WATSON v. FORT WORTH BANK & TRUST
Legal provision: Civil Rights Act of 1964, Title VII
ORAL ARGUMENT OF ART BRENDER, ESQ. ON BEHALF OF THE PETITIONER
Chief Justice William H. Rehnquist: We'll hear argument this morning in Number 86-6139, Clara Watson v. Fort Worth Bank and Trust.
Mr. Brender, you may proceed whenever you're ready.
Mr. Brender: Mr. Chief Justice, and may it please the Court:
The question that this case presents is whether or not Clara Watson can test her employer's subjective employment practices by disparate impact theory.
We ask the Court to reverse the decision of the 5th Circuit.
I'd like to discuss some of the pertinent facts and then discuss the five main points or our reasons for asking the Court to reverse this case.
Those reasons in summary are that the Griggs v. Duke Power that announced the doctrine of disparate impact did not carve out an exception for subjective practices, that the legislative history of the 1972 amendments to Title VII also endorsed both disparate impact and the prior case law.
Furthermore, that the EEOC guidelines have applied disparate impact to subjective practices since at least 1970, that there is no bright line between objective and subjective practices that the Respondent would have this Court adopt, and, lastly, that the adoption of this exception would urge employers or at least foster employers more use of subjective practices.
Clara Watson, after having nine years experience as a cashier at a Montgomery Wards, applied in 1973 to Fort Worth National Bank as a teller.
She applied actually several times before she was hired.
When she was hired, she was hired not as a teller but as a proof operator.
The bank had at that time four black employees, none of them in positions that were visible to the public, and the proof operators and the two proof... I mean printing employees were in the basement and there was a cafeteria employee and a porter.
After three years on the job, she was made a teller at the bank and then, after seven years on the job, in 1980, she first applied for a position, for a supervisory position and actually, in February of '80, there were two supervisory positions that were available; one as the supervisor of the lobby tellers and the other as supervisory of the motor bank tellers.
She applied for each and in each case, she was refused the job and a white male in one instance, white female in another instance, was selected.
One year later, the same two positions became available and she sought both of those positions again with the same result.
A white female was selected for the lobby supervisor and a white male for the motor bank supervisor.
In 1984, even long after the 1981 refusal, she... there were still no black directors, officers or supervisors at the Fort Worth Bank and Trust, despite the fact that the bank is within a few miles of the two main areas of concentration of the black population in Fort Worth, a city that has about a twenty-two percent black population.
The bank system of both hiring and assigning persons to jobs within the bank was one that relied on the decisions of the supervisors, department supervisors, with very little interference from the upper management.
Those supervisors were, of course, all white.
The compensation system that the bank used was a series of... they used evaluations for both supervisory personnel and for line personnel, but those evaluations were filled out by the same supervisors and, once again, all were reviewed by a salary review committee.
The salary review committee rarely changed or affected the decision of those supervisors.
The... both the District Court and the Court of Appeals found that this limited group of white department heads made virtually all the hiring and promotion decisions as well as evaluations used to compute compensation and that the upper management routinely approved those decisions.
Clara Watson, at trial, based on a stipulated data base, presented a very comprehensive statistical study which showed that in the area of hiring, blacks were hired at a rate of 3.5 percent while whites were hired at a rate of 14.8 percent, almost four times greater, and that figure was disputed somewhat by the bank, who claimed that they only hired people... we used all the applications and all the persons who were hired, they disputed that and said we only hire people whose applications were on file for two weeks.
So, during trial, that was also analyzed, although we were able to show that that really wasn't true because--
Unidentified Justice: Well, we don't have a hiring case anymore, do we?
Mr. Brender: --No, but the same supervisors that were making those hiring decisions were also making the promotion decisions.
Unidentified Justice: I mean, all that's left now is the promotion aspect for an individual plaintiff?
Mr. Brender: That's correct, although the hiring case actually has been remanded to the District Court where it's still pending.
Unidentified Justice: Yes, but we're not concerned with that here?
Mr. Brender: That's correct, Justice O'Connor.
Unidentified Justice: Mr. Brender, I assume that the proof offered as trial, the statistical evidence, is exactly the same evidence that would be offered either under a disparate treatment or a disparate impact case.
Mr. Brender: I think that's correct, Justice O'Connor.
We... it could be used either way, and I think this Court's footnote in Teamsters says that the matter of proof may be applied to either theory.
Unidentified Justice: And if it were remanded, you wouldn't have different proof, it would still be the same statistical evidence?
Mr. Brender: Yes, I believe so.
Unidentified Justice: So, what is really at stake here?
Does it turn on the different burden of proof aspects under the two types of cases?
Does it turn on what the defendant's burden is to rebut the statistics, in effect?
Is that how you see the bottom line?
Mr. Brender: Probably both.
As I see it, to apply statistical evidence under a disparate treatment theory or pattern of practice theory, that the level of proof is greater.
This Court has said--
Unidentified Justice: Well, we haven't said that, have we?
Mr. Brender: --Not exactly.
I may be interpreting that.
Unidentified Justice: And I'm just wondering if there's any difference at all in the initial burden of proof under the two types of cases.
Mr. Brender: Well, I think there is in this sense, because at least under Castenada v. Partida, where the Court addressed the two or three standard deviation argument on applying an intent theory, that would be greater that under Griggs, where you could show under the eighty percent rule, the three-fourths rule, you wouldn't necessarily have to have statistical significance.
In other words, two or three standard deviations is either.05 or.01 statistical significance, but the eighty percent rule wouldn't have to come to that level.
