Leading experts weigh in on current policy issues and challenges

Texas Department of Housing and Community Affairs v. The Inclusive Communities Project: A Test of Disparate Impact

This month, the Supreme Court will decide whether the Fair Housing Act treats policies and practices as discriminatory if they have a “disparate impact” based on race, national origin, color, religion, sex, familial status, or disability. Under the disparate impact theory, individuals and groups can challenge a housing or lending policy or practice if it has a disproportionate adverse effect on protected classes, even if unintentional. How could this decision affect how we address and measure discrimination in housing and lending and ongoing segregation in the United States?

The Urban Institute is talking with...
David Stevens David Stevens
Michael Bodaken Michael Bodaken
Solomon Greene Solomon Greene
Rolf Pendall Rolf Pendall
Margery Turner Margery Turner
Laurie Goodman Laurie Goodman
Erika C. Poethig
Moderated by:
Erika C. Poethig
Institute Fellow, Director of Urban Policy Initiatives

Several weeks ago, the New York Times posted an editorial in the wake of the Baltimore riots that charted the fits and starts both Republic and Democratic administrations have taken to enforce the Fair Housing Act of 1968.   As the editorial explained, efforts to enforce aspects of the act, which sought to undo decades of public and private actions that contributed to racial segregation, take political courage that few administrations have mustered.  Now, we wait to see what the Supreme Court will decide about an important tool of under the act that has been essential to combating discriminatory private actions and public policy.  

 This week, we will be discussing the implications of the Supreme Court’s decision.  To begin, I would like colleagues contributing to this discussion to share their perspectives about disparate impact theory and its role in undoing discriminatory practices that lead to segregation.  


Under the disparate impact theory, individuals and groups can challenge a housing or lending policy or practice as discriminatory if it has a disproportionate adverse effect on a given racial or ethnic group, even if there is no intentional discrimination.

I strongly believe intentional discrimination is terrible, businesses should be required to treat all similarly situated applicants equally, and to the extent they don’t they should be vigorously prosecuted to the full extent of the law. But the issue is always, “who is similarly situated?”.  You can only hold a business responsible for what they can control. For example, if a lender applies uniform underwriting standards to all applicants, it will likely result in more mortgage denials for black and Hispanic applicants than for white applicants, because there is a difference in income, wealth and credit experience between the groups. Businesses certainly have a responsibility to support, and at the minimum, not to stand in the way of government programs that push for greater equality of opportunity. But that is different than the disparate impact doctrine, which could hold the private sector guilty of discrimination if their policies resulted in a differential impact on different racial and ethnic groups, despite the fact that these groups have differences in income, wealth and credit experience.

Yes, a lender or other private sector participant can defend themselves against a disparate impact charge by arguing that there is a legitimate business justification for their actions. However, these lawsuits will be expensive to defend, and there is reputational risk for the business. Perhaps more importantly, the threat of such lawsuits makes lenders less willing to exercise any type of judgment. It is easier to defend automated application of pre-set rules, blindly applied—a result that often works to the detriment of exactly the class of people disparate impact suits are meant to protect.   

The real issue is that, as a society, we have not chosen to adopt public policies in which opportunities are really equal for all, which would eventually result in closing the income gap. In fact, the Gini coefficient, which measures the degree of income inequality has been growing. Asking the private sector to compensate for lack of political will is more likely to result in defensive actions than progress.   


I see the “disparate impact” principle as an essential tool for reversing the terrible legacy of past discrimination by both public and private institutions. 

Decades of public policies and institutional practices – like restrictive covenants, redlining, block-busting, exclusionary zoning, racial steering, and predatory lending – have left minorities with lower wealth and weaker credit histories than whites.  Housing market practices that purport to be “colorblind” essentially lock in those disparities , restricting minority homeseekers’ access to homes in opportunity-rich neighborhoods that offer high-performing public schools, safe and healthy environments, and wealth-building through home price appreciation.  

And it’s very difficult to prove intent.  Urban’s most recent paired-testing study of discrimination against minority homebuyers and renters finds that blatant “door slamming” discrimination, where minorities are denied available homes and apartments outright, has become rare.  But minority homeseekers are still told about and shown fewer available housing units, raising their costs of search and constraining their choices.

