So, does the use of US Civil Service Test 21 to screen applicants for the police training program violate the constitution?
The disparate impact alone is not enough to shift the burden to the defendant to show that the discrimination is justifiable. Strict scrutiny requires intent.
SCOTUS says that this test makes perfect sense. Lots of things have disparate impacts: income tax, welfare, etc. We don't want to dilute the meaning of "invidious:" if disparate impact were enough to trigger strict scrutiny in an employment case, there would be all kinds of unintended consequences. Athletic teams, in certain sports, are often dominated by one race, for example.
De facto segregation can have multiple concurrent causes, and it's not necessarily a horribly bad thing.
Now if the court were to strike down this test, it would have wide ranging effects for all kinds of jobs (i.e., this test is used for lots of things. And it has been an important tool in terms of getting rid of cronyism/nepotism/corrupt hiring practices. A merit-based exam serves some good objectives, and before we get rid of it, we should be comfortable with what's going to replace it.
True, though, that it's not necessarily an excellent predictor of success as a police officer (that would be hard to measure; and come to think of it, how do we even define "success" for police officers?), but communication skills are clearly important, and it does at least correlate with success in the training program.
Bar exams, by the way, do a good job correlating with success in law school, but what is success as a lawyer: income? teaching? becoming a judge?
Congress did let Griggs v. Duke Power stand (though it did not apply to state and local government), so the other side has some precedent on its side: there a disparate impact was a sufficient finding. Also, racism is subtle nowadays: almost nobody thinks they're racist-- they all thing a bunch of other people are-- so it's hard to find intentional discrimination because it's cloaked. And making this a constitutional holding makes it awfully permanent, especially when there are very good reasons to get minorities onto the police force (there were riots underway).
Additionally, tests like this cut those who don't do well at them off from all kinds of opportunities: they're a gateway into the middle class, and the best predictor of success on them is the socio-economic status of your parents, which is heavily racially skewed in the US. Upholding the test makes it harder to break that cycle.
Plus, once you've got a test like this, those in the system want to keep raising the standard as a means of making themselves more valuable.
Two bases of law:
So does disparate impact violate the constitution in some way? That was the standard in Washington v. Davis (because Title VII didn't apply), and the court said no. But in 1991 Title VII was amended to include disparate impact, and it now applies to public entities, so now we have the question of whether that statute controls here.
So the test here is the basis for promotion, which makes it instantly controversial. And they went to some lengths to design a decent test, including soliciting input from minority members about what makes a good firefighter leader. So is that in itself a problem? It's disparate treatment: they're intentionally going the extra mile in the construction of the test. The court completely overlooks this, because they've got bigger fish to fry, but basically the principle is that a nod to affirmative action is generally acceptable.
Likewise with the oral exam panels: the examiners were from elsewhere (to avoid local biases, but they're also 66% minority). Again, a deliberate effort to overcome suspicion of prejudice in favor of whites.
You could buy the materials to prepare for the test, but they're not cheap ($500)... will that have a disparate impact? Yes, for two reasons: first, wealthy families are disproportionately white, and second, the families most likely to already have a set of materials in the family (or help studying from a firefighter relative) are disproportionately white.
Ricci personally attended SCOTUS arguments, incidentally, but it's not clear whether the court knew he was there. He's white, of course, but he has a bunch of learning disabilities. Now reading might be a skill that you should take into account when you're promoting firefighters into leadership spots. But he really works his tail off studying for this: 8-13 hours/day... how much of that was on the job?
The results of the test are not as expected. New Haven is 40% black, 20% hispanic, and 30% of the firefighters are black, but only 9% of the officers. This is too much disparity to happen at random: we've got a disparate impact here. Is there disparate treatment (i.e., is there intent to keep the black officers down)?
Well, everyone expects the test to have a disparate impact (else why all the fancy design procedures), so perhaps even giving the test is an intentional act of discrimination, if you know the test is going to have undesirable results. They're masking the means of discrimination behind an attempt to construct a facially neutral test, but the fact that they're still using a test is maybe still discriminatory against blacks. On the other hand, the way the test is set up is discriminatory against whites. So maybe those cancel each other out.
SCOTUS doesn't get into that-- they just say that whatever the facts may show, it's not intent to discriminate against minorities. A disparate treatment allegation just isn't on the table here.
The city is worried about a lawsuit because of the disparate impact, and they also don't want to alienate the 60% of their citizens whose ethnic cohorts didn't do well on the test. So they throw out the test. This makes the white side unhappy: it's clear discrimination.
Summary judgment for the defendants: everyone's test was thrown out, and nobody got any promotions, so nobody can complain. Somehow the court things that throwing out the test because the wrong race succeeded isn't itself racial. 2nd Cir. (including Sotomayor) gives a 1-¶ opinion saying "we agree with the lower court." They duck the issue of why they come out that way.
Kennedy is fairly unsympathetic to affirmative action, and he was the swing vote here. He comes up with a test: if there is disparate treatment, you need to have a "strong basis in evidence" that the disparate impact of not doing so is either illegal or unconstitutional. There definitely was disparate treatment here (throwing out the test because too many whites passed), you need a strong basis in evidence that the disparate treatment was warranted.
This is probably something like strict scrutiny: nobody is going to survive that statement. Anyway, what was the evidence mounted on the side of the black firefighters? The disparate impact of the test, of course: it is representative of a long history of discrimination in the firefighting ranks. But SCOTUS says fear of a lawsuit alleging disparate impact is not itself a strong basis.