So, I think there would be a difference in terms of statistics.
Now, in fact, our statistics were basically statistic significant.
So,... but I do think there is that difference statistically.
The other thing, in the 5th Circuit held this, that in a pattern and practice case, they held it had to be a class action because they refuse... the majority at least refuse to review this under a disparate treatment pattern and practice theory.
There certainly are some cases, and I believe Judge Goldberg cites those in his dissent, where courts have held the statistics are not sufficient in and of themselves to show disparate treatment.
Unidentified Justice: So, the courts below, although you have the statistical evidence, decline to even treat it as a pattern and practice case--
Mr. Brender: The--
Unidentified Justice: --because the class was discharged, she wasn't representative of the class?
Mr. Brender: --I believe that's what they did because they just simply said this was not a... we had disbanded the class or we have upheld the District Court's feeling to that extent, and, but it really wasn't analyzed at all nor, of course, would they allow us to analyze it under the disparate impact.
Unidentified Justice: Suppose the Court had allowed you, what would you have done?
Mr. Brender: --Well, of course, we attempted in the District Court to prove it.
We were pointing in that direction, but I don't think we would have... I don't know that we would have done anything different with the statistical evidence.
Unidentified Justice: Well, what would you have done?
You said the court wouldn't let you analyze it under disparate impact.
What if it had said okay, go ahead, what would you have done?
Mr. Brender: Oh, okay.
I think the disparate impact proof is there.
I mean, I think we did present--
Unidentified Justice: What would you have said?
Mr. Brender: --Maybe I'm not understanding your question, Justice White, but I--
Unidentified Justice: Well, how would you... how could you have won the case?
Mr. Brender: --Oh, Well, I think it would have then been up to the employer to come forward and justify on business necessity their practices, which they didn't do.
They did not attempt to prove business necessity or validation of their processes.
Unidentified Justice: So that in disparate impact case, you really don't need to prove intent, do you?
Mr. Brender: No, no.
You wouldn't have to prove intent.
Unidentified Justice: And in disparate treatment, you do?
Mr. Brender: That's correct.
And in this case, they did not present a statistical case or a statistical expert.
Unidentified Justice: In the disparate impact, is the bottom line, nevertheless, intent in the sense that the employer must justify as in Griggs or what do you conclude, intent or--
Mr. Brender: I don't believe they have to prove intent.
I think that a disparate impact case can be totally devoid of intent.
Unidentified Justice: --But under a pattern and practice case, you don't directly prove intent.
You offer statistics as evidence of... from which the trier of fact could infer.
Mr. Brender: Could infer intent.
Unidentified Justice: Right?
Mr. Brender: That's correct.
Unidentified Justice: And in actuality, the proof is much the same from the plaintiff's side, isn't it?
Mr. Brender: Well, I think proof can be used both ways, but I think the disparate impact probably hones in more on the group discrimination, whereas, you know,--
Unidentified Justice: My impression is that the plaintiff's initial proof would be exactly the same and what you might be quibbling about is what the defendant's burden then is under the two types, and I guess we haven't spoken to that, really.
Mr. Brender: --No.
The only disagreement I have with that is a lot of the language in the disparate... in the pattern and practice cases talks about a pervasive all invasive system of discrimination and so forth that's intent, and I interpret that to mean you've got to basically show statistical evidence of intent at every stage of the process; whereas, I think under disparate impact, you can simply show that the overall result is disproportionate and then the employer must come in and justify those.
Unidentified Justice: And he justifies it by showing job-related.
Mr. Brender: That's correct.
Unidentified Justice: So, he doesn't rebut any notion of intent directly.
Mr. Brender: No, no.
He could rebut the case by trying to show that there is no impact.
In other words, he could come in and try to--
Unidentified Justice: Or he could show, as I understand it, that even if there is an impact, there is a job-related reason for giving it.
Mr. Brender: --That's right.
Either by validating it under the procedures set out in the EEO guidelines or by simply showing that it's necessary to the efficient operation of the business.
Unidentified Justice: And if he fails to do that, he's liable not because he intends anything, but because as the law is interpreted, intended to give relief--
Mr. Brender: That's right.
Unidentified Justice: --Griggs, I guess.
Mr. Brender: That's Griggs, Your Honor.
Griggs says that by the effect of the impact, he must then come in and justify those practices.
If he's unable to justify those practices, then he's liable.
If he is, then he can continue on with the practice and he's not liable despite the impact.
Unidentified Justice: Mr. Brender, may I just... I'm not entirely clear on what you think the employer has to show in a case in which the practice being challenged is the practice of using subjective discretionary approach by the supervisors at each branch office.
You say you proved statistically that this has resulted in discriminatory impact.
How can the employer justify that?
Does he have to prove that he must, that it's necessary for him to use discretionary--
Mr. Brender: Justice Stevens, I think he can do it several ways.
I think that... and the courts have addressed those.
First, he can validate under the EEOC guidelines and, of course, we have cited and talked about the brief of the APA which shows that contrary to what is being argued by the Respondent, subjective procedures are validated.
There are legitimate ways in which you can validate those type of criteria.
The other way--
Unidentified Justice: --By that, do you mean you validate each factor that the local manager uses or validate the practice of using multi-factor discretion?
I'm a little puzzled.
Mr. Brender: --I think one problem with the facts of this case is that the subjective... I don't know whether it's subjective process being used, but there was really no criteria that was out there for which the supervisors were supposed to make decisions.
