Why federal agencies still need to defend hiring standards

Nauval Wildani/Getty Images

In October 2024, the Justice Department’s Civil Rights Division reached a proposed $2.75 million settlement with the Maryland Department of State Police after alleging that the agency’s written exam and physical fitness test disproportionately excluded Black and female applicants and were not job-related or consistent with business necessity. The agreement required monetary relief and priority hiring opportunities for up to 25 previously disqualified candidates who met lawful hiring requirements.Four months later, in February 2025, new Justice Department leadership moved to dismiss the case. Two months after that, President Trump signed Executive Order 14281, “Restoring Equality of Opportunity and Meritocracy,” which directed federal agencies to deprioritize disparate-impact enforcement to the maximum extent permitted by law.For federal managers, the shift is real, but it should be described precisely. The underlying statutes did not disappear. Title VII’s disparate-impact framework remains codified in federal law. The Supreme Court has recognized disparate-impact claims under the Fair Housing Act.Private Title VII plaintiffs can still bring disparate-impact employment claims, and private Fair Housing Act plaintiffs can still bring disparate-impact housing claims. Title VI is different: Under Alexander v. Sandoval, private plaintiffs cannot enforce disparate-impact regulations under Title VI, leaving agency enforcement especially important in that area.The Equal Credit Opportunity Act is also in flux. In April 2026, the Consumer Financial Protection Bureau finalized a rule stating that ECOA does not authorize disparate-impact liability under Regulation B, a position that changes federal regulatory enforcement but does not prevent future litigation over the statute’s meaning.The doctrine being deprioritized has a specific employment-law origin. In Griggs v. Duke Power Co. (1971), the Supreme Court read Title VII to prohibit facially neutral employment practices that screen out protected groups when the practices are not meaningfully related to job performance.Congress later codified the disparate-impact framework in the Civil Rights Act of 1991. Under that framework, a plaintiff must identify a specific employment practice that causes a disparate impact; the employer may defend the practice by showing that it is job-related and consistent with business necessity; and the plaintiff may still prevail by showing that a less discriminatory alternative was available and refused.That is more exact than saying that statistical gaps alone constitute discrimination, although statistical gaps often trigger the burden and cost of defending a selection system.The cost record is substantial. The Maryland matter is one example. Gulino v. Board of Education, a long-running New York City teacher-certification case, produced judgments and payouts on an extraordinary scale, with public reporting placing the city’s potential exposure near $1.8 billion and the New York City comptroller identifying Gulino as a major driver of salary-claim payouts.Federal departments have also entered consent decrees involving police, fire and other public-sector selection systems across several decades. The recurring operational pattern is familiar to HR leaders: A test produces adverse impact, the agency must defend the relationship between the test and the work, and the litigation risk often makes settlement more attractive than years of validation fights.The premise underwriting this enforcement model deserves direct examination. Disparate-impact enforcement does not formally require perfect proportionality, but it often encourages agencies to treat demographic imbalance as presumptively suspicious before causation has been established.That is a serious analytical problem. Groups differ in median age, geography, educational preparation, occupational concentration, language background, prior experience, applicant-pool composition, self-sorting and dozens of other variables that may sit upstream of any hiring decision.A serious federal hiring system must distinguish barriers created by the agency from variation that arises before the agency’s selection process begins.That said, federal managers should not read the executive order or later regulatory changes as permission to abandon validation. A selection device that does not predict job performance is bad management regardless of who passes it.Reading comprehension may be defensible for troopers who prepare incident reports. Physical capacity may be defensible for officers expected to respond to emergencies. Content knowledge may be defensible for teachers responsible for classroom instruction.The defensible question is not whether a test produces identical pass rates across groups; it is whether the test measures something the job actually requires, whether it does so fairly, and whether the agency can explain that relationship before a court, an inspector general, a union, Congress or the public.Three concrete steps follow for federal HR leaders.First, review existing consent decrees, corrective-action plans and workforce-reporting practices for provisions tied to disparate-impact theory, particularly where the agency is relying on older regulatory assumptions that may no longer reflect current federal enforcement policy.Second, document the job-relatedness of any selection device the agency intends to keep, because private Title VII litigation remains available and statistical evidence will continue to appear in court even if federal enforcement priorities have changed.Third, separate workforce reporting that identifies possible agency-created barriers from reporting that merely records demographic variation. The first is a management problem. The second is a descriptive fact.Confusing the two has absorbed years of public-sector bandwidth that should have been spent improving mission performance.The durable lesson for federal hiring is that standards should be defended through evidence, validation and operational relevance rather than through demographic balancing by another name.Robert J. Choi is a former government executive and public-sector consultant. He previously served in the Central Intelligence Agency and as deputy chief people officer at the Metropolitan Transportation Authority.