Caseflicks

Supreme Court of the United States • 1976

Washington v. Davis

426 U.S. 229 | 96 S. Ct. 2040 | 48 L. Ed. 2d 597 | 1976 U.S. LEXIS 154

Full access

Unlock the video and quiz

The written brief is free to read below. Subscribe to watch the video explainer and take the quiz.

Create an account Log in

Takeaway

In short, this case established that a racially disparate impact alone does not violate constitutional equal protection; plaintiffs generally must prove discriminatory purpose, while statutory regimes such as Title VII may impose more demanding disparate-impact rules.

Background

Black applicants challenged the District of Columbia Metropolitan Police Department’s use of “Test 21,” a general federal-service examination testing verbal ability, vocabulary, reading, and comprehension. Applicants had to score at least 40 out of 80 before entering the Department’s 17-week recruit-training program. Black applicants failed the test at a substantially higher rate than white applicants.

The District Court found no allegation or proof that officials had intentionally discriminated. It nevertheless considered the disparate impact sufficient to require an explanation from the government. The court upheld the test because it was reasonably and directly related to the recruit-training program, and because the Department had made affirmative efforts to recruit Black officers. The Court of Appeals reversed. Treating Title VII’s disparate-impact principles as constitutional law, it held that disparate impact alone established a Fifth Amendment violation unless the District proved that Test 21 measured actual police-job performance. The Supreme Court reversed.

Issues

Issue #1

Whether the Court could correct the Court of Appeals’ use of Title VII standards to decide a Fifth Amendment claim even though that error was not presented in the petition for certiorari.

Holding

Yes. The Court could notice the lower court’s plain error under its rules.

Reasoning

The applicants’ motion for partial summary judgment rested solely on constitutional grounds, and the Court of Appeals had held that Title VII’s disparate-impact framework governed that constitutional claim. But Title VII did not supply the cause of action in this case: when the suit began it did not cover federal employees, and the applicants had not amended their complaint to assert a Title VII claim after the statute was extended to government employment.

Because the Court of Appeals had applied the wrong legal standard to the constitutional question, the Court invoked its authority to notice a plain error not raised by the petitioners. Correcting that foundational error was appropriate before deciding whether Test 21 violated the Fifth Amendment.

Issue #2

Whether the Fifth Amendment’s equal-protection component invalidates a facially neutral public-employment test solely because it disproportionately excludes Black applicants.

Holding

No. Disparate racial impact alone does not establish an equal-protection violation; plaintiffs must show that official action had a racially discriminatory purpose.

Reasoning

The Fifth Amendment’s Due Process Clause includes an equal-protection component that bars the federal government from invidious racial discrimination. But the Court distinguished constitutional equal-protection doctrine from Title VII, under which Congress had specifically authorized disparate-impact challenges to employment practices without requiring proof of discriminatory intent.

The Court’s prior jury-selection, voting-district, school-desegregation, and social-welfare cases showed that a constitutional claim ordinarily requires purposeful discrimination. A facially neutral measure may be unconstitutional when it is administered discriminatorily, and discriminatory purpose may be inferred from the totality of circumstances. But a racially disproportionate result, standing by itself, does not trigger strict scrutiny or prove an equal-protection violation.

Disparate impact remains evidentiary relevant because a stark or otherwise unexplained pattern may support an inference of discriminatory purpose. Here, however, Test 21 was facially neutral, sought to measure a legitimate governmental concern—minimum verbal and communicative ability—and was related to the recruit-training program. The Department’s affirmative recruitment of Black officers and its increasing proportion of Black recruits further undermined an inference that the test was a device for racial exclusion.

The Court declined to constitutionalize Title VII’s more demanding validation requirement. Treating every neutral law with unequal racial effects as constitutionally suspect would invite heightened judicial review of a broad range of governmental policies, including tax, welfare, licensing, and public-service measures. The Court concluded that any extension of disparate-impact liability beyond statutory settings was for Congress, not the Court.

Issue #3

Whether Test 21 was invalid under the statutory and civil-service job-relatedness standards argued by the parties because it had not been shown to predict performance as a police officer.