But I think... so, that system that they were using, I think, is kind of far out from where a general subjective hiring practice decision may be or promotion practice, whatever, but I think that... for instance, the practice that they had, which was kind of a gut-level system, I want to--
Unidentified Justice: The people who were familiar with the employees in the best position to make a judgment as to which would be the best supervisor, and there's really no way to prove that that's a necessary way to hire people, is it?
Mr. Brender: --Well, I think there is, but I think you could also validate it.
In other words, you could simply take... for instance, one way you could validate this system would simply be to take upper management and say okay, give us a series of fact situations that you would want your supervisor... the supervisors would handle, from routine things, like a social security check doesn't get credited, to things like, for instance, attempted forgery.
Give us those situations, tell us how a successful supervisor would deal with those in a way that we can rate those things in terms of knowledge of the job, knowledge of the bank procedures, knowledge of maybe the law in some instances, the speed with which they resolve them, and those sorts of things.
And then what you do is you simply take the two groups, you have the group--
Unidentified Justice: It seems to me you're saying don't do it on a subjective basis, that you must have a series of objective criteria that can be reviewed.
Mr. Brender: --Only to validate, only to validate.
In other words, if you go through that process, you take and you test the people who are the successful people, the supervisors, and then you test those who are the applicants but did not succeed, and you find that there is a statistical difference on the side of those who are selected.
In other words, you find that they are... they were, in fact, selected properly, then you could continue on with your gut-level system.
In other words, that only applies--
Unidentified Justice: And if you find they made a few mistakes, why, then, you're in trouble.
Mr. Brender: --If you made some mistakes, you're going to have to either find an alternative process or put in some criteria in this case that would--
Unidentified Justice: And these are all by hypotheses innocent mistakes?
I mean there is no racial animus motivating any of this.
Mr. Brender: --There wouldn't have to be.
Unidentified Justice: Well, I mean, if you can prove racial animus, you don't need any of this theory.
Mr. Brender: Well, if you don't... as I see disparate impact, you could have racial animus.
That doesn't imply intent or lack of intent.
Unidentified Justice: Mr. Brender, the sort of system that you suggest for validating in response to Justice Stevens' question may be find for a company like the Fort Worth Bank and Trust Company that has a number of employees.
How about an employer who employs just a minimum number that's subject to Title VII?
Is he supposed to go through all of that, too?
Mr. Brender: First of all,--
Unidentified Justice: What is it, fifteen people or twenty people?
Mr. Brender: --First of all, Your Honor, I think the thing we've got to realize is that you have to be able to show impact first, and I do think in smaller employers, it's going to be difficult to show an impact simply because of the number of decisions would not be sufficient to show statistically that there's an impact.
But assuming that you could do it, there are other ways to validate.
That would not be the only way.
Unidentified Justice: What would a small employer of twenty people, how could he validate, you know, that sort of subjective system that I think most of those employers probably use?
Mr. Brender: --I think, first of all, there are the transportability provisions to the guidelines.
Those are things that say... for instance, a bank teller's job is not that much different than a bank teller's job anywhere else.
You can take studies that have already been done.
Unidentified Justice: But even an employer of twenty people would have to go to studies and that sort of thing.
Mr. Brender: Only if, only if they have a disparate impact.
Unidentified Justice: Of course, he doesn't know at any given time whether his practices may be shown to have a disparate impact or not.
I take it that doesn't depend on intent.
Mr. Brender: Well, it doesn't depend on intent, but he is under the guidelines required to be looking at his practices and aware of his practices under the provisions of self-help provisions of the guidelines, where an employer is encouraged to try to eliminate racial bias or racial effects in his employment practices.
Unidentified Justice: Well, Mr. Brender, isn't there something short of this so-called validation that would suffice to determine or establish that a particular employer's system of evaluating or promoting employees is generally fair and uniformly applied?
It seems to me there is a body of case law out there to that effect.
Mr. Brender: There is, and we have cited some of those cases in our reply brief, where courts have simply held that on different sorts of things, one of them that comes to mind is the Zahorik v. Cornell University, where their process of assessing people for tenure was found to be... to satisfy business necessity.
Unidentified Justice: I mean, you would concede, I suppose, that for supervisory and managerial positions, it probably isn't possible to always use subjective standards in hiring, that there will almost always be a subjective component in hiring and promotional decisions for that kind of job.
Mr. Brender: Well, I think there's--
Unidentified Justice: Isn't that right?
Mr. Brender: --I think that's true.
I think... however, I do think that you can narrow subjective things.
I don't think there's a fine line between subjective and objective.
For instance, by their definition, a test would be objective, but an essay test could be very, very subjective.
By the same token, things like personal appearance, which were on the rating form that they used and they had what they termed was a professionally-developed form which they just simply didn't use for promotions, but that can be very, very objectively determined.
If the criteria are set out, that's what this case addressed in... this Court addressed in Albemarle, that if the criteria are set out and they are specific, it really narrows subjectivity considerably, and I think attempts... some of the case law simply attempts to try to get at narrower and more objective standards, have been held to be justified by business necessity, and the same thing with regard to the other factor that the Respondent raises about cost studies.
There's certainly no evidence in this record that such studies would be costly.
There's no evidence at all.
They didn't attempt to produce that.
Unidentified Justice: Well, it must go without saying that a strict validation procedure would, indeed, be very costly, and, so, we need to look at this body of case law out there to see what else might--
Mr. Brender: --Well, I think that body of case law can he helpful.
However, I don't think that the estimates of costs are very accurate.
There are, I think, four cites, none of which are specific at all.
In one of them, this OPM study, I believe, says that this is entirely unreliable data.
