concurring in the judgment.
In Steelworkers v. Weber, 443 U. S. 193 (1979), this Court held that § 703(d) of Title VII does not prohibit voluntary affirmative action efforts if the employer sought to remedy a “manifest. . . imbalance] in traditionally segregated job categories.” Id., at 197. As Justice Scalia illuminates with excruciating clarity, § 703 has been interpreted by Weber and succeeding cases to permit what its language read literally would prohibit. Post, at 669-671; see also ante, at 642-643 *648(Stevens, J., concurring). Section 703(d) prohibits employment discrimination “against any individual because of his race, color, religion, sex, or national origin.” 42 U. S. C. §2000e-2(d) (emphasis added). The Weber Court, however, concluded that voluntary affirmative action was permissible in some circumstances because a prohibition of every type of affirmative action would “ ‘bring about an end completely at variance with the purpose of the statute.’” 443 U. S., at 202 (quoting United States v. Public Utilities Comm’n, 345 U. S. 295, 315 (1953)). This purpose, according to the Court, was to open employment opportunities for blacks in occupations that had been traditionally closed to them.
None of the parties in this case have suggested that we overrule Weber and that question was not raised, briefed, or argued in this Court or in the courts below. If the Court is faithful to its normal prudential restraints and to the principle of stare decisis we must address once again the propriety of an affirmative action plan under Title VII in light of our precedents, precedents that have upheld affirmative action in a variety of circumstances. This time the question posed is whether a public employer violates Title VII by promoting a qualified woman rather than a marginally better qualified man when there is a statistical imbalance sufficient to support a claim of a pattern or practice of discrimination against women under Title VII.
I concur in the judgment of the Court in light of our precedents. I write separately, however, because the Court has chosen to follow an expansive and ill-defined approach to voluntary affirmative action by public employers despite the limitations imposed by the Constitution and by the provisions of Title VII, and because Justice Scalia’s dissent rejects the Court’s precedents and addresses the question of how Title VII should be interpreted as if the Court were writing on a clean slate. The former course of action gives insufficient guidance to courts and litigants; the latter course of action serves as a useful point of academic discussion, but fails *649to reckon with the reality of the course that the majority of the Court has determined to follow.
In my view, the proper initial inquiry in evaluating the legality of an affirmative action plan by a public employer under Title VII is no different from that required by the Equal Protection Clause. In either case, consistent with the congressional intent to provide some measure of protection to the interests of the employer’s nonminority employees, the employer must have had a firm basis for believing that remedial action was required. An employer would have such a firm basis if it can point to a statistical disparity sufficient to support a prima facie claim under Title VII by the employee beneficiaries of the affirmative action plan of a pattern or practice claim of discrimination.
In Weber, this Court balanced two conflicting concerns in construing § 703(d): Congress’ intent to root out invidious discrimination against any person on the basis of race or gender, McDonald v. Santa Fe Transportation Co., 427 U. S. 273 (1976), and its goal of eliminating the lasting effects of discrimination against minorities. Given these conflicting concerns, the Court concluded that it would be inconsistent with the background and purpose of Title VII to prohibit affirmative action in all cases. As I read Weber, however, the Court also determined that Congress had balanced these two competing concerns by permitting affirmative action only as a remedial device to eliminate actual or apparent discrimination or the lingering effects of this discrimination.
Contrary to the intimations in Justice Stevens’ concurrence, this Court did not approve preferences for minorities “for any reason that might seem sensible from a business or a social point of view. ” Ante, at 645. Indeed, such an approach would have been wholly at odds with this Court’s holding in McDonald that Congress intended to prohibit practices that operate to discriminate against the employment opportunities of nonminorities as well as minorities. Moreover, in Weber the Court was careful to consider the effects of the af*650firmative action plan for black employees on the employment opportunities of white employees. 443 U. S., at 208. Instead of a wholly standardless approach to affirmative action, the Court determined in Weber that Congress intended to permit affirmative action only if the employer could point to a “manifest . . . imbalance] in traditionally segregated job categories.” Id., at 197. This requirement both “provides assurance . . . that sex or race will be taken into account in a manner consistent with Title VII’s purpose of eliminating the effects of employment discrimination,” ante, at 632, and is consistent with this Court’s and Congress’ consistent emphasis on the value of voluntary efforts to further the antidis-crimination purposes of Title VII. Wygant v. Jackson Board of Education, 476 U. S. 267, 290 (1986) (O’Connor, J., concurring in part and concurring in judgment).