The disparate impact test doesn’t require private-sector businesses to lend, sell, or rent to somebody who really isn’t qualified.   But it does require that they look for alternatives  to rules and procedures that – whether intentionally or not -- systematically disadvantage minority customers, thereby perpetuating past injustices.

Given our history, I think this is a reasonable and necessary public mandate.

Agree w/ Laurie that we need bold public policy to redress discrimination. However, we know that businesses/private sector are not above reproach, even when it appears that their business interests would argue for nondiscrimination. See financial crisis and who was hurt disproportionately by predatory lending.  See also : http://www.nytimes.com/roomfordebate/2015/04/27/can-discrimination-exist-without-clear-intent/statistical-methods-can-demonstrate-racial-disparity.  In the words of Jan Ayers, the author of the article: "Credible statistical methods exist to identify policies that disproportionately burden minorities and are unrelated to a seller’s costs of doing business. Multivariate regressions that often control for dozens of potentially business-justified variables allow courts to test statistically whether corporate policies tend to hurt minorities more than similarly situated whites."  Disparate impact allows for the  modest  redressing of private...as well as public...discrimination..



We all agree that all borrowers should be protected from intentional acts of discrimination.  Housing discrimination has unfortunately played an all-to-large role in this country’s history, which is why Congress passed the Fair Housing Act as part of the Civil Rights Act in 1968.  The question at play here is whether lenders should be prosecuted when neutral policies result in outcomes that impact some groups more than others. 

In the wake of the financial crisis, lawmakers and regulators created well-intentioned rules to hold lenders more accountable to ensure that borrowers can afford to repay their mortgages.  Some of these rules have specific underwriting guidelines that determine who will and who will not get a mortgage. 

Further, the government’s own rules for providing mortgages for federally guaranteed programs (FHA, Veterans Administration and others) can ultimately affect groups differently.    

The concern with applying disparate impact theory to mortgage lending is that it allows the mere existence of a statistical variance to make a discrimination claim, exposing every lending decision to a legal challenge, even if that lending decision is based on sound underwriting and/or compliance with federal regulations.

Consider the implications of the Supreme Court weakening or overturning disparate impact as a tool. What impact will it have on lending, segregation, and state and local policies? How might it affect government programs meant to target affordable homeownership and affordable rental housing? How might it affect new construction and rehabilitation/preservation of existing housing?  

It’s quite difficult to predict how the Court might rule. However, it appears to us that that there are a number of ways to look at possible outcomes.


We have used the following template provided by the Opportunity Agenda-modified somewhat to distinguish the preservation of affordable housing in lower income, racially concentrated communities from new construction in those same communities-to express how NHT might react to a variety of possibilities. This is just a frame for this debate, not our official position. Over time, after the Court issues its decision, we will prepare a more robust response with our Board and fellow housing and community development colleagues.


  • If The Court affirms, upholding disparate impact analysis as part of the Fair Housing Act, we would be pleased with the result but would be quick to point out that such a decision does not in any way indicate that affordable housing preservation in a low income, principally minority neighborhood is in any way affected by the ICP decision, i.e., preservation and rehabilitation of an existing affordable housing development does not constitute disparate impact.  Notably, the ICP case was brought by plaintiffs to demonstrate how the building of new apartments in an already racially concentrated area would constitute disparate impact and it was this affirmative decision to allegedly perpetuate such concentration that would arguably constitute disparate impact. Where, on the other hand, minority residents choose to remain where they already reside and improve their housing and neighborhood, there is no violation. To hold otherwise would mean to restrict federal resources for minority residents already residing in areas of poverty, certainly not a fair reading of disparate impact.
  • If the decision “reverses, holding on statutory grounds that Congress did not intend to prohibit unjustified disparate impact,” we would work with our civil rights colleagues to assure that Congress rectifies this wrongful interpretation of the Fair Housing Act, distinguishing this particular situation from the broader issue of disparate impact. As I have noted above, disparate impact is an important doctrine that can and should be used to redress private and public discrimination.  In particular, we would strongly argue that the Court decision does not in any way affect ‘balanced approach’ to addressing housing need, one that focuses on both “opportunity areas” and revitalization of existing neighborhoods.
  • Finally, should the Court reject disparate impact based on constitutional concerns, we would be quite concerned and would urge Federal, state, and local governments to employ the full array of tools to smoke out covert discrimination and to counter conscious and subconscious bias in housing decision-making. And the courts must heighten their vigilance to detect and address all forms of bias in housing.  Again, we do not believe this should affect the “balanced housing policy” approach that would place federal resources in higher opportunity neighborhoods as well as revitalization of existing neighborhoods, wherever located. 

