concurring in part and dissenting in part.
I concur in the majority’s affirmance of the trial court’s finding of discrimination in job assignments. Likewise, I concur that plaintiffs lack standing to challenge Western Electric’s hiring practices and the alleged sex discrimination in promotions in the Installation facility.
I must voice vigorous disagreement, however, with the reversal of the district court’s findings of discrimination relating to the other promotion practices of Western Electric. I agree with the trial court’s finding that plaintiffs not only established a prima facie case of discrimination but also presented strong evidence which amply demonstrates that the facially neutral promotion practices have a disparate impact on blacks and female employees entitling them to the remedial relief.
The trial court’s opinion adopts as statistical support for its disparate impact finding the data set forth in plaintiffs’ Proposed Findings of Fact (If 104-110). These statistics establish:
1. While blacks constituted 26.4% of those available for promotion from hourly-rated to salaried nonsupervisory jobs, only 9.8% of those promoted were black;
2. While females constituted 18.8% of those available for promotion from hourly-rated to salaried nonsupervisory jobs, only 11.9% of those promoted were female;
3. Only 2.8% of the nonsupervisory employees promoted to installation supervisory jobs were black, even though blacks comprised as much as 23% of the work force from which such promotions were made during this period, and no females were promoted to these supervisory jobs;
4. Only 8.9% of the nonsupervisory employees promoted to service center supervisory jobs were black, and only 5.4% were female, even though blacks constituted as much as 26% and females 27% of the work force from which such promotions were made during this period.
Thus, the district court relied on the disparities between the percentages of blacks and females in the pool of employees eligible for promotion to salaried and supervisory positions and the percentages of blacks and females actually promoted to such positions.
The majority opinion advances two sets of statistical data which allegedly provide a more accurate basis for assessing the impact of Western Electric’s promotion practices. First, the majority suggests the Washington SMSA statistics might contain a more appropriate standard of comparison. SMSA figures would yield a ratio of 4% blacks and 10% females in Western Electric’s salaried and section chief positions. When these percentages are compared to the proportions of blacks and females actually promoted, the conclusion is drawn that no disparate impact exists. In addition, the majority opinion utilizes statistical data which suggests that an employee should have at least ten years experience to be qualified for promotion to salaried positions in the Service Center. Since only 7% of the blacks and 15% of the females in the hourly work force in 1973 had the requisite experience, it is argued that the statistical data offered by plaintiffs fails to establish even a prima facie showing of disparate impact. From the foregoing it is concluded that the trial court erred and that there was a “total failure of proof of any disparate effect in promotions upon blacks or women in the Service Center or upon blacks in Installation.” Ante at 106.
*108The trial court considered the efficacy of using SMSA statistics and experience-related data to determine whether a prima facie showing of discrimination was made. In properly rejecting the SMSA data, Judge Bryan observed:
The facility in question is located in Arlington County, Virginia, a largely residential suburb of the District of Columbia. The Washington SMSA includes not only Arlington County, but the District of Columbia, the Cities of Alexandria and Falls Church, the Counties of Fairfax, Loudoun and Prince William in Virginia, and the Counties of Montgomery and Prince Georges in Maryland. This is too large an area to be considered as the area from which an employer, situated as the defendant is here, draws for its labor market. Here the entry level jobs at both units are, except for secretarial employees, for the unskilled, and the census data does not provide an accurate or reliable indication of persons who are available for work in a particular job with a particular employer. Insofar as vocational and occupational data are concerned, the census data does not, of course, take into account existing discrimination.
Hill v. Western Electric Co., 12 FEP Cases 1175, 1179 (E.D.Va.1976).
It is difficult for me to understand how the over-inclusive census statistics contained in the Washington SMSA can be considered more probative to the issue at hand than the actual work force from which the promotions are made at the specific plant in question.
The majority opinion does not rest on the SMSA statistics alone, but rather relies primarily on the experience-related data to support its conclusion that no disparate impact was proven. The majority’s use of the lengthy-experience “requirement” reflects a fundamental misconception regarding the proper order and nature of proof in disparate impact actions under Title VII.