Costs would not have to be great if, for instance, one simply hired some expert to simply examine the procedure and validate by content type studies, where they simply decide that certain things in the interview are really relevant, they're looked at and those relevant factors are applicable in the job requirements, and then design the application or the interview form so that it goes to those behaviors, and that's not terribly costly to do one of those type studies.
Once again, those costs only have to be incurred after there's an impact, after there's an adverse impact shown.
The studies that we've performed, the statistical studies, showed that, for instance, in the assignment and in the promotion, blacks, for instance, average fourteen months in a particular pay grade as opposed to sixteen months for whites, that blacks averaged about.18 pay grades, increase in pay grade per year, as opposed to.53 per year, and in order to test for things like education and tenure on the job and experience on the job, we did a study that included those of multiple regression, and that showed that over the four-year period of time, blacks were 1.22 to 1.38 pay grades behind comparably-situated whites.
Unidentified Justice: --May I ask what... we're talking about individual claims here, not class claims, aren't we?
Mr. Brender: That's correct.
Unidentified Justice: Does that mean that every individual... say we accept your theory, does that mean that every individual who is a minority person who sought promotion or sought a higher wage during the period in issue would be entitled to recover?
Mr. Brender: I don't believe so, Justice Stevens.
Unidentified Justice: How do you screen out those that do win and those that don't win?
Mr. Brender: Well, what... as I understand it, I think this is what the Court said in Connecticut v. Teale, disparate impact is simply inference.
You're taking the effect on the group and from that, by being a member of the group, you're inferring that you were so injured and that's how you establish liability.
Unidentified Justice: Well, but they're all injured if they all applied.
Mr. Brender: I agree, but if it's not a class action, I don't believe, unless they came in and applied within the 180 days and fulfilled the other agreements, that they would be eligible for any relief.
Unidentified Justice: It seems to me each one would be equally entitled to relief under your theory because there's an absence of intent or any... I may not quite get the understanding here.
Mr. Brender: I think that if they were in the case, but what I'm saying is the statistical proof would not put them in the case.
Unidentified Justice: Why not, if they're a member of the group to which the statistics apply?
Of course, you have to file a lawsuit, I realize that, but I would think all lawsuits would become fungible by members of the class.
Mr. Brender: Well, if there was a class action, they could certainly come in and show they were members of the class.
Unidentified Justice: Right.
That's easy if you've got a class action.
We don't... we have a case where, by hypothesis, we have no class action and no invidious intent.
We have statistics to show that every member of a particular class was not promoted.
There are thirty people in the class and there are five jobs.
They would all prevail.
Mr. Brender: Well, I don't think the proof necessarily says they would all prevail because they couldn't all get the one job.
It just simply says--
Unidentified Justice: How do you decide which would have gotten it?
I don't understand.
Mr. Brender: --Well, all you're showing is that the decision-making was based on an impermissible racial component.
That's what the statistics are saying.
They're not saying that everyone--
Unidentified Justice: I thought they were showing that the permissible factors that were used involving discretion had the unfortunate consequence, unintended, of ending up with either getting many more males than you should have or many more females than you should have.
Mr. Brender: --That's correct.
Unidentified Justice: And whichever way it goes, it seems to me the employer--
Mr. Brender: That wouldn't mean every applicant would then be entitled to the job.
Unidentified Justice: --But how do you decide which ones are and which are not?
Mr. Brender: I think--
Unidentified Justice: Each case would have discretionary decision.
Mr. Brender: --That's right, but I think that applies in class actions, too.
I mean, the court then has to decide whether a person was actually injured or not, and usually they do that sometimes on the percentage allocation of resources.
I see what you mean.
If you have like three or four people who are all complaining about the same--
Unidentified Justice: Well, very frankly, what's running through my mind, I hire a lot of law clerks, and I do it on a discretionary basis, and I suppose statistically I either discriminate against men or women, just to take a simple case, and say all the male applicants who applied and I turned down come in and say, well, statistically, you've got a disproportionate number of women clerks, and I say I guess I have, but I just kind of pick them as they come.
Which law clerks will prevail and which won't?
Mr. Brender: --Well, if you were shown to have a disparate impact because you greatly were favoring one group over another in your hiring practices, and then the question is what individual out of that group would then prevail--
Unidentified Justice: --And there are a lot of them who are qualified that were turned down.
Mr. Brender: --Well, the only one... okay.
Only one person could get the job, let's assume.
Unidentified Justice: Right.
Mr. Brender: If that's the case, then the Court is simply... it's a matter of damages, as I see it.
Unidentified Justice: Well, why?
Why is it any different from the ordinary case?
Let's say you have an employer who will not hire blacks.
He turns down five blacks for a single job and hires a white individual for it.
Wouldn't each one of those five blacks have a cause of action?
Mr. Brender: --They would have--
Unidentified Justice: Do you think just one would?
Mr. Brender: --I think... yeah.
I think they would have the cause of action.
I don't think they would have the relief, is what I'm saying.
Unidentified Justice: What good is the cause of action without relief?
What do you mean?
They have a cause of action for what?
Mr. Brender: By that, I don't mean they would not have total relief.
They might have... let's say that they had a back wage award which would be determined by the fact that only one of them would have gotten the job, so that let's just say that they were disadvantaged by one year's worth of half wages, and you had two people who were equally disadvantaged by the process, then they would have to... the way it's commonly done in class actions is that they would simply split that back wage award.
Unidentified Justice: Anyway, I suppose your answer to all this is if this is a problem, it's not a problem you're creating, this must be a problem under Griggs, too.
