Adverse Impact: What It Means for Employers

What it means in practice

Adverse impact is the measurement at the heart of "disparate impact" discrimination. Disparate impact asks whether a neutral rule or test excludes one group at a much higher rate than another. It differs from disparate treatment, which is intentional discrimination. The EEOC gives the contrast: testing only African American applicants for reading ability is disparate treatment, while a physical agility test for everyone that disproportionately screens out women raises a disparate impact question. The main federal source of the measurement rules is the Uniform Guidelines on Employee Selection Procedures (UGESP) at 29 CFR part 1607, adopted in 1978 under Title VII of the Civil Rights Act of 1964.

What counts as a selection procedure. UGESP applies to tests and other selection procedures used as a basis for any employment decision, including hiring, promotion, demotion, referral and retention; selection for training or transfer can count if it leads to one of those decisions (29 CFR 1607.2(B)). A "selection procedure" is any measure, combination of measures or procedure used as a basis for an employment decision, from paper-and-pencil tests and performance tests to physical, educational and work experience requirements, informal interviews and unscored application forms (29 CFR 1607.16(Q)). Recruitment practices are not selection procedures (29 CFR 1607.2(C)). The EEOC also lists cognitive and personality tests, background and credit checks and performance appraisals as examples of selection procedures.

Who is covered. Title VII applies to employers with 15 or more employees for each working day in each of 20 or more calendar weeks in the current or preceding calendar year, and the protected bases are race, color, religion, sex and national origin. UGESP applies to persons subject to Title VII and other federal EEO requirements, but by its own terms not to the Age Discrimination in Employment Act or to disability under sections 501, 503 and 504 of the Rehabilitation Act (29 CFR 1607.2(D)).

How to calculate adverse impact. The joint agency Questions and Answers on UGESP describe four steps: (1) compute each group's selection rate by dividing the number selected by the number of applicants from that group; (2) find the group with the highest rate; (3) divide each group's rate by the highest rate to get an impact ratio; and (4) check whether any ratio is below four-fifths, or 80%. The analysis starts with the "total selection process": the combined effect of every step leading to the final decision. If the total process shows no adverse impact, the agencies in usual circumstances will not expect the individual components to be evaluated; if it does, each component should be evaluated (29 CFR 1607.4(C)). Men are compared with women, and each race or ethnic group with the highest-rate group; the Q&A says subgroups such as Black women need not be compared.

Worked example 1: a hiring screen. Hypothetical: 120 men and 90 women apply for warehouse jobs. The employer hires 48 men and 27 women. The men's rate is 48 / 120 = 40%; the women's rate is 27 / 90 = 30%. The impact ratio is 30% / 40% = 75%. Because 75% is below 80%, the four-fifths rule generally treats this as evidence of adverse impact against women. Had 34 women been hired, their rate would be about 37.8%, the ratio about 94%, and the rule would generally not indicate adverse impact.

Worked example 2: several groups. The Q&A gives hiring rates of 60% for whites other than Hispanics, 45% for American Indians, 48% for Hispanics and 51% for Blacks. Dividing each by 60% gives impact ratios of 75%, 80% and 85%. On that information alone, adverse impact is indicated for American Indians but not for Hispanics or Blacks.

Worked example 3: small numbers. Also from the Q&A: an employer selects 3 of 20 male applicants (15%) and 1 of 10 female applicants (10%). The ratio is 10% / 15% = 66 2/3%, which fails the four-fifths test, yet the agencies say the number of selections is too small to warrant a finding of adverse impact without more information, such as rates over a longer period. The reverse also holds: with large numbers, a difference smaller than 20% can be adverse impact if it is significant in both statistical and practical terms (29 CFR 1607.4(D)).

Who counts as an applicant. The Q&A says an applicant is a person who has indicated an interest in being considered, whether on a form or orally, depending on the employer's practice. A person who voluntarily withdraws at any stage is no longer counted, but standards that disproportionately discourage applicants of a group may need justification.

The Title VII burden of proof. Section 703(k), added by the Civil Rights Act of 1991 and posted with the Title VII text on eeoc.gov, says a disparate impact violation is established only if (i) the complaining party shows that the employer uses a particular employment practice that causes a disparate impact and the employer fails to show the practice is "job related for the position in question and consistent with business necessity," or (ii) the complaining party shows an alternative employment practice and the employer refuses to adopt it. Each particular practice must generally be shown to cause the impact, unless the decision-making process cannot be separated for analysis. Business necessity is not a defense to intentional discrimination (section 703(k)(2)). Separately, section 703(h) permits professionally developed ability tests that are not designed, intended or used to discriminate, and section 703(l) forbids adjusting scores, using different cutoff scores or otherwise altering test results on the basis of race, color, religion, sex or national origin.

