YOUR QUESTION
You may have heard about a major change to Regulation B. They eliminated disparate impact. I also learned that they changed a few other areas that were working to reduce discrimination. As an underwriter, I think this is wrong-headed. I think this reduces fair lending protection.
We met with our lawyer because we have a second review process, which weeds out potential discrimination in our loan process. Our lawyer says there is a shift away from not having to prove intent to discriminate to now having to prove intent. She says that this is a problem because proving intent is extremely difficult. In other words, discrimination is now possible without having to prove intent to discriminate – only the outcome matters.
So, if I get this right, even if the outcome is discrimination, the company that discriminated won't be held responsible if you can't prove an intent to discriminate. I don't understand why disparate impact protection is being weakened. It’s scary!
Do the changes to Regulation B basically eliminate disparate impact?
OUR COMPLIANCE SOLUTION
OUR RESPONSE
I am going to be blunt: the CFPB's April 2026 Final Rule ("Rule") amending Regulation B eliminates the "effects test" – that is, "disparate impact" – of the Equal Credit Opportunity Act (ECOA), while also restricting special-purpose credit programs (SPCPs), and narrowing the definition of "discouragement" of applicants or prospective applicants. This is clearly a significant regulatory shift away from fair lending restrictions.
However, saying it eliminates disparate impact and fair lending is not accurate. The Rule eliminates disparate impact liability specifically under ECOA and Regulation B. That's significant, but ECOA is only one of several legal frameworks that govern lending discrimination. The Rule does not affect several others that remain fully intact.
The Fair
Housing Act (FHA) still recognizes disparate impact for mortgage lending. The
Supreme Court confirmed this in Texas Department of Housing v. Inclusive
Communities Project (2015), and the Rule expressly does not touch FHA
liability. So a mortgage lender whose policies produce racially skewed outcomes
can still face a disparate impact challenge under the FHA, which is a
completely separate statute.
State fair lending laws are arguably the bigger remaining protection. Many states – for instance, California, New York, Illinois, and others – have their own anti-discrimination statutes that incorporate disparate impact standards, and federal rulemaking cannot preempt those. State attorneys general were among the most vocal opponents of the Rule precisely because they intend to continue using their own authorities.
The Department of Justice retains independent enforcement tools. And the Community Reinvestment Act, which addresses lending patterns in lower-income communities, operates on its own separate framework.
HOW DID THIS HAPPEN?
The CFPB received over 64,500 public comments, including ours. The overwhelming majority of comments opposed the Rule. Nevertheless, the Rule is now law. The compliance effective date is July 21, 2026. Whatever the comments offered, pro or con, the Rule largely finalizes a November 2025 proposal, with only clarifying edits rather than substantive revisions.
Since your question specifically involves the change to disparate impact, I will discuss it primarily. The other changes are also very significant and should be incorporated into your policies and procedures.
Eliminating the “effects test,” a change supposedly meant to lower compliance costs, actually gives lenders greater freedom to target protected groups.
WHAT IS THE EFFECTS TEST?
The purpose of the “effects test” is ultimately to protect against disparate impact. The "effects test" is actually a legal doctrine used to determine if a lender’s facially neutral policy creates a discriminatory, disproportionate impact on a protected class (for instance, race, gender, or age). It means a creditor can be liable for discrimination, even without discriminatory intent, if their practices have a discriminatory effect.
Most regulators know full well that they can challenge lending policies that, while appearing neutral, create a negative impact on protected groups. Most compliance lawyers know full well that a financial institution can expose itself to a disparate impact violation by creating a pattern or practice that results from defective lending policies. And most financial institutions know, or should know, that if a policy has a discriminatory effect, they must prove that a legitimate business necessity justifies it.
What the CFPB has done is to remove the “effects test” from Regulation B, thereby promulgating that ECOA does not recognize disparate impact liability. The focus now is on the intent to discriminate.
Disparate impact is different from “disparate treatment,” which involves explicit, intentional discrimination by the lender acting with the intent to treat applicants differently, though it does not require malicious prejudice.
REMOVING DISPARATE IMPACT LIABILITY
The Rule removes disparate impact claims from ECOA enforcement, shifting the focus to intentional discrimination. It flows from President Trump's April 2025 executive order titled "Restoring Equality of Opportunity and Meritocracy," which directed federal agencies to deprioritize enforcement of statutes and regulations incorporating disparate impact standards.
Notably, as I’ve stated above, the Rule has no bearing on disparate impact liability under the Fair Housing Act or applicable state laws. However, this is a significant rollback of fair lending enforcement under ECOA because it is eliminating disparate impact liability (while also tightening the anti-discouragement requirements and restricting race-based special lending programs).
Critics, including consumer advocates and state attorneys general who submitted comments, raised concerns that the changes weaken protections against discriminatory lending. Both America's Credit Unions and the American Bankers Association supported the changes, with the ABA saying the framework would “advance the purposes of the ECOA, encourage prudent, risk-based underwriting, and discourage arbitrary government enforcement.”
