(dissenting).
This case seems to me unlike most cases resulting in a divided panel. Instead of a head-on clash of arguments (there is/is not probable cause, sufficient evidence, due process, etc.), the difference between my brothers and me lies in our choice of a starting point. I insist that, because there has been no discrimination against women in the process of filling the positions at issue here, there is no prima facie case to begin with. My brothers’ position has to be that although women have not been discriminated against, the use of a criterion of choice — height—which in other circumstances can be a means of discriminating against women, is sufficient to make out a prima facie case here. Because this seems to me to provide a remedy when there has been no wrong and to carry abstract disparate impact analysis far beyond that articulated in Connecticut v. Teal, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982), I dissent.
At the outset, I must assume some responsibility for what I consider to be current confusion as to the relationship between the situation in this case and that in Connecticut v. Teal. In our first opinion, (Costa I), I described Teal as the “mirror image of this case and specifically declined to follow the Teal approach. In part because of that language I agreed that in light of the Supreme Court’s affirmance of Teal we should rehear this case and make sure that the two cases differ significantly. I think they do.
The plaintiff in this case had urged us to accept the Second Circuit’s opinion in Teal as guidance for our resolution of her claim. We could (and perhaps should) have dismissed Teal as simply irrelevant to a situation where the challenged barrier never had any disproportionate effect on women. Instead, we described the cases as “mirror images”, a description which has caused no small amount of confusion and which I feel some obligation to attempt to clarify. In describing Teal as the mirror image of this case, I meant to highlight the differences rather than the similarities between the two cases. Both cases involved a pass-fail hiring barrier that statistics demonstrated had a greater impact on members of the protected class than on non-members. Both also involved an additional aspect of the hiring process that operated to remove the impact of that barrier. But there the similarity ends. They are mirror images and not identical situations because the aspects of the process that operated to remove the overall impact of the barrier were at differ*9ent ends of the process. In Costa there was no discrimination to begin with; in Teal the initial discrimination was compensated for at the end. That timing is a crucial distinction between the two cases because, as we pointed out in Costa I, the initial decision in this case to hire only women meant that the pass-fail barrier never had any disparate effect on women. There was, therefore, no question of remedying the discriminatory effect of an element in a process with a non-discriminatory bottom line. There was never any discriminatory effect.1
As I understand the debate between the majority and the dissent in Teal, the issue was whether an employer should be excused for utilizing a barrier that statistics showed disproportionately affected blacks as long as it added enough blacks at the end of the hiring process to meet the appropriate racial ratio. The majority concluded that a non-discriminatory “bottom line” was no justification for discriminating against individuals along the way. They found it “clear that Congress never intended to give an employer license to discriminate against some employees on the basis of race or sex merely because he favorably treats other members of the employees’ group.” 457 U.S. at 455, 102 S.Ct. at 2535. In Justice Powell’s view, the result reached by the majority is an unprecedented expansion of the theory of disparate impact. As he points out, Title VII plaintiffs have been allowed to proceed under either a theory of disparate treatment — by establishing that they, as individuals, have been discriminated against — or of disparate impact — by establishing that they have been victims of a facially neutral practice that excludes a disproportionate number of members of his protected group. Because the latter theory necessarily requires reference to the group rather than to the individual, Justice Powell would not allow a plaintiff proceeding under this theory to “have it both ways”, to infer discrimination by reference to the group and then ignore the fact that the group as a whole has not been injured. In his view, this is “to confuse the individualistic aim of Title VII with the methods of proof by which Title VII rights may be vindicated.” Id. at 460, 102 S.Ct. at 2537 (Powell, J., dissenting).
Whatever the merits of the dispute in Teal, neither side addressed the issue raised by plaintiff here. My brothers, I submit, have made a leap beyond the position taken by the majority and criticized by the dissent in Teal. For, as far as I can tell, the Supreme Court has never held, or even intimated, that a Title VII plaintiff should be able to recover for discrimination on the basis of her sex when it is impossible that she was disadvantaged in relation to any man. We held in Costa I, and my brothers still agree, that at the only point in time when the height requirement had any effect on plaintiff, she was in competition only with other women for a job as a police officer.2 As we pointed out in Costa I, courts that have addressed similar sitúa-, tions agree that when there is no competition between the sexes, there can be no discrimination on the basis of sex. See Stroud v. Delta Air Lines, Inc., 544 F.2d 892 *10(5th Cir.1977); Gerdom v. Continental Airlines, Inc., 648 F.2d 1223 (9th Cir.1981).
The majority reasons that “[wjere we to focus in this case on the city’s hiring of only women, we would be focusing on the ‘bottom line’ ”, a focus forbidden by Teal. To the contrary, I submit that in this case, the city’s hiring of only women is the introductory paragraph. It is the key to everything else that follows. According to Teal, once a disparate effect is demonstrated, no amount of undoing of that effect will suffice. In this case, there is no effect, so we need not worry about the employer’s “bottom line”.
