Disparate treatment vs disparate impact: the two Title VII theories across the hiring funnel

Disparate treatment is intentional discrimination: a candidate is treated less favorably because of race, color, religion, sex or national origin, so the case turns on the employer's motive. Disparate impact is a neutral practice, such as a test, a degree requirement or a screening cutoff, that excludes one group at a higher rate and cannot be justified as "job related for the position in question and consistent with business necessity"; intent does not matter. Both theories sit in Title VII, at 42 U.S.C. 2000e-2 sections (a) and (k). As of October 2026 the EEOC prioritizes disparate treatment and has stopped litigating disparate impact, but the statute and private lawsuits under it are unchanged.

This page sets the two theories side by side for hiring: how each is proved, how employers defend, where each shows up in a typical funnel, and what changed in 2025 and 2026. For the impact theory applied to AI screening tools, see Title VII disparate impact and AI hiring tools; for the arithmetic, see the four-fifths rule.

What is the difference between disparate treatment and disparate impact?

The Supreme Court drew the line in a footnote that courts still quote. In Teamsters v. United States (1977) it called disparate treatment "the most easily understood type of discrimination," where the employer "simply treats some people less favorably than others," and said that under the disparate impact theory, "proof of discriminatory motive ... is not required."

The damages row comes from 42 U.S.C. 1981a, which allows compensatory and punitive damages only for intentional discrimination, "not an employment practice that is unlawful because of its disparate impact," and caps them at $50,000 for employers with 15 to 100 employees, $100,000 for 101 to 200, $200,000 for 201 to 500, and $300,000 above 500.

How is disparate treatment proved?

Most hiring cases have no direct evidence of motive, so courts use the burden-shifting framework from McDonnell Douglas Corp. v. Green (1973). In its original hiring form:

1. The candidate's prima facie case: they belong to a protected group, applied and were qualified for a job the employer was filling, were rejected, and the position stayed open or went to someone else. The Court noted the elements adapt to different facts.
2. The employer's reason: the employer must "articulate some legitimate, nondiscriminatory reason" for the rejection.
3. Pretext: the candidate gets a fair chance to show the stated reason was a pretext for discrimination, for example because comparable candidates of another group were treated differently.

Two further rules matter. Under 2000e-2(m), a violation is established when a protected characteristic "was a motivating factor for any employment practice, even though other factors also motivated the practice"; if the employer proves it would have made the same decision anyway, 2000e-5(g)(2)(B) limits relief to declaratory and injunctive relief and fees, with no damages or order to hire. And since Ames v. Ohio Department of Youth Services (2025), a candidate from a majority group meets the same standard as anyone else, with no extra "background circumstances" showing. A company-wide policy of intentional discrimination is litigated as "pattern or practice" disparate treatment, the theory in Teamsters.

How is disparate impact proved?

Griggs v. Duke Power Co. (1971) struck down a high school diploma requirement and two aptitude tests that disqualified Black applicants at a much higher rate without being shown to predict job performance. The Court said Title VII "proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation," and that "good intent or absence of discriminatory intent does not redeem" procedures that act as "built-in headwinds" unrelated to measuring job capability. Congress wrote the burden of proof into 2000e-2(k) in 1991:

1. The candidate identifies a particular practice and shows it causes a disparate impact; each practice must be identified separately unless the steps of the decision cannot be separated.
2. The employer shows the practice is "job related for the position in question and consistent with business necessity."
3. The candidate can still win by showing a less discriminatory alternative practice that the employer refused to adopt.

Interviews are not exempt. In Watson v. Fort Worth Bank & Trust (1988), a bank with no formal criteria left promotion decisions to supervisors' subjective judgment, and the Court held that "subjective or discretionary employment practices may be analyzed under the disparate impact approach in appropriate cases." It reasoned that otherwise an employer could add "brief interviews" to any test and escape Griggs entirely. The statistical first step is usually a selection-rate comparison; the four-fifths rule page works one through.

Where does each theory show up in a hiring process?

The same structured method reduces both risks: one set of job-related questions for every candidate, written scoring anchors, and notes that record what each person actually said. How to reduce bias in interviews covers the method, and what is a structured interview the basics.

Can one decision raise both theories at once?

Yes, and the Supreme Court has said how to resolve the conflict. In Ricci v. DeStefano (2009), New Haven discarded promotion exam results after white candidates outperformed minority candidates, fearing a disparate impact suit. The Court held that "race-based action like the City's in this case is impermissible under Title VII unless the employer can demonstrate a strong basis in evidence that, had it not taken the action, it would have been liable under the disparate-impact statute." Throwing out results because of who passed is disparate treatment.

The practical lesson for hiring teams: check a test, screen or interview format for impact before you use it, not after you see who passed. And never fix an impact problem by adjusting scores or cutoffs by group; Title VII forbids that in 2000e-2(l). The same logic is why deliberate diversity preferences are judged as treatment, not impact; see is diversity hiring legal.

What changed in 2025 and 2026?

What did not change: the text of 2000e-2(k), which only Congress can amend; the Supreme Court decisions above, which bind courts while the OLC opinion binds only the executive branch; a candidate's right to file a charge and, after a notice of right to sue, go to court; and state anti-discrimination laws. Disparate treatment, the EEOC's stated priority, is if anything more exposed than before. Age and disability have their own versions of both theories; see age discrimination in hiring and ADA accommodations in interviews.

When should a hiring team think about which theory?

• Before adopting any rule that applies to everyone (a requirement, test, knockout question, AI screen or cutoff), ask the impact question: does it screen out a group, and can you show why it matters for this job? Write the job-related reason down.
• Whenever a person makes a judgment about an individual (a screen, an interview rating, a debrief, an offer), ask the treatment question: would we have decided the same way for a comparable candidate of another group, and does the record show the reason?
• When numbers look uneven, investigate the practice and its job relevance; do not change individual outcomes by group.
• When a rejected candidate complains, the record decides most cases; see what happens when a rejected candidate files an EEOC charge.

What do people get wrong about the two theories?

• "No intent, no problem." True only for treatment. Impact liability never required intent.
• "The EEOC stopped enforcing disparate impact, so it is gone." The statute stands, and private plaintiffs and states can still use it.
• "Business necessity covers any sensible reason." Business necessity is a defense to impact only. It is never a defense to intentional discrimination; the statute says so in 2000e-2(k)(2).
• "Balanced final hires mean we are fine." A challenger can target the particular step that caused the disparity, and treatment claims are about individuals, not totals.
• "We can fix a skewed result by re-weighting scores." That is score adjustment by group, which 2000e-2(l) prohibits, and a treatment claim in waiting.
• "BFOQ can justify a race preference." The statute allows a bona fide occupational qualification only for religion, sex or national origin; see what a BFOQ is.