Concept
Neutral Laws and Disparate Impact
Rational basis review
Neutral Laws and Disparate Impact
The Fourteenth Amendment forbids a State to “deny to any person within its jurisdiction the equal protection of the laws.”1 Earlier sections showed that the tier of review depends on how a law classifies. Sometimes the text settles that question. A minimum age for a driver's license sorts by age on its face,2 and a rule that elementary school teachers must be citizens sorts by alienage, since it burdens aliens and no one else. When a law discriminates on its face, its text establishes the prima facie case.3 Even then the tier follows the line drawn. A statute that penalizes the aged draws no strict scrutiny, because age is not akin to the classifications the Court has found suspect.4
Many laws never mention race or sex and still fall harder on one group than on another. Such a law is facially neutral, and its uneven effect is called disparate impact. The question is whether that effect turns a neutral law into a classification that invites heightened scrutiny. Washington v. Davis answers no, unless the plaintiff also proves discriminatory purpose: impact alone cannot create a racial classification.5
The answer chooses the tier. A court that finds purpose has found a racial classification, and strict scrutiny technically applies, but in practice the law always falls, because a purpose to harm racial minorities is not even legitimate, let alone compelling. A court that finds no purpose applies only rational basis review, and the government wins.6 A plaintiff who proves a disparity but not a purpose therefore lands in the deferential tier, which is why this material belongs in a chapter on rational basis.
The disparity is identical in both cities, and only the proof of purpose changes the tier. The rest of this section follows that proof: Washington v. Davis fixed the rule, Feeney defined purpose, and Arlington Heights showed how a court finds purpose when no lawmaker admits it.
Washington v. Davis
Washington v. Davis supplies the rule. The respondents challenged Test 21, a written aptitude test used to screen potential police officers.7 It sought to ascertain whether applicants had reached a particular level of verbal skill, and the government gave it generally to prospective government employees.8 The respondents invoked the Fifth Amendment's Due Process Clause, and the Court analyzed their claim as one of equal protection.9
No one alleged intentional discrimination. The respondents claimed only that Test 21 bore no relationship to job performance and “has a highly discriminatory impact in screening out black candidates.”10 The district court found that a higher percentage of African Americans than whites failed the test and that it had not been validated to measure later job performance. It denied relief anyway, because the test was reasonably related to the requirements of the police recruit training program and was not designed or operated to discriminate against otherwise qualified Black applicants.11 The court of appeals called the lack of discriminatory intent irrelevant and held impact alone sufficient to establish a violation unless the petitioners proved that the test adequately measured job performance.12
The Supreme Court rejected that approach. Disproportionate impact, it said, “is not irrelevant, but it is not the sole touchstone of an invidious racial discrimination forbidden by the Constitution. Standing alone, it does not trigger the rule, that racial classifications are to be subjected to the strictest scrutiny and are justifiable only by the weightiest of considerations.”13 The holding matters because of what the tier does. Under strict scrutiny, racial classifications are presumptively unconstitutional, and the government must show that the classification is narrowly tailored to a compelling interest.14 Because Test 21 drew no racial classification, the Court asked only whether it was rational, and a standardized aptitude test for screening potential police officers easily was.15
The Court's reasoning starts from the neutrality of the test. It had difficulty understanding how a law establishing a racially neutral qualification for employment denies equal protection simply because a greater proportion of one racial group fails to qualify.16 A government may modestly seek to upgrade the communicative abilities of its employees rather than accept a lower level of competence, particularly where the job demands skill in speaking and writing. The Black respondents could no more claim a denial of equal protection than could white applicants who also failed, and proof that more Black than white applicants were disqualified would not change the conclusion.17 The Court put the rule in one sentence: it had not held that “a law, neutral on its face and serving ends otherwise within the power of government to pursue, is invalid under [equal protection] simply because it may affect a greater proportion of one race than of another.”18
Impact keeps a role as evidence. In the jury cases, the total or seriously disproportionate exclusion of Black citizens from venires may for all practical purposes demonstrate unconstitutionality, because the discrimination is very difficult to explain on nonracial grounds.19 More generally, the justices permit evidence of disparate impact to prove discriminatory motives, and when plaintiffs establish a disparate impact, the government must offer a plausible race-neutral reason for the skewed outcome.20 Impact is evidence of purpose, not a substitute for it.