The Weber view of Congress’ resolution of the conflicting concerns of minority and nonminority workers in Title VII appears substantially similar to this Court’s resolution of these same concerns in Wygant v. Jackson Board of Education, supra, which involved the claim that an affirmative action plan by a public employer violated the Equal Protection Clause. In Wygant, the Court was in agreement that remedying past or present racial discrimination by a state actor is a sufficiently weighty interest to warrant the remedial use of a carefully constructed affirmative action plan. The Court also concluded, however, that “[sjocietal discrimination, without more, is too amorphous a basis for imposing a racially classified remedy.” Id., at 276. Instead, we determined that affirmative action was valid if it was crafted to remedy past or present discrimination by the employer. Although the employer need not point to any contemporaneous findings of actual discrimination, I concluded in Wygant that the employer must point to evidence sufficient to establish a firm basis for believing that remedial action is required, and that a statistical imbalance sufficient for a Title VII prima facie *651case against the employer would satisfy this firm basis requirement:
“Public employers are not without reliable benchmarks in making this determination. For example, demonstrable evidence of a disparity between the percentage of qualified blacks on a school’s teaching staff and the percentage of qualified minorities in the relevant labor pool sufficient to support a prima facie Title VII pattern or practice claim by minority teachers would lend a compelling basis for a competent authority such as the School Board to conclude that implementation of a voluntary affirmative action plan is appropriate to remedy apparent prior employment discrimination.” Id., at 292.
The Wygant analysis is entirely consistent with Weber. In Weber, the affirmative action plan involved a training program for unskilled production workers. There was little doubt that the absence of black craftworkers was the result of the exclusion of blacks from craft unions. Steelworkers v. Weber, 443 U. S., at 198, n. 1 (“Judicial findings of exclusion from crafts on racial grounds are so numerous as to make such exclusion a proper subject for judicial notice”). The employer in Weber had previously hired as craftworkers only persons with prior craft experience, and craft unions provided the sole avenue for obtaining this experience. Because the discrimination occurred at entry into the craft union, the “manifest racial imbalance” was powerful evidence of prior race discrimination. Under our case law, the relevant comparison for a Title VII prima facie case in those circumstances — discrimination in admission to entry-level positions such as membership in craft unions — is to the total percentage of blacks in the labor force. See Teamsters v. United States, 431 U. S. 324 (1977); cf. Sheet Metal Workers v. EEOC, 478 U. S. 421, 437-439 (1986) (observing that lower courts had relied on comparison to general labor force in finding Title VII violation by union). Here, however, the evidence of past discrimination is more complex. The num*652ber of women with the qualifications for entry into .the relevant job classification was quite small. A statistical imbalance between the percentage of women in the work force generally and the percentage of women in the particular specialized job classification, therefore, does not suggest past discrimination for purposes of proving a Title VII prima facie case. See Hazelwood School District v. United States, 433 U. S. 299, 308, and n. 13 (1977).
Unfortunately, the Court today gives little guidance for what statistical imbalance is sufficient to support an affirmative action plan. Although the Court denies that the statistical imbalance need be sufficient to make out a prima facie case of discrimination against women, ante, at 632, the Court fails to suggest an alternative standard. Because both Wygant and Weber attempt to reconcile the same competing concerns, I see little justification for the adoption of different standards for affirmative action under Title VII and the Equal Protection Clause.
While employers must have a firm basis for concluding that remedial action is necessary, neither Wygant nor Weber places a burden on employers to prove that they actually discriminated against women or minorities. Employers are “trapped between the competing hazards of liability to minorities if affirmative action is not taken to remedy apparent employment discrimination and liability to nonminorities if affirmative action is taken.” Wygant v. Jackson Board of Education, 476 U. S., at 291 (O’Connor, J., concurring in part and concurring in judgment). Moreover, this Court has long emphasized the importance of voluntary efforts to eliminate discrimination. Id., at 290. Thus, I concluded in Wy-gant that a contemporaneous finding of discrimination should not be required because it would discourage voluntary efforts to remedy apparent discrimination. A requirement that an employer actually prove that it had discriminated in the past would also unduly discourage voluntary efforts to remedy apparent discrimination. As I emphasized in Wygant, a chai-*653lenge to an affirmative action plan “does not automatically impose upon the public employer the burden of convincing the court of its liability for prior unlawful discrimination; nor does it mean that the court must make an actual finding of prior discrimination based on the employer’s proof before the employer’s affirmative action plan will be upheld.” Id., at 292. Evidence sufficient for a prima facie Title VII pattern or practice claim against the employer itself suggests that the absence of women or minorities in a work force cannot be explained by general societal discrimination alone and that remedial action is appropriate.
In applying these principles to this case, it is important to pay close attention to both the affirmative action plan, and the manner in which that plan was applied to the specific promotion decision at issue in this case. In December 1978, the Santa Clara Transit District Board of Supervisors adopted an affirmative action plan for the Santa Clara County Transportation Agency (Agency). At the time the plan was adopted, not one woman was employed in respondents’ 238 skilled craft positions, and the plan recognized that women “are not strongly motivated to seek employment in job classifications where they have not been traditionally employed because of the limited opportunities that have existed in the past for them to work in such classifications.” App. 57. Additionally, the plan stated that respondents “recognize[d] that mere prohibition of discriminatory practices is not enough to remedy the effects of past practices and to permit attainment of an equitable representation of minorities, women and handicapped persons,” id., at 31, and that “the selection and appointment processes are areas where hidden discrimination frequently occurs.” Id., at 71. Thus, respondents had the expectation that the plan “should result in improved personnel practices that will benefit all Agency employees who may have been subjected to discriminatory personnel practices in the past.” Id., at 35.