Mr. Brender: Well, I think it would be a problem under intentional discrimination, too.
I mean, it wouldn't be specific to impact analysis, it would be the same thing if it was disparate treatment analysis because you could show the same thing statistically.
You know, it's a problem of relief and I've dealt with it in class actions, not on individual contexts, and that's usually the way it's been done in class context, but I don't think it would have anything to do, you know, relative to the matter of proof because all we're talking about is an inference of discrimination, liability, in other words, and then the damage situation would be another portion.
I can see that in terms of the injunctive relief might be difficult if you have one position and two people who have filed a claim, which can happen outside of the class action, and/or there's just a limited number of applicants that have been determined to be eligible for those said positions.
But, once again, I don't think that's particular to this particular method of proof.
I'd like to at this time reserve my time for rebuttal.
Chief Justice William H. Rehnquist: Thank you, Mr. Brender.
We'll hear now from you, Mr. McGee.
ORAL ARGUMENT OF BRUCE W. McGEE, ESQ. ON BEHALF OF THE RESPONDENT
Mr. McGee: Thank you, Mr. Chief Justice, and may it please the Court:
Much of Petitioner's account of this case today is irrelevant and perhaps, to some extent, even inaccurate.
What is clear and what is important for this Court's consideration is that the judgment of the Court of Appeals found that my client was not a discriminator.
That is to say that the Court of Appeals... that the Petitioner did not prevail in its disparate treatment claim against the bank.
It's also clear that--
Unidentified Justice: Did not prevail on discrimination.
Mr. McGee: --That is correct, Justice White.
It is also clear that the judgment of the Court of Appeals found that this was the judgment of the trial court as well.
The 5th Circuit did--
Unidentified Justice: Mr. McGee, did the trial court consider the statistical evidence in the disparate treatment claim?
Mr. McGee: --Yes, Justice O'Connor, it did.
Unidentified Justice: And held that it didn't raise any inference?
Mr. McGee: --They considered it both in whether or not it raised the prima facie case under McDonnell Douglas and then whether or not it had been used to show, by the Petitioner to show that the articulated reasons by the bank were a pretext and they found that the District Court considered those fully or at least there was no suggestion that it did not, and concluded that the statistics had been given full weight by the trial court.
The difference in the two types of causes of action or methods of proof that the Petitioner seeks today are quite clear and have a dramatic effect on this case.
On the one hand, the case was tried under the disparate treatment method, which is the McDonnell Douglas, Furnco, Burdine method of proof.
The court, applying 5th Circuit precedent, applied those tests and determined that the Petitioner had made a prima facie case of her claim of racial discrimination, but that the bank had articulated a reasonable legitimate excuse, a reason for its actions--
Unidentified Justice: What was that?
Mr. McGee: --They described four different reasons for the non-selection.
With the case of the first selection, the court determined that the bank chose Mr. Burt over Ms. Watson because of considerations of his experience, his previous supervisory experience, his close working relationship with Mr. Shipp, who was the bank's operation officer at that time.
Also, his greater or better qualifications than Ms. Watson.
With regard to the second decision, that being the one where Ms. Cullar was chosen as the motor bank supervisor, the same types of considerations were made.
The court concluded that Mr. Burt had a greater opinion of Ms. Cullar's qualifications than those of Ms. Watson and that those reasons were not pretext for discrimination.
Likewise, with the other two selections, the same type of considerations were made and the same type of determinations was made.
The qualifications were not identical in each case, but they were similar.
They were all these discretionary intuitive-type of judgments that this person is the best qualified for this particular job.
That highlights the distinction between the disparate impact test which would require no proof of discriminatory treatment, no proof of intent or motivation, but would rather require only that the Petitioner here show that there was a disproportionate representation of blacks in the bank's work force.
Unidentified Justice: Well, Mr. McGee, it seems to me the initial proof would be the same under the statistics, under either theory.
Mr. McGee: The initial proof would--
Unidentified Justice: The initial proof by the Plaintiff and apparently there was enough to make out a prima facie case of disparate treatment here.
Mr. McGee: --That is correct, Justice O'Connor.
Unidentified Justice: And I assume the exact same proof would be offered and considered under disparate impact.
Mr. McGee: I know of nothing that could be offered at trial that would be any different.
This case was tried over four days.
Unidentified Justice: Right.
So, it just turns on what the Defendant's burden is after that?
Mr. McGee: --That is correct.
It is a greater burden under the impact theory that's thrust upon the Defendant that would affect this case.
Unidentified Justice: And what do you think the difference is in the Defendant's burden under the two?
Mr. McGee: Well, with this particular case,--
Unidentified Justice: And do you think this Court has spoken to that?
Mr. McGee: --No, I do not.
I don't believe this Court has spoken to it.
However, by the very nature of the impact burden of proof or the burden that's put upon the Defendant, the requirement of validation of his selection process, there is, in this case, a virtually insurmountable burden of proof.
The nature of subjective criteria are such that despite what the APA says--
Unidentified Justice: Well, I don't think it's established that there is a need to prove validation, is there?
I mean, I'm just trying to get out the bottom line of what the Defendant has to show in these cases under the two theories.
Mr. McGee: --As I read Griggs and those that follow, there is a requirement that the Petitioner or that the Plaintiff point out a specific facially-neutral arbitrary-type of standard or requirement that causes the impacts complained of.
That's not been done here.
So, under my theory or my view of this case, the Petitioner would still lose because she has never made her prima facie case, but the trial court could and very well likely would conclude that they had and then would put us to validating our entire selection process without ever having been pointed to the one specific point in that selection process that caused the impact.