As colleagues in this debate have noted, the disparate impact standard under the Fair Housing Act applies to both housing provision (the production, sale and rental of housing) and housing credit (i.e., mortgage lending).  I am going to pick up on the latter thread, since I agree with Marge’s comments on the critical importance of the disparate impact standard in ensuring that everyone has equal access to housing, regardless of their race, ethnicity, national origin, family status or ability, but I strongly disagree with the contention that the disparate impact standard discourages neutral lending practices or impedes credit flows to underserved borrowers.

The disparate standard does NOT prohibit lenders from adopting uniform underwriting standards or other neutral policies and practices that may have a disproportionate effect on protected classes such as racial minorities or households with children. Rather, it requires lenders to offer evidence that a practice that has a demonstrated and statistically significant discriminatory effect serves an important business objective. A question arises as to why any lender would want to continue a practice that did have a discriminatory effect but served no business objective. The standard is aimed solely at weeding out lending policies that serve no purpose other than to perpetuate inequality and exclusion from housing and credit markets.  

David and Laurie both raise concerns that the disparate impact rule will encourage costly lawsuits against lenders and Laurie suggests that, in response, lenders will avoid innovating in their products or incorporating judgment into underwriting that may actually benefit underserved borrowers.  In response, I’d point out that lenders are exposed to lawsuits for any number of violations of federal laws, and laws promoting civil rights and fairness shouldn’t play second fiddle to those requiring, say, transparency or preventing unfair competition. The promise of equal opportunity in this country is just as great as unfettered competition, and both are enshrined in our Constitution and statutes.  Simply equating the existence of a law with its abuse by overeager litigants is smoke screen—courts already have plenty of tools available to readily dismiss truly frivolous lawsuits. The real question is whether the standard is sufficiently well crafted so that courts can interpret it and so that it effectively deters public policies and private practices that would violate the Fair Housing Act.  Under that standard, the rule has been reviewed and upheld by all 11 federal circuit courts to review it.  The standard isn’t novel: it has been the law of the land and honed in the courts for decades.  During this time lenders have innovated and they've exercised judgment and discretion in underwriting.  And where this innovation and discretion has in fact had devastating consequences for protected groups, such as during the subprime mortgage crisis, the disparate impact standard under the Fair Housing Act was an essential tool to reign in abuses. Moreover, to strengthen the standard and minimize lawsuits, in 2013 HUD adopted a rule clarifying exactly what this longstanding policy requires and removing the inconsistencies in enforcement. Today and going forward, lenders can and should be informed about the law's requirements, enabling them to more effectively avoid legal action.  Simply put, the standard is finely crafted tool, not a blunt instrument, to accomplish its goal of improving equal opportunity and choice in housing and lending.


Earlier this week, I participated in a panel with Raj Chetty, discussing the implications of his latest research showing that poor kids growing up in racially segregated metros are less likely to achieve upward economic mobility than those from less segregated metros.  So the persistence of residential segregation imposes real costs on all of us, and on our country’s future. 

As I said earlier, I see the disparate impact principle as an important tool for addressing residential segregation and unequal housing outcomes for whites and people of color.  But no single tool will be sufficient for tackling this complex and long-standing tangle of problems.  Whether or not the Supreme Court upholds the disparate impact standard, public policies (and private sector practices) should advance more proactive strategies as well. 

Discrimination, information gaps, stereotypes and fears, and disparities in purchasing power all work together to perpetuate housing and neighborhood inequality.  The evidence argues for: enforcement—to combat persistent discrimination; education—about the availability and desirability of housing options in diverse neighborhoods; affordable housing development—to expand choice and open up exclusive communities; reinvestment—to equalize the quality of minority neighborhoods; and new incentives—to encourage and nurture stable diversity.