As previously noted, the trial court did consider the “experience” claims of Western Electric when determining whether discrimination existed. The promotion practices utilized by Western Electric, however, precluded an initial precise definition of the pool of qualified employees. Under Western Electric’s promotion procedures, employees had to be placed on a list by their section chief in order to be eligible for promotion. Western Electric provided the section chiefs no written guidelines setting forth the qualifications necessary, or the criteria used, for promotion of employees. The factors employed by the section chiefs in determining whether to place an employee on the list were vague and subjective. Furthermore, Western Electric stipulated that no specific number of years of experience is necessary to be qualified for promotion. Accordingly, the trial court used the racial and sexual composition of the entire hourly-employee work force as the most probative labor market percentages. At this juncture, the “experience” needs of Western Electric were properly addressed by the trial court to determine whether Western Electric successfully rebutted the prima facie showing of discrimination. In light of the fact that Western Electric had totally failed to apprise section chiefs of any promotion qualification requirements, the order in which the trial court evaluated the proof was clearly justified.1
*109The trial court’s inquiry did not begin and end with the statistical data. Judge Bryan properly allowed Western Electric to present evidence which would cast doubt on the propriety of using the entire hourly-wage work force in determining whether promotion practices had a disparate impact on blacks and females. Western Electric strenuously argued before the trial court that experience is relevant in making promotion decisions. In assessing the credibility of the “experience defense” proffered by Western Electric, the trial judge noted that a lengthy-experience requirement could be used to perpetuate discrimination in hiring. Although plaintiffs have no standing to launch a frontal assault on Western Electrie’s hiring practices, the trial court could legitimately consider evidence of such discrimination when evaluating the experience claim.2 See Rowe v. General Motors Corp., 457 F.2d 348, 356 (5th Cir. 1972); cf. Hazelwood School District v. United States, 433 U.S. 299 at 309 n. 15, 97 S.Ct. 2736, 53 L.Ed.2d 768. Furthermore, contrary to the majority’s conclusion, the relevance of experience in determining whether an employee is qualified for promotion is not overlooked in the trial court’s remedial order, nor is it necessarily obviated in the master’s duty to award back pay related to promotional transfers.3
*110In addition, in rejecting the experience factor as a threshold issue, the district court viewed not only the statistical proof but also the overall record regarding the subjective employment practices in which promotions were made. I think it significant to highlight this other evidence, not otherwise discussed in the majority opinion. Judge Bryan, in an exhaustive and analytical opinion, wrote:
Promotion — Service Center.
Promotion within hourly positions is done strictly by departmental seniority unless an individual is determined to be unqualified. As yet, no one has ever been found to be unqualified.
The potential problem with this scheme of promotion is that it perpetuates past discrimination and reflects discrimination in original hiring. The fact that promotions are always made from within the particular section in which the vacancy occurs adds to this, since transfer among sections is prohibited. However the Court does not find any discrimination in the system of promotion.
Hourly to Salaried Non-Supervisory
Promotion of hourly workers to salaried non-supervisory positions is made from a list of recommended employees. The initial recommendation to place an employee upon the list is made by the section chief. This recommendation is reviewed by two higher supervisory levels. The specific promotion recommendation is made by an advancement committee but sometimes by the section chief. The final decision to promote is made by the assistant manager and manager.
The statistics support, and the Court finds an adverse impact on blacks and females and discrimination in this class of promotion (Plaintiff’s Proposed Findings of Fact 1f|f 104 — 110). The promotional procedure itself is supportive of the Court’s finding, because an employee cannot be promoted unless he is placed upon the list and the only way that can be done is by the section chief. Section chiefs are given no written guidelines for this task and the factors employed by the section chiefs are necessarily vague and subjective. Additionally, vacancies are not posted and there is no way for an individual to apply for a particular position. The section chief’s decision is final and unreviewable.
Non-Supervisory to Supervisory
Supervisory promotions are made from a Management Potential Inventory. In order to be listed, an employee must first complete a request form. Annually, all supervisors meet and there determine who shall be placed upon the list. The actual promotion decision is made and reviewed up four levels of supervisory command.