My dispute with my brothers, I believe, comes down to what is required for a plaintiff to demonstrate disparate impact. I have no question that in another context, where the sexes are in competition, the statistics provided here, showing that as of 1971-74 the average height of women between the ages of 25 and 34 was 5 feet, 4.1 inches while that for men in the same age bracket was 5 feet, 9.6 inches, would establish a prima facie case of disparate impact and require an employer to demonstrate that his height requirement was job related. My point in Costa I and the point which I reiterate here is that the fact that a hiring barrier excludes more women than men is irrelevant when the women cannot lose out to men. When only women are competing for a job as a police officer, a height requirement may be unfair but it is not a violation of Title VII.
OPINION EN BANC
COFFIN, Circuit Judge.This case has a long history. On May 3, 1982, we held that plaintiff could not prevail in her claim of employment discrimination based on a 5 feet 6 inches height requirement. Although the height requirement would, in other contexts, have a disparate impact on women, we determined that the height requirement could have had no such impact in this case because only women were in competition for the job. Costa v. Markey (Costa I), 677 F.2d 158 (1st Cir. 1982).
Subsequently, we agreed to rehear the case in light of the Supreme Court’s decision in Connecticut v. Teal,-U.S.-, 102 S.Ct. 2525, 73 L.Ed.2d 130 (U.S.1982). After reviewing supplemental briefs and hearing oral argument, a majority of the panel determined that Teal required that we reverse our original position and allow plaintiff to recover based on her evidence that the height requirement would exclude more women than men, regardless of the fact that in this case the only job applicants were women. Costa v. Markey (Costa II), 706 F.2d 1 (1st Cir.1982).
On March 9, 1983, we granted rehearing en banc. Counsel were requested to file supplemental memoranda addressing the question whether
“this court [should] view the relevant job here at issue (1) as a single job open only to women or (2) as one of a set of police force jobs in general open to both men and women? If the latter, are not men and women in competition for the same jobs?”
For reasons set out below, a majority of the court now agrees that Teal does not require that we reverse our original determination that plaintiff has not made out a prima facie case of disparate impact. Nor are we persuaded by any other factors regarding the hiring process at issue here that plaintiff was discriminated against on the basis of her sex. Accordingly, we reaffirm the judgment we rendered on May 3, 1982, denying plaintiff recovery.
I.
Plaintiff’s initial argument before us relied in part on the Second Circuit’s opinion in Teal v. Connecticut, 645 F.2d 133 (2d Cir.1981), aff’d, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982), that Title VII plaintiffs may recover if they are disadvantaged by one component of a selection process that has a disparate impact on a protected group of which they are members, even if their employer “remedies” the disparate impact by subsequent affirmative measures. Since then, the Supreme Court has affirmed the Teal approach, holding *11that a nondiscriminatory “bottom line” is not a defense to a prima facie case of employment discrimination.
In Costa I, we specifically declined to follow the Teal approach. In a position similar to that of Justice Powell, dissenting in Teal, we concluded that if the overall effect of a hiring process is not discriminatory, we saw no reason to look behind that result to possibly discriminatory elements of the process.1 We no longer have the liberty of offering that broad justification for our result in this case, but we can still point out that Teal and this case are distinguishable in at least one crucial respect. In Teal, the employer sought to remedy the effects of a discriminatory barrier by affirmatively hiring members of the disadvantaged minority group. In this case, although plaintiff was discriminated against on the basis of her height, there was never any discrimination of the type for which Title VII provides a remedy. Because the sexes were not in competition, the height requirement had no disparate effect on women. There was, therefore, no question of remedying the discriminatory effect of an element in a process with a nondiscriminatory “bottom line”.
A brief review of the facts should make the absence of discriminatory effect clear. The failure to hire at issue here arose from an effort in 1974 by the City of New Bed-ford, Massachusetts (city) to hire two female police officers. The city makes appointments for police officers from a list of eligible applicants that is certified to it by the state civil service personnel division. Applicants’ names appear on the list in order of their scores on a written exam, adjusted for various statutory and court-ordered preferences. Ranking on the list is not affected by an applicant’s height.
Prior to February 1974, the city had two distinct police categories: males were police officers and females were police women. Male applicants were ineligible for positions as police officers if they failed to meet a minimum height requirement of 5 feet 6 inches. In February, 1974, the city abandoned these separate job categories, and thereafter both men and women competed for positions as police officers. Women applicants were also required to meet the 5 feet 6 inches height requirement.
In the spring and summer of 1974, two female police officers retired. Because the city needed female officers to perform certain duties, related, for example, to female prisoners, and because there were no women high enough on the integrated list to be hired, the city sought and received approval from the Massachusetts Commission Against Discrimination to hire from a list of exclusively female applicants. The most recent integrated list was therefore separated and the city hired the top two women on the women-only list who also met the 5 feet 6 inches height requirement. Although plaintiff’s name appeared first on the list, she was rejected because she did not meet the height requirement.