I don't think that that validation can be done.
What the APA talks about as validation are really objective-cation of a subjective system, which is quite a long ways from validating the bank's selection system.
The trial court found that the bank was absolutely free of any racial discriminatory intent or motivation.
It selects... it uses and has selected a process which it believes best serves its needs, its legitimate business needs, as an employer, and there was--
Unidentified Justice: Well, not necessarily.
It may be wrong.
I mean, isn't the big difference between validation under the impact test and what you urge be applied here, isn't the big difference that the employer, if he's using an objective test under Griggs, he has to be right?
He has to be not merely... the reason he failed to hire this individual has to be not merely something that has nothing to do with racial discrimination or any other unlawful discrimination, but it has to be correct.
Mr. McGee: --That is correct.
Unidentified Justice: Whereas, under your theory, this bank would be entirely entitled to put the wrong people in the position, so long as they didn't pick the wrong people because of racial animus or sex animus.
Mr. McGee: Obviously, errors can occur and we would submit that we have the right to make those errors.
Unidentified Justice: You're not asserting that you established below or that the Court thought you established below that, in fact, your system's a good one, that you're picking the best people?
All the Court really found is that it may be terrible, but whatever is terrible about it had nothing to do with racial or sexual animus.
Mr. McGee: That's correct.
The system was not on trial at the District Court.
Unidentified Justice: Let me just go on further.
But the question, it seems to me, this raises is there's no racial animus in the objective... say you require a high school diploma.
That's not job-related, but it has the impact of denying a lot of minority applicants jobs.
Well, what if your use of this discretionary factor also has the same impact?
You didn't mean it.
Just like requiring a high school diploma.
You didn't mean to hurt anybody, but, in fact, you did.
If this system, in fact, causes precisely the same harm, why should you not have to have precisely the same defense?
Mr. McGee: --Well, the difference is, Justice Stevens, that with the facially-neutral objective requirements, there is a uniform or automatic disqualification of certain people.
It's applied across the board.
But in our... my client's employment system, each selection, each of these four selections that are involved in this case, were done on an individual basis.
The criteria differed to some extent in each case.
There was no automatic disqualification.
The impact, if you will, is not really shown because of the system itself, but rather the results of things that may be totally unrelated to the subjective system that is being criticized.
Unidentified Justice: What if it goes on for ten years and you just get hundreds and hundreds of black applicants turned down and you never hire anybody but white; would you say the same thing?
Mr. McGee: This Court has provided an adequate method for challenging that type of result.
The pattern and practice tests that are referred to in McDonnell Douglas and Justice Powell's opinion make clear that an individual can show that the employer's procedure or practices in carrying on its business are evidence of racial intent itself, and if they are sufficiently great of the type you described, I feel certain that the Plaintiff would be able to obtain a finding that, in fact, any reasons offered by the Defendant were, in fact, a pretext for his racially-discriminatory intention.
If my client is found wrong here today, even though he's not... he's been found not to have any racial intention or any discriminatory intention or motivation, the result will not be that be... that it switches to an objective system for making its promotional decisions.
It will, instead, not go to objective tests because we don't believe that any tests exist today which will give my client the type of decisions they choose and we don't believe that they can be formulated or devised in any time in the foreseeable future at a cost that would be acceptable to my client.
But what my client will continue to do is to make the subjective decisions that it has in the past because it believes this is the best and most fair way to satisfy its legitimate business need to supply itself with employers and supervisors, rather, who are the best qualified.
But what it will be forced to do to protect itself in the future is to overlay on that subjective system a numerical quota system, and I do not believe that this is what the Congress intended when it enacted Title VII.
The subjective system that is challenged sounds, on the basis of the briefs on the Petitioner's side, to be maybe a little bit sloppy and somewhat inferior, and particularly when you have thrown up on the other side the concept of objective tests, which have the sound of being scientific and fair, but, really, the decisions that are being made at my client's place of business are the types of decisions that are made all over this country by employers of all sizes, particularly those that are smaller, like my client, but all employers.
The... no one would suggest, I don't believe, that the employer would prefer to continue to have the right to make his judgments on promotion based upon his view of what is most important to him, who best typifies those criteria and those traits which he thinks are most important to his business decisions.
The experience that the employer has with dealing with these people who are applicants for promotion puts him in the best position to determine who has the characteristics, such as leadership or ability to relate to other people or the respect of the co-workers, the honesty, integrity, those types of subjective discretionary criteria that were employed here.
Unidentified Justice: Mr. McGee, I'm not sure that validation of an objective test is a very simple operation either.
I mean, I've heard argument in other cases bemoening the difficulty of validating an objective test and what many employers in that situation simply say is just what you've said, that the way to solve the problem is simply to make sure that whatever kind of an objective test you apply, it ends up with a work force that is more or less balanced by race, by sex, and so forth.
Now, what's the reason for treating your employer differently just because he's using a subjective test?
I mean, I gather that we've made the decision that if you use an objective test, you not only have to be free of discrimination, you have to be right.
You have to be using tests that really do work to get the best people.
Why should... morally, why should we have a different test for subjective employers?
Mr. McGee: I think because while validation of objective testing devices may be hard, it has been demonstrated that it can be done.
There has absolutely been no demonstration even though many years of efforts and studies have been done, there's been no indication that subjective criteria can be validated.
I believe that the result will be that a subjective system, if challenged under the theory proposed by the Petitioner, will result in an automatic finding of liability if there's an imbalance.