The statistics from 1965 support a charge of discrimination (Exhibit P-24S). In addition, the promotion procedure is subject to most of the same objections as “Hourly to Salaried Non-Supervisory.” The process is basically informal and non-structured. There are no written guidelines for evaluating potential supervisory personnel and the promotion decision is based upon the subjective evaluations of supervisors. The process is secret — vacancies are not posted and no one is allowed to “apply” for a job, only the list.
The defendant seeks to offset the plaintiff’s statistics with statistics of its own. No one has been promoted since 1972. In 1972, five people — including 1 black and 2 females — were promoted. Defendant claims that the overall disparate statistics are due to the fact that generally, only those employees with ten *111or more years of experienced [sic] are promoted. Only one person hired since 1965 — a woman — has been promoted to supervisor. Nevertheless, the statistics are such that they cannot be explained away in this manner. They result, the Court finds, from past discrimination, and warrant, at the very least, injunctive relief.
Promotion — Installation.
Supervisory Positions
Annually, the department chiefs and the district manager select names of non-supervisory employees and place them on a Management Potential Inventory. When a supervisory vacancy occurs, an employee is selected from this list to fill it.
The statistical evidence supports a finding of discrimination. There has only been one black and no female supervisors. The defendant once again asserts that length of employment is the basic criterion for promotion. In addition, there have been only five promotions since 1970 and none since 1973. In fact, since 1972 there has been a net downgrading from supervisory to hourly of 28 positions.
The objections to the actual promotion procedure are similar to those for the Service Center. An employee must be recommended to be placed upon the Management Potential Inventory; and this decision is unstructured and subjective. Again, as in the case of the Service Center, injunctive relief is warranted.
12 FEP Cases at 1181-83.
In view of this additional evidence and the exhaustive and specific findings of fact by the trial court, I find it difficult to say there is a total failure of proof by plaintiffs to show discrimination in promotional practices. Even if it is assumed that the statistical comparison employed by the district court fails to furnish a precise measure of Western Electric’s conduct, the additional findings made by the trial court warrant injunctive relief. As this court stated in Patterson v. American Tobacco Co.:
The fact that the company’s appointments since 1965 exceed the ratio of qualified blacks and women in the workforce does not exonerate the company for the violations of the Act which the district court found. The tardy appointments of blacks and women to supervisory positions long after the passage of Title VII and the present lack of published job descriptions and objective selection procedures fully justify the injunctive relief the district court ordered.
535 F.2d at 275.
Any number of cases, including decisions of this circuit, have emphasized that subjective practices utilized by defendant constitute strong evidence of discrimination. See, e. g., Parson v. Kaiser Aluminum & Chemical Corp., 575 F.2d 1374, 1384-85 (5th Cir. 1978); Roman v. ESB, Inc., 550 F.2d 1343, 1351 (4th Cir. 1976); Stewart v. General Motors Corp., 542 F.2d 445, 450-51 (7th Cir. 1976), cert. denied, 433 U.S. 919, 97 S.Ct. 2995, 53 L.Ed.2d 1105 (1977); Patterson v. American Tobacco Co., 535 F.2d at 272-73; Muller v. United States Steel Corp., 509 F.2d 923, 928 (10th Cir.), cert. denied, 423 U.S. 825, 96 S.Ct. 39, 46 L.Ed.2d 41 (1975); United States v. N. L. Industries, Inc., 479 F.2d 354, 368 (8th Cir. 1973); Rowe v. General Motors Corp., 457 F.2d at 358-59.
For the foregoing reasons, I would defer to the trial court’s careful analysis.
. Assuming, as the majority opinion apparently does, that ten years of experience was the prerequisite for promotion, plaintiffs could have attacked that qualification requirement as a facially neutral employment practice having a disparate impact on blacks and females. The statistics relied on by the trial court clearly support such a claim. Western Electric would then have been required to show that such a stringent experience requirement was justified by “business necessity.” See Griggs v. Duke Power Co., 401 U.S. 424, 432, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). In the instant case, however, plaintiffs can hardly be faulted for not directly attacking a job qualification requirement which Western Electric stipulated did not exist. Furthermore, in suggesting the ten year statistics as a basis for comparison, the majority suggests a defense which Western Electric candidly denies. Western states in its brief:
Western does not contend that it established the “business necessity” of a particular number of years of experience as a qualification for promotion, but that experience is relevant to the threshold inquiry of whether Western’s *109promotion practices have had an adverse impact.