Job related and consistent with business necessity. The EEOC fact sheet says an employer can meet this standard by showing the procedure is necessary to the safe and efficient performance of the job and evaluates skills related to the particular job, not general skills. UGESP describes three ways to show validity: criterion-related validity (the procedure predicts or correlates with important elements of job performance), content validity (its content represents important aspects of the job) and construct validity (it measures characteristics shown to be important for the job) (29 CFR 1607.5(B)). UGESP does not require validity studies where no adverse impact results (29 CFR 1607.1(B)). Where validation is not feasible, procedures should be as job related as possible and minimize adverse impact (29 CFR 1607.6(B)).

Less discriminatory alternatives. Where two or more procedures serve the employer's legitimate interests and are substantially equally valid, UGESP says the user should use the one shown to have the lesser adverse impact, and a validity study should include a search for suitable alternatives (29 CFR 1607.3(B)). The EEOC fact sheet describes a settlement in which Ford agreed to replace an apprenticeship cognitive test that had a statistically significant disparate impact with a procedure designed to reduce adverse impact.

Automated and software screening tools. The federal texts reviewed for this entry do not set a separate rule for software. UGESP's definition of a selection procedure covers "any measure, combination of measures, or procedure" used as a basis for an employment decision, and section 703(k) speaks of "a particular employment practice." The EEOC fact sheet notes that many tests can be administered online. Ranking or filtering tools in an applicant tracking system are therefore measured with the same selection-rate arithmetic as any other step.

Age: reasonable factors other than age. Under the ADEA (workers 40 and over, employers with 20 or more employees), a neutral practice that adversely affects older workers is discriminatory unless it is justified by a reasonable factor other than age (RFOA) (29 CFR 1625.7(c)). The person challenging the practice must isolate and identify the specific practice causing the disparity, while the employer bears the burdens of production and persuasion on the RFOA defense (29 CFR 1625.7(c)-(d)). The practice must be reasonably designed and administered to achieve a legitimate business purpose. Listed considerations include training managers to apply the factor fairly, limiting subjective judgments and assessing the impact on older workers (29 CFR 1625.7(e)). The defense is not available against a disparate treatment claim, or when age itself is a limiting criterion. See Age Discrimination in Employment Act.

Recordkeeping. UGESP says users should keep records showing impact by sex and by the race and ethnic groups in 29 CFR 1607.4(B). Users with 100 or fewer employees, and others not required to file EEO-1 reports, may instead keep yearly records of the number of persons hired, promoted and terminated for each job by sex, and where appropriate by race and national origin; the number of applicants for hire and promotion; and the selection procedures used (29 CFR 1607.15(A)(1)). Larger users should determine adverse impact at least annually for each group that is at least 2% of the relevant labor force or workforce (29 CFR 1607.15(A)(2)). Under 29 CFR 1602.14, application forms and other hiring and promotion records must be kept one year from the record or personnel action, whichever is later, and all relevant records must be kept until final disposition once a Title VII charge is filed. On July 23, 2026 the EEOC proposed rescinding the EEO-1 through EEO-6 reports and their related recordkeeping requirements; a proposal does not change current rules.

Current federal enforcement posture. Executive Order 14281 (April 23, 2025) states a policy "to eliminate the use of disparate-impact liability in all contexts to the maximum degree possible" and directs agencies to deprioritize enforcement of statutes and regulations to the extent they include disparate-impact liability, naming 42 U.S.C. 2000e-2, the Title VII section that contains 703(k). On June 9, 2026, the Justice Department's Office of Legal Counsel told the EEOC Chair that EEOC interpretations, including UGESP, embrace an unconstitutional reading of Title VII; that business necessity requires only that a practice rationally serve a valid business purpose; and that plaintiffs must show specific causation and an equally effective alternative. The opinion says it does not address disparate impact under the ADEA. The EEOC rescinded its affirmative action guidelines (29 CFR part 1608) on July 6, 2026. OPM removed UGESP references from federal civil service rules on July 31, 2026, stating the change does not alter federal agencies' obligations under Title VII. The Labor Department's rule rescinding the Executive Order 11246 regulations, including OFCCP's copy of UGESP at 41 CFR part 60-3, takes effect October 26, 2026, and says it does not affect other agencies' application of UGESP. As of October 22, 2026, 29 CFR part 1607 still appeared in the eCFR, and the Title VII text on eeoc.gov still contains section 703(k).

Where states differ. State fair employment laws can add their own rules. As one example, the California Civil Rights Department announced that regulations effective October 1, 2025 make clear that using an automated-decision system may violate California law if it harms applicants or employees based on protected characteristics, and require employment records, including automated-decision data, to be kept for at least four years. Rules vary by state, so check your state agency.

FAQ: Is the four-fifths rule the law? The Q&A calls it a "rule of thumb," not a legal definition, used to flag serious discrepancies.

FAQ: Does passing the four-fifths rule prove there is no discrimination? No. The Q&A says it speaks only to adverse impact; unlawful discrimination may still be shown through other evidence.

FAQ: Does the four-fifths rule apply to age? UGESP excludes the ADEA (29 CFR 1607.2(D)); age claims use the reasonable-factor-other-than-age standard instead.

Why it matters and common mistakes

How Remindax helps

Related terms

See all terms in Employment Law & Compliance or browse the full HR Glossary.