HOW DID DISPARATE IMPACT WORK?
Disparate impact was the primary evidentiary tool available to federal and state governmental entities and private parties to enforce the anti-discrimination provisions under ECOA and Regulation B. Under that framework, a lender could be found liable if its facially neutral policies, such as requiring minimum credit score thresholds or debt-to-income ratios, produced discriminatory outcomes against a protected class, even without proof of intentional bias.
PROS AND CONS
Having carefully reviewed the Rule since its inception through its publication in the Federal Register on April 22, 2026, I have witnessed critics taking one side against the supposed harm, and supporters taking the other. So, let’s check out their core arguments.
Critics
Consumer advocates and some legal
experts have warned that eliminating disparate impact would weaken a
longstanding tool for identifying systemic discrimination. The concern is
practical: intentional discrimination is very hard to prove. Lenders
rarely put discriminatory intent in writing. Many discriminatory outcomes arise
from policies that appear neutral on the surface but are built on biased
assumptions or historical data, such as using zip codes, certain income types,
or automated underwriting models trained on historically biased lending data.
Without the disparate impact framework, those outcomes become much harder to
challenge.
In my view, this surely matters for
communities of color, who have historically faced patterns of credit
discrimination and redlining. Critics argue that the Rule fundamentally
reshapes 50 years of fair lending enforcement.
Supporters
Proponents frame the removal of disparate
impact differently. America's Credit Unions, for instance, argued that removing
disparate impact language will "reduce uncertainty and avoid chilling
innovative, inclusive credit programs." Basically, this argument suggests that
the threat of disparate impact liability sometimes leads lenders to avoid
offering certain products or serving certain markets altogether, out of fear
that any outcome disparity could trigger enforcement action that, in turn,
could limit access to credit. Personally, I find this argument unpersuasive. It
may contain some logical grounding, but it's weak as a standalone justification
for several reasons.
I have discussed this argument
with several legal and compliance professionals. So, first, as to the merits, disparate
impact liability does create legal risk for lenders who can't fully predict how
a new product will perform across demographic groups before they launch it. A
lender designing a novel underwriting model, say, one using alternative data,
such as rent payment history, to expand credit access, might hesitate if any
demographic disparity in outcomes could trigger an enforcement action. I acknowledge
that the legal risk may cause a “chilling" dynamic on new loan products.
But I think the argument breaks
down. The same logic could justify removing virtually any anti-discrimination
standard by arguing that liability for discrimination "chills"
business activity. The question isn't whether liability has costs, but whether
those costs outweigh the benefits of the protection. Once you understand that
conclusion clearly, you realize the argument reverses the direction of the
evidence.
Disparate impact liability has
historically been used to challenge exclusionary practices – such as
redlining, steering, and discriminatory appraisal criteria – that restricted
credit access for minority communities. The programs it most plausibly
"chills" are those that produce discriminatory outcomes, not
innovative, inclusive ones. A genuinely inclusive credit program, by
definition, tends to improve demographic outcomes rather than worsen them,
so that it wouldn't trigger disparate impact liability in the first place.
The inclusive credit programs are
also internally in tension with the restrictions on Special Purpose Credit
Programs, which were actually tools designed to expand credit access to
underserved groups. Thus, the Rule simultaneously restricts those while
claiming to promote inclusion, a contradiction that critics have noted.
WHAT’S A
STRONGER ARGUMENT FOR THE RULE?
I believe a more persuasive
argument for removing disparate impact isn't the "inclusive credit"
framing at all; it is the legal and textual one, specifically, that the ECOA's
statutory text doesn't clearly authorize effects-based claims the way the Fair
Housing Act does, and that the Supreme Court has never definitively ruled it
applies under ECOA. That's a legitimate legal argument about statutory
interpretation, even if you disagree with the policy outcome. The
"chilling innovation" framing, though, is much harder to defend
empirically.
MY VIEW
I will make no new friends in the
mortgage banking community with my view, but I think this is a classic
example of reframing a rollback of a protection as a benefit to the very people
that the protection was designed to help! The structure of the argument,
couched in a talking point like "removing this anti-discrimination tool
actually helps minorities," is rhetorically sophisticated but requires
strong empirical support to be persuasive. That support hasn't really been
provided. The CFPB's own rulemaking was predominantly opposed by consumer
advocates, state attorneys general, and members of Congress – not exactly a flashing
signal that the "inclusive credit" argument landed with the
communities most affected.
This article, the CFPB Eliminates Disparate Impact, published on April 29, 2026, is authored by Jonathan Foxx, PhD, MBA, the Chairman & Managing Director of Lenders Compliance Group, founded in 2006, the first and only full-service, mortgage risk management firm in the United States, specializing exclusively in residential mortgage compliance.