*654The long-term goal of the plan was “to attain a work force whose composition in all job levels and major job classifications approximates the distribution of women ... in the Santa Clara County work force.” Id., at 54. If this long-term goal had been applied to the hiring decisions made by the Agency, in my view, the affirmative action plan would violate Title VII. “[I]t is completely unrealistic to assume that individuals of each [sex] will gravitate with mathematical exactitude to each employer . . . absent unlawful discrimination.” Sheet Metal Workers, 478 U. S., at 494 (O’Connor, J., concurring in part and dissenting in part). Thus, a goal that makes such an assumption, and simplistically focuses on the proportion of women and minorities in the work force without more, is not remedial. Only a goal that takes into account the number of women and minorities qualified for the relevant position could satisfy the requirement that an affirmative action plan be remedial. This long-range goal, however, was never used as a guide for actual hiring decisions. Instead, the goal was merely a statement of aspiration wholly without operational significance. The affirmative action plan itself recognized the host of reasons why this goal was extremely unrealistic, App. 56-57, and as I read the record, the long-term goal was not applied in the promotion decision challenged in this case. Instead, the plan provided for the development of short-term goals, which alone were to guide respondents, id., at 61, and the plan cautioned that even these goals “should not be construed as ‘quotas’ that must be met.” Id., at 64. Instead, these short-term goals were to be focused on remedying past apparent discrimination, and would “[pjrovide an objective standard for use in determining if the representation of minorities, women and handicapped persons in particular job classifications is at a reasonable level in comparison with estimates of the numbers of persons from these groups in the area work force who can meet the educational and experience requirements for employment.” Id., at 61.
*655At the time of the promotion at issue in this case, the short-term goals had not been fully developed. Nevertheless, the Agency had already recognized that the long-range goal was unrealistic, and had determined that the progress of the Agency should be judged by a comparison to the qualified women in the area work force. As I view the record, the promotion decision in this case was entirely consistent with the philosophy underlying the development of the short-term goals.
The Agency announced a vacancy for the position of road dispatcher in the Agency’s Roads Division on December 12, 1979. Twelve employees applied for this position, including Diane Joyce and petitioner. Nine of these employees were interviewed for the position by a two-person board. Seven applicants — including Joyce and petitioner — scored above 70 on this interview, and were certified as eligible for selection for the promotion. Petitioner scored 75 on the interview, while Joyce scored 73. After a second interview, a committee of three agency employees recommended that petitioner be selected for the promotion to road dispatcher. The County’s Affirmative Action Officer, on the other hand, urged that Joyce be selected for the position.
The ultimate decision to promote Joyce rather than petitioner was made by James Graebner, the Director of the Agency. As Justice Scalia views the record in this case, the Agency Director made the decision to promote Joyce rather than petitioner solely on the basis of sex and with indifference to the relative merits of the two applicants. See post, at 662-663. In my view, however, the record simply fails to substantiate the picture painted by Justice Scalia. The Agency Director testified that he “tried to look at the whole picture, the combination of [Joyce’s] qualifications and Mr. Johnson’s qualifications, their test scores, their experience, their background, affirmative action matters, things like that.” Tr. 68. Contrary to Justice Scalia’s suggestion, post, at 663, the Agency Director knew far more than *656merely the sex of the candidates and that they appeared on a list of candidates eligible for the job. The Director had spoken to individuals familiar with the qualifications of both applicants for the promotion, and was aware that their scores were rather close. Moreover, he testified that over a period of weeks he had spent several hours making the promotion decision, suggesting that Joyce was not selected solely on the basis of her sex. Tr. 68. Additionally, the Director stated that had Joyce’s experience been less than that of petitioner by a larger margin, petitioner might have received the promotion. Id., at 69-70. As the Director summarized his decision to promote Joyce, the underrepresentation of women in skilled craft positions was only one element of a number of considerations that led to the promotion of Ms. Joyce. Ibid. While I agree with Justice Scalia’s dissent that an affirmative action program that automatically and blindly promotes those marginally qualified candidates falling within a preferred race or gender category, or that can be equated with a permanent plan of “proportionate representation by race and sex,” would violate Title VII, I cannot agree that this is such a case. Rather, as the Court demonstrates, Joyce’s sex was simply used as a “plus” factor. Ante, at 636-637.
In this case, I am also satisfied that respondents had a firm basis for adopting an affirmative action program. Although the District Court found no discrimination against women in fact, at the time the affirmative action plan was adopted, there were no women in its skilled craft positions. Petitioner concedes that women constituted approximately 5% of the local labor pool of skilled craft workers in 1970. Reply Brief for Petitioner 9. Thus, when compared to the percentage of women in the qualified work force, the statistical disparity would have been sufficient for a prima facie Title VII case brought by unsuccessful women job applicants. See Teamsters, 431 U. S., at 342, n. 23 (“[F]ine tuning of the statistics could not have obscured the glaring absence of minority line drivers. . . . [T]he company’s inability to rebut the in*657ference of discrimination came not from a misuse of statistics but from ‘the inexorable zero’”).
In sum, I agree that respondents’ affirmative action plan as implemented in this instance with respect to skilled craft positions satisfies the requirements of 'Weber and of Wygant. Accordingly, I concur in the judgment of the Court.