Defenders of the purpose requirement offer two arguments. The first is formal: on the anticlassification view, equal protection means that government may not treat similarly situated individuals differently, and a discriminatory purpose is part of the definition of that wrong.21 The second is practical: too many laws would be vulnerable to challenge if a discriminatory impact sufficed, and courts would constantly be examining them closely.22
The dissent shows what the rule can cost. It stressed that the petitioners, the officials defending the test, did not maintain that Test 21 scores correlated with job performance, and that their own validity study could not discern a significant positive relationship between training averages and job performance. In its view, nothing proved a correlation, direct or indirect, between Test 21 and the performance of a police officer's job.23 The majority needed no such proof, because a neutral test reviewed only for rationality need be no more than a reasonable way to seek better communication skills.24
The rule has a boundary, and it appears when officials administer a neutral law along ethnic lines. In Yick Wo v. Hopkins, a board granted permits to operate laundries in wooden buildings to all but one of the non-Chinese applicants and to none of the 240 Chinese applicants. A Chinese alien who had run a laundry for many years was refused a permit, convicted of a misdemeanor for operating without one, and imprisoned.25 That pattern could be explained only by race, so impact alone sufficed. In Davis, by contrast, one test faced every candidate, and the Court noted that no one challenged the individual selection of police officers.26
Two questions remain after Davis, and the next parts of this section answer them. The first is what discriminatory purpose means, because Davis required it without defining it.27 The second is how a plaintiff proves purpose when lawmakers do not announce it. On an exam, a stem that reports unequal pass rates on a neutral, evenly administered test should send you looking for evidence of purpose beyond the numbers. Without such evidence, the test is reviewed for rationality and usually stands. If purpose is proved, the test is treated as a racial classification and strict scrutiny follows, which in practice means the test falls.28
Feeney and Discriminatory Purpose
Davis required discriminatory purpose without saying what it is, and Personnel Administrator of Massachusetts v. Feeney supplied the definition.29 The case involved a preference for veterans in government employment that had a significant disparate impact on women, few of whom were veterans.30 Under the Massachusetts statute, every veteran who qualified for a state civil service position had to be considered for appointment ahead of any qualifying nonveteran, and the preference operated overwhelmingly to the advantage of men.31 When the suit began, over 98% of the veterans in Massachusetts were men and only 1.8% were women, while more than a quarter of the state's population were veterans.32 Because the impact was undisputed and severe, the case isolates what Davis left open: what a challenger must show about the lawmaker's state of mind. The Court relied on the reasoning of Davis and Arlington Heights to reject the challenge.33
Sex discrimination follows the same logic as race. Davis upheld a test that white candidates passed in proportionately greater numbers than Black candidates, since nothing showed that racial discrimination entered into the test's establishment. Arlington Heights upheld a zoning decision that tended to perpetuate segregated housing, since apart from its effect the decision was only the application of a neutral zoning policy. Those principles, the Court said, apply with equal force to alleged gender discrimination.34 The tier explains why the question matters. Classifications by gender must bear a close and substantial relationship to important governmental objectives, but when the basic classification is rationally based, uneven effects on particular groups within the class are ordinarily of no constitutional concern.35
The first step went against the challenger. Veteran status is not uniquely male, and although few women benefit from the preference, the class of nonveterans is not substantially all female. Significant numbers of nonveterans are men, and all nonveterans, male and female, are placed at a disadvantage. Too many men are affected to permit the inference that the statute is a pretext for preferring men over women.38
The second step is where the definition lives. Discriminatory purpose, the Court held, “implies more than intent as volition or intent as awareness of consequences.” It implies that the decisionmaker selected or reaffirmed a particular course of action at least in part “because of,” not merely “in spite of,” its adverse effects upon an identifiable group.39 The two phrases sort lawmakers by what they were after. A legislature that rewards veterans while knowing that most are men treats the cost to women as a by-product and acts in spite of it. A legislature that adopts the same rule in order to keep women out of the jobs has made the effect its object. American law uses two different definitions of purpose, depending on the field, and Feeney chose the one that requires desire: intent rests on the wish to bring about the consequences, not on knowledge of them.40 The same principle governs claims of race discrimination.41
The line sits at aim rather than awareness because of how legislation works. Most laws classify, and many affect certain groups unevenly. The calculus of those effects, the manner in which a law reverberates in society, is a legislative and not a judicial responsibility.42 A rule that charged lawmakers with discriminatory purpose whenever they foresaw an uneven effect would invite the very flood of challenges that defenders of the purpose requirement fear.43 Critics answer that officials who act with full knowledge of the consequences for a historically disadvantaged group should at least bear a burden of justification, and that such knowledge should raise a rebuttable presumption that the action is unlawful.44 That view is not the rule.
