fair housing protections, essentially catching discrimination in the act. Researchers have adapted the tool to systematically measure how often discrimination occurs across housing markets and what forms it takes.
Despite its power, paired testing can't capture all forms of discrimination that might occur during a housing search. For example, it doesn't encompass differences in advertising practices that may limit a homeseeker’s knowledge about available housing options. And it can't measure differences in treatment that might occur after an initial inquiry––when homeseekers submit applications, seek mortgage financing, or negotiate lease terms. Moreover, the latest results don't reflect the experience of the average or typical minority homeseeker, because testers presented themselves as unambiguously well-qualified for the advertised homes and apartments about which they inquired.
For all these reasons, the latest study probably understates the total level of discrimination that occurs in the marketplace today. Nonetheless, the long-term trends in patterns of discrimination suggest that the attitudes and actions of rental and sales agents have changed over time, and that fair housing enforcement and public education are working.
What do our latest results mean for policy?
Fair housing enforcement and education are still needed as long as significant discrimination persists. And today's discrimination is very difficult for victims to detect, so enforcement strategies can't rely primarily on individual complaints of suspected discrimination. HUD should encourage the local fair housing organizations it funds to conduct more proactive testing, especially in the sales market, where discrimination appears higher than in the rental market.
Proactive testing can reveal discriminatory practices that would otherwise go unpunished. And when housing providers know that testing is ongoing, they are more likely to comply with the law. In addition, more locally targeted testing may be needed to pinpoint the types of neighborhoods, housing providers, or homeseekers where discrimination is most prevalent. In particular, minority homeseekers with lower incomes, less wealth, weaker English language fluency, or blemished credit may face higher levels of discrimination than documented in this national study.
Finally, local fair housing organizations should expand and strengthen their relationships with Hispanic and Asian communities to tackle the discrimination experienced by all people of color. Historically, the fair housing movement has focused on discrimination against blacks. Although some local organizations have extended their scope in light of changing demographic realities, others have yet to do so.
What do these results tell us about the persistence of residential segregation and neighborhood inequality? I’ll have thoughts on that question tomorrow.