Federal fair lending law recognizes three types of lending discrimination: disparate treatment, disparate impact, and redlining. They’re defined by how the discrimination operates, not by whether the lender meant any harm. A loan officer can discriminate without saying a discriminatory word, an algorithm can discriminate without anyone reviewing its output, and a bank can discriminate against a whole neighborhood without ever meeting the people it excludes. All three violate the Equal Credit Opportunity Act, the Fair Housing Act, or both.
Disparate Treatment
Disparate treatment is intentional. A lender treats you differently because of who you are: your race, color, religion, national origin, sex, marital status, age, receipt of public assistance income, or your exercise of rights under federal consumer credit law. For mortgage transactions, the Fair Housing Act adds familial status and disability to that list.
The unequal treatment doesn’t have to be dramatic to count. A loan officer who processes one applicant’s file in three days but sits on a comparable applicant’s file for two weeks is treating them differently. So is a lender who “forgets” to mention a lower-rate product to one borrower while volunteering it to another. What matters is whether the difference lines up with a protected characteristic and lacks a legitimate business explanation.
Steering
Steering is one of the most common forms of disparate treatment in mortgage lending. The originator pushes you toward a more expensive product even though you qualify for better terms, often because the pricier loan pays a higher commission. Federal rules prohibit mortgage originators from directing borrowers into loans that pay the originator more unless the loan genuinely serves the borrower’s interest. Steering minority applicants into FHA loans with mortgage insurance premiums while offering similarly qualified white applicants conventional loans without that cost is the textbook illegal pattern.
To counter steering, rules require mortgage originators to present meaningful options: the loan with the lowest rate, the loan with the lowest rate and fewest risky features, and the loan with the lowest upfront costs. Those options have to come from a significant number of the lenders the originator works with.
How It Gets Proven
Cases almost never hinge on a smoking-gun email. The usual approach is circumstantial. You show you belong to a protected class, that you were qualified for the loan, that you were denied or given inferior terms, and that comparable applicants outside your protected class got better outcomes. The burden then shifts to the lender to give a legitimate non-discriminatory reason. If that reason doesn’t hold up, the claim stands.
Fair housing organizations build these cases with matched-pair testing. Two testers with similar credit profiles and incomes apply for the same product, and investigators compare how the lender treats each. The Department of Justice and HUD use the same method in their investigations.
Disparate Impact
Disparate impact is harder to see because the lender isn’t targeting anyone. A policy that looks perfectly neutral, applied to everyone equally, still ends up excluding a protected group at a much higher rate than others. When that happens, the policy can violate fair lending laws regardless of what the lender intended.
The classic example is a minimum loan amount. Set the floor at $60,000, and every applicant faces the same rule. But if home values in predominantly minority neighborhoods cluster below that number, the policy effectively locks those neighborhoods out. The same logic reaches minimum income requirements, blunt debt-to-income cutoffs, and credit-score thresholds that correlate with race more than they predict default.
Disparate impact claims work in three steps. You show the policy produces a statistically significant disparity against a protected group. The lender then gets a chance to show the policy meets a legitimate business need with no less discriminatory alternative available. Even if the lender clears that bar, you can still win by identifying a less harmful alternative the lender refused to adopt.
Algorithmic Bias
Automated underwriting has made disparate impact more common and more difficult to spot. Algorithms trained on historical lending data can inherit the biases baked into that data. They can also lean on variables that function as proxies for protected traits: zip codes stand in for race because of decades of residential segregation, and even data points like phone type, shopping timing, or purchase locations can correlate with protected characteristics strongly enough to produce discriminatory results.
Technological complexity is not a defense. The CFPB has said lenders cannot justify noncompliance with adverse action notice requirements by pointing to an opaque model. A lender’s inability to explain its own algorithm doesn’t shield it from liability; the outcomes still count.
Redlining and Reverse Redlining
Redlining is discrimination by geography. Instead of evaluating individual borrowers, a lender draws lines around whole neighborhoods and restricts credit inside them. Historically, those lines tracked racial composition almost perfectly. The name comes from mid-century maps that federal housing agencies literally marked in red to flag minority neighborhoods as too risky for investment.
Explicit maps are gone, but the practice continues in quieter forms. A bank may decline to advertise in majority-minority neighborhoods, place no branches there, discourage applications from those zip codes, or apply tougher underwriting to properties in them. Regulators often uncover the pattern by analyzing Home Mortgage Disclosure Act data: when a lender approves loans across a metro area but skips its neighborhoods of color, that gap can support a redlining claim without any internal document telling employees to avoid those areas.
Reverse Redlining
Reverse redlining is the mirror image. Rather than withhold credit from minority neighborhoods, predatory lenders flood them with expensive high-risk loans. Borrowers who would qualify for standard-rate products elsewhere get pushed into subprime mortgages with inflated fees, steep prepayment penalties, and unnecessary add-ons like credit insurance. The absence of mainstream lenders in redlined neighborhoods creates the vacuum that predatory lenders fill.
Signs You May Have Experienced Lending Discrimination
Some warning signs show up during the application itself, and others only become visible later. Watch for:
- A denial or worse-than-expected terms after you were pre-qualified or verbally encouraged to apply.
- An originator who steers you toward one product and won’t discuss alternatives you’ve read about or asked for.
- Substantially different treatment (timing, documentation demands, follow-up) from what similarly qualified friends or family describe with the same lender.
- A denial notice that gives vague reasons like “did not meet internal standards” without specifics you can act on.
- A lender with no branches, marketing, or approvals in your neighborhood while operating actively in demographically different areas nearby.
Federal law backs one specific protection here: when a lender denies your application, approves it on worse terms than you asked for, or takes other negative action on an account, they must send an adverse action notice within 30 days. That notice must state the specific reasons for the decision or tell you clearly how to request them within 60 days. “Your score was too low” isn’t enough on its own; the reasons must be specific enough that you can see what went wrong and evaluate whether discrimination played a role.
What to Do If You Suspect Discrimination
Start documenting right away. Save every email, letter, loan estimate, and disclosure. Note the dates of phone calls, who you spoke with, and what they said. If the terms shifted between conversations, keep both versions. If a denial notice doesn’t include specific reasons, request them in writing.
The Home Mortgage Disclosure Act requires lenders to publicly report detailed data on nearly every mortgage application, including loan amount, property location, action taken, and the borrower’s race, ethnicity, sex, and age. That data is available through the CFPB and can help you see whether a lender’s approval pattern raises broader questions than your individual file could answer.
Watch the Deadlines
Every route for challenging discrimination has its own clock:
- A HUD administrative complaint under the Fair Housing Act must be filed within one year of the last discriminatory act.
- A private federal lawsuit under the Fair Housing Act must be filed within two years of the discriminatory practice. Filing a HUD complaint pauses that two-year clock while the agency proceeding is pending.
- A private lawsuit under the ECOA must be filed within five years of the violation.
Where to Report
For housing-related lending discrimination, you can contact the Department of Justice at 1-833-591-0291, email fairhousing@usdoj.gov, or file online. HUD accepts complaints at 1-800-669-9777 and through its online portal. For any type of credit discrimination, including non-mortgage lending, the CFPB accepts complaints at 1-855-411-2372 or through its website; the lender generally has 60 days to respond, and the agency publishes anonymized complaint data publicly.
Filing an administrative complaint doesn’t stop you from consulting a private attorney. Both the ECOA and the Fair Housing Act require courts to award reasonable attorney’s fees to a successful plaintiff, which is why many fair lending attorneys take these cases on contingency.