Plaintiff brought suit under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e-2, asserting that application of the height requirement was discriminatory because it had a disparate impact on women. The district court agreed with plaintiff and granted her back pay and other relief.2 We *12reversed, concluding that if the height requirement was applied to women only, it could have had no disparate effect on women. Plaintiff, therefore, had failed to establish a prima facie case of employment discrimination.
After lengthy consideration, we are satisfied that our original reasoning was sound. Plaintiff simply failed to make it over the initial hurdle of demonstrating that the challenged barrier had a discriminatory effect on women. Absent a discriminatory effect, there is no need to engage in further inquiry regarding the employer’s non-discriminatory “bottom line”.
Another plaintiff — perhaps one denied employment because of height a year after initiation of the separate lists — might be able to argue that because the job sought was that of a “police officer”, the duties of which were basically the same for men and women, the sexes were in competition and the height requirement did have a disparate impact on women. In such a case, the city’s height requirement might operate to exclude more women than men. In this case, however, the evidence is uncontroverted that competition between men and women was at least temporarily suspended so that the city could hire two needed female police officers. Without the height requirement, the city would still have hired only two women. The only difference is that plaintiff would have been one of them.
We are compelled to decide the case on the facts before us. On these facts, the height requirement cannot be viewed as having a disparate effect on women. In the absence of a discriminatory effect, Teal simply does not apply. See Teal, 457 U.S. at 452, 102 S.Ct. at 2534 (“[a] non-job-related test that has a disparate racial impact, and is used to ‘limit’ or ‘classify’ employees, is ‘used to discriminate’ within the meaning of Title VII whether or not it was ‘designed or intended’ to have this effect and despite an employer’s efforts to compensate for its discriminatory effect.”) (Emphasis added.)
II.
Besides relying on Teal, plaintiff has an additional basis for challenging our decision in Costa I. In Costa I, we relied on a line of cases in which courts have refused, in the absence of competition between the sexes, to find that a facially neutral employment barrier discriminated against women on the basis of sex. See, e.g., Stroud v. Delta Airlines, Inc., 544 F.2d 892 (5th Cir.1977). Plaintiff suggests that subsequent authority has cast doubt on the reasoning of those courts. Her argument, however, is unpersuasive. Neither Gerdom v. Continental Airlines, Inc., 692 F.2d 602 (9th Cir.1982), nor County of Washington v. Gunther, 452 U.S. 161, 101 S.Ct. 2242, 68 L.Ed.2d 751 (1981), found disparate impact where the sexes were not in competition. Both cases dealt with disparate treatment, the proof of which is not based on the effect on two competing groups of a facially neutral barrier, but on intentional discrimination against an individual because of his or her race or sex. We remain unpersuaded that a hiring barrier can have a disparate impact on a protected group when competition for the job is only among members of that protected group.
We reaffirm our original decision of May 3, 1982. The part of the district court’s judgment relating to denial of employment from the all-women list of August 12, 1974 is reversed.
. In Costa I, we also described Teal as standing for an approach to reviewing disparate impact discrimination claims that we rejected. In a position similar to that of Justice Powell, dissenting in Teal, we concluded that if the overall effect of a hiring process is not discriminatory, we saw no reason to look behind that result to possibly discriminatory elements of the process. We no longer have the liberty of offering that broad justification for our result in this case, but we can still point out that Teal and this case are different in the crucial respect that the challenged hiring barrier in one had a disparate effect on women and that in the other did not. Nothing in Teal requires us to find a discriminatory effect where there is none.
. It has not been argued and there is no evidence that the relative numbers of men and women hired would have varied according to the number or quality of the eligible women applicants. The City requested permission to hire from a list of women because of its need for female police officers and the fact that, wholly apart from the height requirement, there were no women high enough on the integrated list to be hired. Thus it is doubtful that after the lists were separated the number of women hired would have increased had there been no height requirement.
. We also described Teal and the case before us as “mirror images”, a description which has caused some confusion as this case has evolved. Use of the term was intended to highlight the differences rather than the similarities between the two cases. Both cases involved a pass-fail hiring barrier that statistics demonstrated had a greater impact on members of the protected class than on non-members. Both also involved an additional aspect of the hiring process that operated to remove the impact of that barrier. But there the similarity ends. They are mirror images and not identical situations because the aspects of the process that operated to remove the overall impact of the barrier were at different ends of the process. In Teal the initial discrimination was compensated for at the end; in Costa the separation of the hiring lists meant that there was no discrimination to begin with.
. The district court also agreed that plaintiff had been discriminated against by an earlier application of the height requirement. During the integrated list period, plaintiffs name was incorrectly placed higher on the list than it should have been. Based on this placement, plaintiff was interviewed for a position as a police officer, but was rejected because of her *12height. The district court held that, although plaintiff was deprived only of an “invalid appointment” as a police officer, defendants nonetheless had violated Title VII. The court denied plaintiff back pay, but issued a declaratory judgment in her favor. Since neither party appealed this judgment, we have had no occasion to reconsider it.