Unidentified Justice: The fact is that it's doable for an objective test, but the cost of doing it is usually more than the cost of simply adopting a relatively balanced work force, and it's my impression that that's the way most employers meet the problem, and if that's the case, then I see no difference between their situation and your client's.
Mr. McGee: Well, I am not familiar with what other employers do and whether that is how they respond to the question, but there's no question in my mind that that's what I would have to recommend to my client.
The numerical quota system is the only appropriate response to a finding of the type Petitioner seeks.
My problem with that, Justice Scalia, is that that is not the intention of Congress, at least not the way I read Title VII and the history of Title VII.
It's just the contrary.
It places the decision of who is promoted, instead of being free of racial consideration, race becomes the paramount consideration.
To me, that's just an absolute abomination of the purpose of Title VII.
The cases of this Court have uniformly, I believe, recognized a distinction between the objective types of criteria as opposed to the subjective criteria that we're considering here today.
The Griggs case, of course, dealt with the objective test of the high school education requirement, and also the standardized testing.
That type of objective standard was easily identifiable.
The causation could be pointed to and the impact shown.
Those type of objective standards have been found in other cases, such as Dothard, involving the minimum height and weight requirement, the New York Transit Authority case, which dealt with the exclusionary rule that disqualified persons who had a history of drug use, Teale that dealt with a written test, and Albemarle that dealt with written tests.
But opposed to those type of cases are the McDonnell Douglas, the Furnco, the Aikens and the Burdine cases, which recognize a different set of proof and a different set of facts.
The situation in McDonnell Douglas was an individual who complained that he was not hired because of his participation... was not hired because of his race.
The reason articulated by the employer was that he was not hired or not rehired because of his participation in illegal activities.
The Court held and pointed out the distinction between that type of an articulated reason that is subjective and inherently unique to the employer as opposed to those which were present in Griggs which had the effect of disqualifying large groups of people automatically, and I think that's the key to the distinction between McDonnell Douglas and Griggs, is that, on the one hand, you have these objective artificial rules which disqualify large groups of people, regardless of consideration of their individual qualifications, where, with McDonnell Douglas, you have a consideration of qualifications.
Here, in my case, the bank recognized that Clara Watson was fully qualified for the job she sought or at least possessed the minimum qualifications as well as did others.
Some of the selections that were involved here today, there were four or five other people considered, including in that group was at least one other black person.
The bank considered the qualifications of each of those people on an individual basis and reached the conclusion that the person selected was the best qualified and, therefore, the others were less qualified.
But that's the type of selection process that was referred to in the Johnson v. Transportation Authority opinion from last term.
The purpose of Title VII impact cases is to remove these arbitrary barriers which prevent large groups of minorities and women from being qualified.
But once they've been qualified, then the decisions as to whether they were chosen or not chosen should be considered under disparate treatment theory.
If there is a pattern and practice of racial discrimination or discrimination on other prohibited bases, then that will come out as a part of the disparate treatment proof, and if it is proved by the statistical evidence, it can be dealt with and can be the basis for finding that the defendant is guilty of discriminatory intent.
Unidentified Justice: What is the EEOC's view?
Mr. McGee: The EEOC views the case exactly the same way I do.
They have filed a brief before this Court which supports--
Unidentified Justice: Have they always had that view?
Mr. McGee: --No, Justice White.
I think their view is best described as having been inconsistent.
They have had varying views over the history of Title VII.
Unidentified Justice: Well, what was its view immediately before the present view?
Mr. McGee: Let me see if I can go back in time.
Unidentified Justice: Well, it shouldn't be too long.
Mr. McGee: --In 1978, when the uniform guidelines were adopted, they had the view that subjective selection processes, such as present here, might not be verifiable or validateable and, therefore, need only be validated under the federal law, which comments to the uniform guidelines suggest would be under the disparate treatment theory as announced in McDonnell Douglas.
Prior to that, there was the 1970 guidelines, had some rather harsh validation requirements on all forms of selection processes.
Unidentified Justice: Including subjective?
Mr. McGee: Objective and subjective.
And the '78 guidelines were seen as a response to the complaints of the APA and other professionals that they were basically unachievable goals and objectives and put the defendant, the employer, in the position of basically being out of compliance automatically, and that is the reason for the '78 guidelines and why the EEOC supports our position here today.
The... one other thing that I'd like to point out in response to a question to Mr. Brender, the numbers in this case do not make disparate impact an appropriate method of proof.
There are only two black employees who sought promotions, Clara Watson and one other employee.
Those numbers are entirely too small to make any kind of determination of statistical impact or statistical significance, and--
Unidentified Justice: By your own figures, how many employees are there involved in the whole bank?
Mr. McGee: --How many employees are involved in the whole bank?
Unidentified Justice: Yes.
Mr. McGee: There are approximately eighty employees.
Unidentified Justice: And how many are those are Negroes?
Mr. McGee: Approximately nine or ten.
Unidentified Justice: And one's a cook and one's a janitor.
What are the others?
Mr. McGee: There are no employees that I'm aware of in those positions any longer, Justice Marshall.
Unidentified Justice: I thought one Negro was in the cafeteria.
Mr. McGee: The bank does not have a cafeteria at this time nor did it have at the time of the trial.
They did have a--
Unidentified Justice: What jobs do the nine have?
Mr. McGee: --There is one employee, I believe, that has the position that you're referring to, it's not a... they do not have a cafeteria, but there's someone in a position that's similar to what you have described.
The other positions are all employees of the bank in either operations or in the teller's--
Unidentified Justice: Like what?
