On Monday, the US Department of Housing and Urban Development (HUD) will officially propose a major change to a well-settled standard in how the agency and the courts review claims of discrimination under the Fair Housing Act of 1968.
The standard, referred to as “disparate impact,” recognizes that groups protected under the Fair Housing Act can experience discrimination through policies and practices even if there is no evidence of an explicit intention to discriminate. And in today’s environment, implicit or unintentional acts of discrimination may be more common and harder to detect.
The proposal threatens to halt 50 years of progress toward fair and equal access to housing and credit, with sweeping implications for people of color, women, families with children, and people with disabilities.
How housing discrimination can be implicit or unintentional
Today, explicit discrimination from lenders, landlords, realtors, or local governments (though it still occurs) is likely less common thanks to federal, state, and local laws that prohibit it and cultural norms that at least discourage blatant racism.
America’s housing ecosystem is home to widening disparities in homeownership, persistent racial segregation, and growing areas of concentrated poverty. Nearly two decades of Urban Institute paired-testing research documents the complex ways access to housing remains unequal.
HUD and the courts have consistently recognized (PDF) that housing policies and practices can adversely affect groups such as racial, ethnic, and religious minorities; women; families with children; and people with disabilities and violate the Fair Housing Act—even if discrimination is implicit. Here are some examples:
- A federal disaster recovery program could allocate funds to rebuild homes in a way that devalues homes in black neighborhoods, as we saw in New Orleans after Hurricane Katrina.
- A lender might provide incentives for higher interest rates in a way that encourages loan officers to market high-cost and high-risk mortgages to people of color, as we saw in the lead-up to the foreclosure crisis.
- A provision in a local zoning ordinance could limit the location of group homes, unfairly denying people with disabilities access to housing in residential areas.
Though these policies can seem neutral on their face, they can unfairly exclude groups of people and perpetuate unfair outcomes.
That’s partly why every federal appellate court to consider this issue—including the US Supreme Court—has upheld and applied the disparate impact standard under the Fair Housing Act. These decisions recognize that discrimination can operate in many ways and that an action’s consequences, not just motives, matter.
How the proposal would make it easier to discriminate and harder to reduce disparities
Over several decades, courts have worked out a straightforward process for evaluating claims of disparate impact, which HUD standardized and clarified in a 2013 rule. Under the existing rule, courts engage in a three-step process of “burden shifting” that recognizes legitimate business interests while guarding against discriminatory outcomes.
The new proposed HUD rule would eliminate this burden-shifting framework and replace it with an onerous five-part test to prove discrimination. And plaintiffs would bear the entire burden of proving each element of the test.
Among these elements, plaintiffs would have to establish that the challenged policy is “arbitrary, artificial, and unnecessary” and allege a “robust causal link” between the policy and the adverse impact on a protected class. The proposed rule would also require plaintiffs to demonstrate by a “preponderance of evidence” that a less discriminatory policy would work just as well to fulfill the defendant’s business purpose, without imposing greater costs or burdens on the defendant.
Technology brings new potential for discrimination
Perhaps most troubling, HUD proposes to make it easier for businesses and policymakers to escape liability when they rely on algorithms to make housing and lending decisions.
Algorithms are increasingly used in nearly every facet of housing and lending, including tenant screening, home appraisals, and code enforcement. Even in areas that have long used computer models to assess risk, such as credit underwriting, algorithms are incorporating a broader array of seemingly neutral inputs, such as data from social media networks.
The proposed rule would allow a defendant to defeat a disparate impact claim when the inputs used in an algorithm do not include attributes of protected classes (e.g., race or gender) or proxies for them. In addition, defendants could escape liability whenever an algorithm was created by a “recognized third party.” These loopholes, if adopted, could undermine the rule entirely.
As Cathy O’Neil and Virginia Eubanks show in their recent books, Weapons of Math Destruction and Automating Inequality, algorithms and inputs reinforce disparities. It’s not necessarily the case that algorithms are any more fair, or less discriminatory, than humans when information on race is excluded, at least according to a recent study from researchers at Dartmouth. At the Urban Institute, we’ve documented how data that are representative can still be biased and how, without close scrutiny, algorithms can amplify those biases.
HUD’s proposed changes threaten to raise the bar for documenting disparate impact so high that it likely eliminates civil rights lawyers’ ability to use it as a tool to identify and address discriminatory practices and outcomes.
Fighting discrimination can make markets work for everyone
As our country becomes more diverse and the mechanisms that perpetuate inequality become more complex, the Fair Housing Act ensures no one is shut out of the economic opportunities central to our collective success and prosperity.
Both discriminatory practices and the law will continue to evolve, and disparate impact standards are more important than ever to protect against unfair policies and practices.
When Lyndon B. Johnson signed the Fair Housing Act, less than a week after the assassination of Dr. Martin Luther King Jr., he famously declared, “We have come some of the way, not near all of it. There is much yet to do.” LBJ’s words ring as true today as they did half a century ago.
Tune in and subscribe today.
The Urban Institute podcast, Evidence in Action, inspires changemakers to lead with evidence and act with equity. Co-hosted by Urban President Sarah Rosen Wartell and Executive Vice President Kimberlyn Leary, every episode features in-depth discussions with experts and leaders on topics ranging from how to advance equity, to designing innovative solutions that achieve community impact, to what it means to practice evidence-based leadership.