Feeney's own record shows what acting in spite of looks like. The preference began in 1896, when there were no women veterans and the statute covered only Civil War veterans, most of them beyond middle age and relatively few of them competing for public jobs. Its impact on women was then slight.45 Nothing in the record showed that the preference was originally devised or later re-enacted to accomplish the collateral goal of keeping women in a stereotypic and predefined place in the Massachusetts Civil Service. The substantial edge granted to veterans, the Court said, may reflect unwise policy, but the challenger had failed to demonstrate that the law reflected any purpose to discriminate on the basis of sex.46
Foreseeability is evidence, not proof. Mere awareness of a foreseeable and anticipated disparate impact will not by itself support a finding of discriminatory purpose, though the Court has held that foreseeable consequences may be used as one of several kinds of proof from which an inference of purpose is drawn.47 In discerning the purposes behind facially neutral policies, the Court has considered the degree, inevitability, and foreseeability of any disproportionate impact and the alternatives reasonably available.48 A challenger who shows only awareness has shown that lawmakers acted in spite of the effect, and needs further evidence to show that they acted because of it.
An exam stem that describes a neutral law with a severe, even predictable, effect on women or on a racial group is testing this definition. Look for facts showing why the lawmakers acted. A recital that they knew, or were warned, proves awareness only. Without proof of purpose, the neutral line is reviewed for rationality and usually stands, as the veterans' preference did.
Proving Discriminatory Purpose
Davis and Feeney tell a plaintiff what must be proved about a facially neutral law, but lawmakers seldom announce that they are acting because of race. The harder question is how a court finds a purpose that no one admits. Village of Arlington Heights v. Metropolitan Housing Development Corp. supplies the framework. The Village had denied a rezoning request tied to the developer's plan to erect integrated housing.50
The plaintiff need not prove that discrimination was the sole or dominant purpose. Rarely can a legislature or administrative body operating under a broad mandate be said to have acted on a single concern, or even on one purpose that was “dominant” or “primary.” Proof that discriminatory purpose was a motivating factor in the decision is enough.51 The Court had earlier observed that the search for legislative purpose is elusive enough without requiring that primacy be ascertained, and that removing even a subordinate purpose may shift altogether the consensus of legislative judgment supporting a statute.52 Courts refrain from reviewing the merits of decisions that balance numerous competing considerations, absent arbitrariness or irrationality. But racial discrimination is not just another competing consideration, and once it is shown to be a motivating factor, that deference is no longer justified.53
Direct evidence of motive is scarce, so the Court called for a sensitive inquiry into whatever circumstantial and direct evidence is available.54 A court weighs the following kinds of evidence together instead of scoring them one by one:
- The impact of the decision, meaning whether it bears more heavily on one race than another.55
- The historical background of the decision, particularly if it reveals a series of official actions taken for invidious purposes.56
- The specific sequence of events leading up to the decision.57
- Departures from the normal procedural sequence.58
- Substantive departures, particularly if the factors the decisionmaker usually treats as important strongly favor a decision contrary to the one reached.59
- The legislative or administrative history, especially contemporary statements by members of the decisionmaking body, minutes of its meetings, and reports. In extraordinary instances the members might be called to testify, although privilege frequently bars such testimony.60
Impact comes first on the list but almost never carries a case alone. Absent a pattern as stark as that in Gomillion or Yick Wo, impact is not determinative, and the court must look to other evidence.61 Yick Wo showed the pattern among applicants, where no Chinese applicant among 240 received a permit while nearly every other applicant did.62 Gomillion showed it in a boundary line. A statute altered Tuskegee “from a square to an uncouth twenty-eight-sided figure” and was alleged to remove all but four or five of its 400 Black voters without removing a single white voter. Justice Frankfurter wrote that, if the allegations were proved, the conclusion that the legislation was solely concerned with fencing Black citizens out of town would be “tantamount for all practical purposes to a mathematical demonstration.”63 In both cases the disparity was so extreme that no race-neutral explanation fit. A weaker disparity leaves room for such explanations, and the plaintiff must then turn to the other kinds of evidence.