Holding

No. The test was sufficiently job related because the record supported the District Court’s finding that it predicted success in the police recruit-training program.

Reasoning

Although the applicants had not pleaded a Title VII claim, the parties and lower courts had litigated the statutory issues under standards similar to Title VII’s job-relatedness requirement. The District Court therefore resolved the statutory claims when it granted summary judgment to the District and federal defendants, and the Court of Appeals should have affirmed that part of the judgment as well.

The Court concluded that an entrance test may be validated by showing that it predicts an applicant’s ability to succeed in a job-related training program; the government need not always prove a direct correlation between the entrance-test score and later performance on the job. Police training necessarily prepares recruits for the demands of police work, and minimum verbal and communication skills are useful, if not essential, for mastering that training.

A Civil Service Commission validation study and expert evidence supported the District Court’s factual finding that Test 21 scores positively correlated with success in the recruit school. That finding was not clearly erroneous. Griggs and Albemarle did not require a different result on this record, and the Court rejected a remand to investigate changes to the training program during the litigation. Any present-day deficiency under Title VII, the Court stated, should be challenged through the procedures applicable under that statute.

Concurrences

Justice Stevens

Reasoning

Justice Stevens agreed with the judgment and generally accepted that purposeful discrimination is required for a constitutional violation. He stressed, however, that purpose and impact are not cleanly separable in every case. Objective evidence of what government actually did can be highly probative of intent, particularly because governmental decisions are often collective, compromised, and motivated by several considerations. In extreme cases such as Gomillion and Yick Wo, the disparity may be so dramatic that describing the inquiry as one of purpose rather than effect makes little practical difference.

He would rest the constitutional result on narrower grounds than the majority. In his view, Test 21 served the legitimate and neutral purpose of imposing a uniform minimum literacy standard relevant to police work; there was no evidence that the passing score was arbitrarily high; and variations among high schools justified a separate uniform examination. He did not rely on the District’s good-faith recruitment efforts, because neither good faith nor recruitment efforts would save a test that was otherwise constitutionally invalid.

On the statutory question, Justice Stevens emphasized that this was not formally a Title VII case and that Title VII standards could not simply be transplanted wholesale into the distinct statutory provisions at issue. He nevertheless agreed that the law allowed a police department to use a test predictive of success in a training program, even without independent proof that it predicted field performance. He also agreed that the District Court’s finding of a correlation between Test 21 and training performance had enough evidentiary support to survive appellate review.

Dissents

Justice Brennan

Reasoning

Justice Brennan, joined by Justice Marshall, first argued that the Court should not have decided the statutory issues at all. The majority faulted the Court of Appeals for importing Title VII standards into a constitutional claim, yet then decided statutory questions under 5 U.S.C. § 3304 even though the applicants had not pleaded a claim under that provision either. In his view, the Court should not reach statutory questions unnecessary to resolving the case.

If the statutory issue was to be decided, Justice Brennan would have affirmed the Court of Appeals because the District had not carried its burden of proving that Test 21 was job related. The parties and lower courts had treated Title VII principles as controlling, and the Civil Service Commission itself viewed its job-relatedness standard as consistent with Title VII. Under those principles, once disparate impact was shown, the employer had to validate the test through a meaningful connection to job performance.

The claimed validation study showed only a correlation between Test 21 scores and average scores on written recruit-school examinations. It did not establish that the recruit-school examinations measured mastery of the curriculum, that the curriculum taught skills necessary for police work, or that either training scores or Test 21 scores predicted job performance for Black officers. Indeed, all recruits ultimately passed the training examinations, sometimes with additional assistance, which weakened the claim that a particular examination average measured meaningful training success.

Justice Brennan read Griggs and Albemarle to require that a discriminatory employment test be demonstrably related to job performance, either directly or through professionally acceptable validation methods. Permitting a general verbal-ability test to be validated merely by its correlation with another written test risked allowing employers to reproduce the same verbal-skills measure at two stages and call the resulting correlation job related. He warned that the majority’s approach could significantly weaken statutory protections against discriminatory employment practices.