delivered the opinion of the Court.
Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U. S. C. §2000e et seq., makes it an unfair employment practice for an employer to discriminate against any individual with respect to hiring or the terms and condition of employment because of such individual’s race, color, religion, sex, or national origin; or to limit, segregate, or classify his employees in ways that would adversely affect any employee because of the employee’s race, color, religion, sex, or national origin.1 §2000e-2(a). Griggs v. Duke Power Co., 401 U. S. 424, 431 (1971), construed Title VII to proscribe “not only overt discrimination but also practices that are fair in form but discriminatory in practice.” Under this basis for liability, which is known as the “disparate-impact” theory and which is involved in this case, a facially neutral *646employment practice may be deemed violative of Title VII without evidence of the employer’s subjective intent to discriminate that is required in a “disparate-treatment” case.
I
The claims before us are disparate-impact claims, involving the employment practices of petitioners, two companies that operate salmon canneries in remote and widely separated areas of Alaska. The canneries operate only during the salmon runs in the summer months. They are inoperative and vacant for the rest of the year. In May or June of each year, a few weeks before the salmon runs begin, workers arrive and prepare the equipment and facilities for the canning operation. Most of these workers possess a variety of skills. When salmon runs are about to begin, the workers who will operate the cannery lines arrive, remain as long as there are fish to can, and then depart. The canneries are then closed down, winterized, and left vacant until the next spring. During the off-season, the companies employ only a small number of individuals at their headquarters in Seattle and Astoria, Oregon, plus some employees at the winter shipyard in Seattle.
The length and size of salmon runs vary from year to year, and hence the number of employees needed at each cannery also varies. Estimates are made as early in the winter as possible; the necessary employees are hired, and when the time comes, they are transported to the canneries. Salmon must be processed soon after they are caught, and the work during the canning season is therefore intense.2 For this *647reason, and because the canneries are located in remote regions, all workers are housed at the canneries and have their meals in company-owned mess halls.
Jobs at the canneries are of two general types: “cannery jobs” on the cannery line, which are unskilled positions; and “noncannery jobs,” which fall into a variety of classifications. Most noncannery jobs are classified as skilled positions.3 Cannery jobs are filled predominantly by nonwhites: Filipinos and Alaska Natives. The Filipinos are hired through, and dispatched by, Local 37 of the International Longshoremen’s and Warehousemen’s Union pursuant to a hiring hall agreement with the local. The Alaska Natives primarily reside in villages near the remote cannery locations. Non-cannery jobs are filled with predominantly white workers, who are hired during the winter months from the companies’ offices in Washington and Oregon. Virtually all of the non-cannery jobs pay more than cannery positions. The predominantly white noncannery workers and the predominantly nonwhite cannery employees live in separate dormitories and eat in separate mess halls.
In 1974, respondents, a class of nonwhite cannery workers who were (or had been) employed at the canneries, brought this Title VII action against petitioners. Respondents alleged that a variety of petitioners’ hiring/promotion practices— e. g., nepotism, a rehire preference, a lack of objective hiring criteria, separate hiring channels, a practice of not promoting from within — were responsible for the racial strati*648fication of the work force and had denied them and other nonwhites employment as noncannery workers on the basis of race. Respondents also complained of petitioners’ racially segregated housing and dining facilities. All of respondents’ claims were advanced under both the disparate-treatment and disparate-impact theories of Title VII liability.
The District Court held a bench trial, after which it entered 172 findings of fact. 34 EPD ¶34,437, pp. 33,822-33,836 (WD Wash. 1983). It then rejected all of respondents’ disparate-treatment claims. It also rejected the disparate-impact challenges involving the subjective employment criteria used by petitioners to fill these noncannery positions, on the ground that those criteria were not subject to attack under a disparate-impact theory. Id., p.33,840. Petitioners’ “objective” employment practices (e. g., an English language requirement, alleged nepotism in hiring, failure to post noncannery openings, the rehire preference, etc.) were found to be subject to challenge under the disparate-impact theory, but these claims were rejected for failure of proof. Judgment was entered for petitioners.