Hunter v. Underwood shows impact and historical background working together. The Court invalidated a 1901 Alabama constitutional provision that disenfranchised anyone convicted of a crime involving moral turpitude. The plaintiffs established impact by showing that the provision had disenfranchised ten times as many Black as white citizens, and purpose through historical materials showing that a “zeal for white supremacy ran rampant” in the convention that adopted it.64 The state answered that the real purpose was to disenfranchise poor whites as well as Black citizens. That answer failed, because it conceded that discrimination against Black citizens was a motivating factor.65
Purpose proved in one setting can sometimes be carried to another. In a school desegregation case, the plaintiffs proved that the board had intentionally segregated schools in one part of the district but could not prove a similar purpose for other segregated schools in the same district. The Court let them satisfy the purpose element of their prima facie case for the other schools by in effect transferring the board's invidious intent from one part of the district to another.66
Comparators can defeat an inference of purpose as well as support one. The defendant in Wayte v. United States argued that his prosecution for failing to register for the draft was discriminatory because the enforcement policy targeted vocal non-registrants who were exercising their First Amendment right to protest. The Court concluded that the policy did not in fact target them, because other violators who were similarly situated but had not protested had also been prosecuted.67 Yick Wo is the mirror image: the similarly situated applicants outside the disfavored group received the permits that the Chinese applicants were denied.68
Statistics show how little a real but diffuse disparity can prove. In McCleskey v. Kemp, the petitioner offered the Baldus study, which analyzed 230 variables that could explain sentencing disparities on nonracial grounds. One model concluded that, even after accounting for 39 nonracial variables, defendants charged with killing white victims were 4.3 times as likely to receive a death sentence as defendants charged with killing Black victims, and that Black defendants like McCleskey who killed white victims faced the greatest likelihood of a death sentence.69 The Court held the study insufficient. To prevail, McCleskey had to prove that the decisionmakers in his own case acted with discriminatory purpose, and he offered no evidence specific to his case, relying solely on the study.70 Because discretion is essential to the criminal justice process, the Court would demand exceptionally clear proof before inferring that it had been abused, and the study was clearly insufficient to support an inference that any decisionmaker in his case acted with discriminatory purpose.71 His claim against the legislature failed for the same want of proof. He would have had to show that the Georgia Legislature enacted or maintained the death penalty statute because of an anticipated racially discriminatory effect.72 The Court had found in Gregg v. Georgia that the capital sentencing system could operate in a fair and neutral manner, and there was no evidence then, and none now, that the legislature enacted the statute to further a racially discriminatory purpose.73
The challengers in Arlington Heights lost because their record held a disparity and little else. The Court observed that a discriminatory effect can be considered in assessing intent where, for example, government action departs from official policy or produces a clear pattern unexplainable on grounds other than race, and it found neither. The Village had consistently applied its “buffer zone” policy in the past. Critics note that the Court never explained why a clear pattern was absent when racial minorities made up 18% of Chicago-area residents but African Americans were only about 1/20th of 1% of the Village's population (27 of 64,000 residents), and the Court implicitly accepted that the rezoning denial was valid under rational basis review.74 The opinion itself shows what would have changed the case. If the property had always been zoned R-5 but suddenly was changed to R-3 when the town learned of the developer's plans for integrated housing, the Court said it would have “a far different case.”75 The property and the plan are the same in both versions, and only the sequence of events differs. That sequence turns the routine application of a policy into evidence that the policy was invoked because of the project's racial makeup. Housing plaintiffs have also found another route. Lower federal courts generally agree that proof of a racially discriminatory effect establishes a prima facie case that a zoning decision violates the Fair Housing Act, even without evidence of discriminatory intent, and the Act has become a powerful weapon against exclusionary zoning.76
Proof that discrimination was a motivating factor does not end the case. The burden then shifts to the government to show that it would have taken the same action anyway, and the court decides whether it would have. If the court decides that it would have, there is no discriminatory purpose; if not, the court holds that there is one.77 An additional lawful aim does not satisfy that burden, as the state learned in Hunter. Only proof that the discriminatory aim added nothing to the result does.
An exam stem on this topic hands you a neutral law, a disparity, and scattered facts about timing, process, and statements, and the task is to sort those facts. A sudden change in zoning or policy just after a project is announced, a skipped review, or a recorded remark about the group to be kept out all point toward purpose. A long-standing policy applied consistently, ordinary procedure, and a bare statistic point away from it. A large disparity with nothing more, as in the capital-sentencing study, fails at step three, and the answer is rational basis review.