Reply Brief of appellant at 19 (emphasis added).
I would make one additional comment on the majority’s treatment of the experience factor. By using ten years experience as the factor which defines the qualified work force, the majority, in effect, evaluates lengthy experience to a promotion qualification. The majority opinion justifies this conclusion by noting that “one naturally and inevitably looks to those who have acquired experience and demonstrated skills” when selecting supervisors. Ante at 105. The fact that an employer would probably look to its experienced employees when determining who should be promoted may negate an inference of discriminatory intent, but it does not resolve an adverse impact claim. By definition, an adverse impact cause of action arises when job qualifications which are facially neutral and neutral in terms of intent fall more harshly on minorities. See, e. g., Griggs v. Duke Power Co., supra; Stewart v. General Motors Corp., 542 F.2d 445, 450 (7th Cir. 1976), cert. denied, 433 U.S. 919, 97 S.Ct. 2995, 53 L.Ed.2d 1105 (1977); Patterson v. American Tobacco Co., 535 F.2d 257, 268 (4th Cir.), cert. denied, 429 U.S. 920, 97 S.Ct. 314, 50 L.Ed.2d 286 (1976); United States v. Dillon Supply Co., 429 F.2d 800, 804 (4th Cir. 1970). When examining such job qualifications
the applicable test is not merely whether there exists a business purpose for adhering to a challenged practice. The test is whether there exists an overriding legitimate business purpose such that the practice is necessary to the safe and efficient operation of the business. Thus, the business purpose must be sufficiently compelling to override any racial impact; the challenged practice must effectively carry out the business purpose it is alleged to serve; and there must be available no acceptable alternative policies or practices which would better accomplish the business
purpose advanced, or accomplish it equally well with a lesser differential racial impact.
Robinson v. Lorillard Corp., 444 F.2d 791, 798 (4th Cir. 1971), cert. dismissed, 404 U.S. 1006, 92 S.Ct. 573, 30 L.Ed.2d 655 (1972).
. To support its finding of discrimination in hiring the trial court relied on the following:
(a) Only 12.7% of the black applicants for entry-level jobs in defendant’s Service Center from 1970 through 1974 were hired as compared to 29.8% of the white applicants; while 1,489 of the 3,382 applicants (or 44.4%) were black, only 189 of the 754 applicants hired (or 25.1%) were black.
(b) Only 17.9% of the black applicants for entry-level jobs in defendant’s installation organization from 1968 through 1971 were hired as compared to 45.7% of the white applicants; while 1,293 of the 2,760 applicants (or 46.8%) were black, only 232 of the 903 applicants hired (or 24.7%) were black.
(c) Only 16.9% of the female applicants for entry-level jobs in defendant’s Service Center from 1970 through 1974 were hired, as compared to 26.8% of the male applicants; while 1,443 of the 3,511 applicants (or 41.1%) were female, only 244 of the 799 applicants hired (or 30.5%) were female.
12 FEP Cases at 1179.
. With regard to priority promotions, the trial court’s remedial order states:
Promotion shall be offered only to those eligible claimants who are employed at the time by the defendant, and who are qualified on a job-related, non-discriminatory basis.
If a claimant is found to be eligible by the Master, but subsequently is found to be unable to perform the duties of a position at the time the defendant otherwise would be required to make a priority offer of that position to that claimant, no obligation to make such an offer shall be imposed on defendant.
*110With regard to back pay, the order states:
[T]he Master may take into account the eligible claimant’s actual employment history and such other factors as he may deem relevant to that claimant’s performance potential, and may adjust the formula figure up or down accordingly, stating the reasons for such adjustment.
Defendant shall have the opportunity to seek reduction of the net back pay award for each eligible claimant by showing higher actual earnings, or earnings obtainable through due diligence, or demonstrable factors probative on the question of how the claimant might have performed had no discrimination occurred.