No matter what the Supreme Court decides in Texas Department of Housing and Community Affairs v. The Inclusive Communities Project, Inc., there will remain other important tools to combat housing discrimination.  Disparate treatment of borrowers and overt discrimination will remain illegal under the Fair Housing Act, as they should.   Furthermore, the Community Reinvestment Act obligates bank lenders to provide credit to the underserved communities where they operate.


Considering these factors, if the Supreme Court does decide that disparate impact cases cannot be brought under the Fair Housing Act , there should not be an negative effect on access to credit for underserved communities. In fact, instead of devoting resources to meeting an amorphous disparate impact standard, lenders will instead be able to redirect those efforts to actually providing credit to those communities that need it most.  

Unless the Supreme Court decides Inclusive Communities much more radically than most observers expect, the federal government will still need to take actions that affirmatively further fair housing. That obligation, in effect since passage of the Fair Housing Act, requires HUD grantees—states, cities, housing agencies, and others—to build inclusion into new communities and promote access to opportunity for everyone.

HUD long delayed effective measures to carry out this provision of the Fair Housing Act. But several state legislatures and courts have required public policies that promote neighborhood inclusion and housing choice for decades, showing that it’s possible to reverse income and racial segregation. This year, HUD will finalize its rule on how states and cities need to plan for inclusion. Seventy-four regions already did similar plans as part of their Sustainable Communities Regional Planning Grants. This experience has demonstrated that local governments, fair housing advocates, housing agencies, and community groups can work together to align housing programs with transportation, land use, and economic development to build more competitive and livable regions.

Though long overdue, it’s not too late to plan enthusiastically for inclusion. Today’s cities, suburbs, and rural communities are segregated because of the laws, rules, and practices of the past 200 years. But unlike many other wealthy societies, we’re still growing. By current projections, the U.S. will grow by nearly 50 million people between 2010 and 2030 alone. Our new families will need more housing in new and renewed neighborhoods. With laws, rules, and practices that truly reflect a longtime national principle—E Pluribus Unum—the America of 2065 could be full of more inclusive communities.

I appreciate Solomon’s point that lenders can be sued for many practices in addition to disparate impact. But there is a difference between being sued for disparate impact and being sued for, say, unfair business practices. In the disparate impact suits, once there is a showing of different results involving a protected class, the lender has the burden of proof to show that (1) there was as legitimate business reason for his actions, and (2) that all similarly situated people were treated the same.  Good underwriting, especially for potential borrowers with limited income or wealth or thin or less-than-pristine credit, involves judgment.  The lender who wants to make a loan to such a borrower wants to have a reasonable likelihood that the loan will be repaid (to the benefit of lender and borrower). And  this often requires balancing one factor—say, a strong rent payment record—against another, say, limited net assets.  Those judgments are vulnerable to disparate impact claims, as they will not necessary have the same impact on different racial/ethnic groups due to the differential distribution of income/wealth and credit scores. Moreover, the lawsuits are expensive to defend, and there is reputational risk when the lawsuit is filed, even if the lender eventually prevails. By contrast, in an unfair business practices lawsuit, the burden of proof is on the plaintiff to show that the lender did in fact engage in unfair business practices.

I disagree that the question is just about whether the standard is sufficiently well crafted so that courts can interpret it and so that it effectively deters public policies and private practices that would violate the Fair Housing Act. In fact, I would argue that placing the burden of proof on the lender, without the plaintiff even having to allege a purpose to discriminate, encourages the lenders, fearful about getting hit with a disparate impact lawsuit, to be overly formulaic in their lending practices. That is, they ignore mitigating factors that would actually open the credit box, benefitting those with less than pristine credit, who are often minorities. A Supreme Court ruling barring suits based on disparate impact alone  would give lenders space to exercise flexibility in their lending decisions, while, as Dave Stevens pointed out, leave ample scope to fight discrimination, covert as well as overt.    

Thank you to all of our contributors to this Urban Institute policy debate, especially Michael Bodaken from the National Housing Trust and David Stevens from the Mortgage Bankers Association.  I think this conversation revealed some very different perspectives about disparate impact — even among scholars at the Urban Institute — and I hope it helped our readers understand some of the implications of a forthcoming Supreme Court decision on the matter.  Please stay tuned for future policy debates on Urban’s new web platform

Comments are now closed. Thanks, all, for participating in this informative discussion! I hope you'll continue the dialogue on Twitter and on our blog.