How many are on the floor of the bank?
Mr. McGee: --There are approximately four.
Fifty percent of those people are on the floor of the bank.
Unidentified Justice: Four?
Mr. McGee: Yes.
Unidentified Justice: Out of how many?
Mr. McGee: Out of eight.
Unidentified Justice: So, only half of them are Negroes.
Mr. McGee: That's correct.
How many Negroes are on the--
Unidentified Justice: That was what I asked.
Mr. McGee: --Yes.
Half of those people are operating in positions similar to the one Ms. Watson's in.
Unidentified Justice: You mean they're tellers?
Mr. McGee: Some are tellers, some are customer service personnel, but they all function in the same area of the bank.
Unidentified Justice: You know, one time I had a case where they reported that they had a Negro employed as a telephonic expert.
They were talking about a telephone operator.
I mean, I didn't want to get in that category.
Mr. McGee: I don't think that's what we have here.
I think all of the positions, including the position that you referred to, are all legitimate important positions to the bank, including, at the time of trial, one black employee who was serving as an assistant supervisor in the motor bank facility.
This is the same employee who had sought some of the earlier promotions that Ms. Watson sought.
The subjective selections that my client uses are not inherently bad.
They have just been thrust in this position by the Petitioner.
They are no different than the type of selection process that Justice Stevens referred to with his law clerks.
If, for instance, Justice Stevens determined that he chose to retain one of his law clerks for a second term, and he wanted to choose among the law clerks he had which one that he would retain for that second term, he's going to necessarily rely upon his intuitive subjective judgment as to which one is the best qualified, will most fulfill his needs as a law clerk, and I think it's just ludicrous to suggest that Justice Stevens should have to hire an industrial psychologist to devise some form of test so that his decisions can be validated so as he can avoid a claim of discriminatory impact.
Unidentified Justice: Actually, that's a bad example.
That would be easy to validate.
You would pick the masochist, I think.
I was going to say if I did that, maybe I'd have better law clerks.
Mr. McGee: That's important.
Unidentified Justice: We won't quote you.
Mr. McGee: McDonnell Douglas makes clear and Burdine makes clear that the law does not require that the employer have the best or the most balanced work force.
It's only that he make those decisions free of discriminatory intent and animus, and that is what the trial court found and is what the Court of Appeals has affirmed.
They found that my client was not a discriminator, and we suggest that the opinion of the--
Unidentified Justice: But when you say he was not a discriminator, what you mean is he did not intentionally discriminate?
The problem the case really raises is a very difficult one.
There are a lot of people who don't think they're discriminators but who maybe really are without knowing it, and that's what the... that's the troublesome problem this case raises.
Mr. McGee: --Well, I agree that that is the problem with this case, but this case differs from those, such as Griggs, where you're dealing with non-intent.
By definition, these judgments and selections that my client made were intentional decisions.
They made a judgment to choose this person or to not choose that person.
Those are intent-based decisions by definition and should be properly analyzed under disparate treatment tests.
We would urge that the opinion of the court below be affirmed.
Chief Justice William H. Rehnquist: Thank you, Mr. McGee.
Mr. Brender, you have two minutes remaining.
ORAL ARGUMENT OF ART BRENDER, ESQ. ON BEHALF OF THE PETITIONER -- REBUTTAL
Mr. Brender: Thank you, Mr. Chief Justice.
Justice White, the EEOC position has changed much more recently because, in 1986, in the 8th Circuit, in EEOC v. Rath Packing Company, the EEOC prosecuted the Rath Packing Company for subjective hiring and promotion practices under the disparate impact theory and prevailed.
The court so found.
Their guidelines for eighteen years have required that subjective practices be treated under the disparate impact theory and the questions and answers that are supposed to guide employers in applying the guidelines state that, too, and we have that in our reply brief.
There's no difference between the provisions that the Respondent cites for subjective procedures and the ones for objective procedures.
Just like in the case studies for costs, there is no difference between objective... the cost for objective and subjective validation.
They just don't make that distinction.
Unidentified Justice: Well, the Solicitor General, speaking on behalf of the EEOC and the United States, said that the subjective methods must be justified, need only be shown that the selection procedure is reasonably related to the requirements of the job or other business needs, and has been applied in a non-pretextural fashion.
That's what its brief says.
Mr. Brender: I know.
He kind of combined--
Unidentified Justice: Let me just ask you.
Assume that that is the right test, assume that that is the right test, was that test satisfied in this case?
Was that kind of a justification offered?
Mr. Brender: --No.
Well, the problem is--
Unidentified Justice: Or found or applied by--
Mr. Brender: --No.
They provided no justification whatsoever for their practices.
I mean, they just said we have them, we try to select the best person.
Of course, the problem is the best person was always white.
There were many more decisions, by the way, in our statistics.
We took into account every increase in pay grade.
So, that was--
Unidentified Justice: --What I'm trying to find out, if that's the test... was that the test that was applied in this case by the courts below or not?
Mr. Brender: --No.
He has kind of a hybrid test there where he uses disparate impact analysis not to justify the particular decision, but to justify the particular... they don't address what the burden of proof is going to be on the employer in doing that, and I don't find any support in any case law for that test.
It wasn't applied in this case.
Chief Justice William H. Rehnquist: Thank you, Mr. Brender.
The case is submitted.
Mr. Brender: Thank you, Your Honor.
Chief Justice William H. Rehnquist: We'll hear argument next in Number 87-5002, Ellis T. McCoy v. The Court of Appeals of Wisconsin.