Scalia: we've been putting this issue off for decades-- sooner or later we're going to have to say whether disparate impact is enough to violate equal protection. Of course, the inevitable consequence of making disparate impact a cause of action is quotas (in hiring). Actually, practically all significant statutes have some disparate impact. The tax code, social security, etc. He's basically implying that there should not be a statutory basis for suits based merely on disparate impact. If disparate impact (as in Title VII) is an adequate basis for employment suits, then isn't it also for other violations of equal protection (education, etc.)?
Officially quotas are a dirty word, but the concept is kind of lurking behind a lot of affirmative action. So the court does a lot of dancing around these issues.
Unlike the lower courts, the dissent doesn't allege that there was no discrimination. It says that the real problem is the test itself. The attack is twofold. First, there's the obvious disparate impact. And second, maybe this is deliberate.
The dissent also has major federalism and separation of powers arguments: the court doesn't need to intervene when minorities are getting the long end of the stick, only the short end.
So did Brown mean "act with all deliberate speed to get to substantive racial equality," or "end state-sponsored discrimination?" The majority would pick the latter, and the dissent the former.
The reason objective tests were developed was to diminish the effects of prejudice (particularly, out east, with respect to Jewish applicants to educational institutions). So it's ironic to downplay merit and substitute racial categories.
And it's really strong, since it refers to private housing.
This is 6 years prior to Brown and contemporaneous with military desegregation.
SCOTUS almost never hears private cases, so what are they doing here? Maybe sending a message that they don't like segregation. They say this becomes a public matter when you call upon the apparatus of the state to enforce your private contract. And the court isn't allowed to do this-- it would be a violation of equal protection.
So parties can discriminate all they like, but they can't get the court to help them out. And there was concern about the case: people thought this meant every dispute was public.
Of course, 20 years later, we had Jones v. Alfred Meyer holding that the civil rights statutes prohibited discrimination by private people. So it's a statutory violation, which is easier to find than a constitutional one. As a result, Shelley is sort of a dead letter, even though it is famous.
First off, just after the civil war, there was a lot of law passed designed to stick it to the southern racial ideology and the southern slavery economy.
However, when these statutes got to court (see the Civil Rights Cases), the courts weren't very sympatheic. Feelings about the south receded somewhat, and the courts wound up not being champions of integration (this changed later, of course). So the federal courts effectively rendered most of these statutes nugatory. But congress never really did repeal them. So, as attitudes began to change, they started to become interesting, and might have given recourse, had congress not passed the modern civil rights statutes.
The big statute comes in 1964. 42 USC §1980 et seq.
These aren't based on 14A's permission to congress to enforce stuff by appropriate legislation-- they're grounded in the commerce power. So, after Lopez, holding that there are limits to ICC powers, should we be nervous that the civil rights statute is vulnerable? Seems quite unlikely, but some folks are nervous, and would like to see it grounded under 14A.
For the moment, we're mostly concerned with Title VIII of the 1968 act: federal fair housing (42 USC §3601-14). Originally it covered race, religion, color, or national origin. Nowadays we've got race, family status, and handicap status as well.
There's always going to be a case that challenges the policies behind any given law, and this one is a classic. A bit like Ricci: in order to achieve the end of integration, we engage in discrimination.
Majority: plain meaning rule-- this is not legal.
Dissent: some great quotes. Literal interpretation of statutes can kill them. And this is sad: we're destroying one of the most integrated housing projects in America.
Basically what we have here is a clash between policies. We want integration, and we don't want discrimination. And, on practical terms, this is voluntary integration: all the residents know what they're getting into, and the elect to do it. But to the majority, if we start tolerating overt discrimination (i.e., under these circumstances if we say the benefits of integration justify a litte of this discrimination), we're on a slippery slope. Also, the quotas here are cieling quotas on integration: they act to prevent white flight. So we're making sure that the minorities stay minorities. The plaintiff here was a black applicant who was denied, incidentally. So this is overt discrimination, and we just can't tolerate it.
Clearly we've got disparate impact legislation. SCOTUS says disparate impact alone is not enough-- there has to be intent (or a statutory violation). But there is a fair housing law, so maybe zoning that has a disparate impact is covered by federal fair housing laws. SCOTUS says Arlington Heights wins on the constitutional issue, but remands to the lower court (where Arlington Heights had lost previously) for consideration on the fair housing issue. And wouldn't you know: Arlington Heights loses.
So we get Mount Laurel II, a very strong opinion, as opposed to the restraint of Arlington Heights. And even though a few apartments got built, there wasn't much effect. And the legislature allowed affluent suburbs to bargain with other areas, paying them money to cover their Mount Laurel deficit, and not having to take the apartments.
Once again, a clash of policies: majorities have to integrate because it's a good thing, but you can't make minorities integrate because it hurts them.
The dissent has nice points: everyone on the petty jury was hispanic, as was the prosecurity, and the judge. And also if hispanics are upset about hispanic representation on grnd juries, a 79% majority can certainly change the jury commission (elected, and also hispanic, it seems).
So, can lawyers for a company being sued by a black plaintiff use their challenges racially? No. This is public action (as in Shelley v. Kraemer), so you can't do it.
Of course, cynically, you just don't say you're doing this on the basis of race. But we can look for disparate impact over time. Anyway: with respect to all peremptory challenges, you can't base them on race.
Scalia's dissent points out that what's sauce for the goose also sauces the gander: this affects minorities' ability to ensure a racially diverse jury.
Another point: once you put the judge as overseer of jury selection, we lose some of the benefit of the adversarial process.