On appeal, a panel of the Ninth Circuit affirmed, 768 F. 2d 1120 (1985), but that decision was vacated when the Court of Appeals agreed to hear the case en banc, 787 F. 2d 462 (1985). The en banc hearing was ordered to settle an intracircuit conflict over the question whether subjective hiring practices could be analyzed under a disparate-impact model; the Court of Appeals held — as this Court subsequently ruled in Watson v. Fort Worth Bank & Trust, 487 U. S. 977 (1988)— that disparate-impact analysis could be applied to subjective hiring practices. 810 F. 2d 1477, 1482 (1987). The Ninth Circuit also concluded that in such a case, “[o]nce the plaintiff class has shown disparate impact caused by specific, identifiable employment practices or criteria, the burden shifts to the employer,” id., at 1485, to “prov[e the] business necessity” of the challenged practice, id., at 1486. Because the en banc holding on subjective employment practices reversed *649the District Court’s contrary ruling, the en banc Court of Appeals remanded the case to a panel for further proceedings.
On remand, the panel applied the en banc ruling to the facts of this case. 827 F. 2d 439 (1987). It held that respondents had made out a prima facie case of disparate impact in hiring for both skilled and unskilled noncannery positions. The panel remanded the case for further proceedings, instructing the District Court that it was the employer’s burden to prove that any disparate impact caused by its hiring and employment practices was justified by business necessity. Neither the en banc court nor the panel disturbed the District Court’s rejection of the disparate-treatment claims.4
Petitioners sought review of the Court of Appeals’ decision in this Court, challenging it on several grounds. Because some of the issues raised by the decision below were matters *650on which this Court was evenly divided in Watson v. Fort Worth Bank & Trust, supra, we granted certiorari, 487 U. S. 1264 (1988), for the purpose of addressing these disputed questions of the proper application of Title VII’s disparate-impact theory of liability.
II
In holding that respondents had made out a prima facie case of disparate impact, the Court of Appeals relied solely on respondents’ statistics showing a high percentage of nonwhite workers in the cannery jobs and a low percentage of such workers in the noncannery positions.5 Although statistical proof can alone make out a prima facie case, see Teamsters v. United States, 431 U. S. 324, 339 (1977); Hazelwood School Dist. v. United States, 433 U. S. 299, 307-308 (1977), the Court of Appeals’ ruling here misapprehends our precedents and the purposes of Title VII, and we therefore reverse.
“There can be no doubt,” as there was when a similar mistaken analysis had been undertaken by the courts below in Hazelwood, supra, at 308, “that the . . . comparison . . . fundamentally misconceived the role of statistics in employment discrimination cases.” The “proper comparison [is] between the racial composition of [the at-issue jobs] and the racial composition of the qualified . . . population in the relevant labor market.” Ibid. It is such a comparison — between the racial composition of the qualified persons in the labor market and the persons holding at-issue jobs — that generally forms *651the proper basis for the initial inquiry in a disparate-impact case. Alternatively, in cases where such labor market statistics will be difficult if not impossible to ascertain, we have recognized that certain other statistics — such as measures indicating the racial composition of “otherwise-qualified applicants” for at-issue jobs — are equally probative for this purpose. See, e. g., New York City Transit Authority v. Beazer, 440 U. S. 568, 585 (1979).6
It is clear to us that the Court of Appeals’ acceptance of the comparison between the racial composition of the cannery work force and that of the noncannery work force, as probative of a prima facie case of disparate impact in the selection of the latter group of workers, was flawed for several reasons. Most obviously, with respect to the skilled non-cannery jobs at issue here, the cannery work force in no way reflected “the pool of qualified job applicants” or the “qualified population in the labor force.” Measuring alleged discrimination in the selection of accountants, managers, boat captains, electricians, doctors, and engineers — and the long list of other “skilled” noncannery positions found to exist by the District Court, see 34 EPD ¶ 34,437, p. 33,832 — by comparing the number of nonwhites occupying these jobs to the number of nonwhites filling cannery worker positions is nonsensical. If the absence of minorities holding such skilled positions is due to a dearth of qualified nonwhite applicants (for reasons that are not petitioners’ fault),7 petition*652ers’ selection methods or employment practices cannot be said to have had a “disparate impact” on nonwhites.
One example illustrates why this must be so. Respondents’ own statistics concerning the noncannery work force at one of the canneries at issue here indicate that approximately 17% of the new hires for medical jobs, and 15% of the new hires for officer worker positions, were nonwhite. See App. to Brief for Respondents B-l. If it were the case that less than 15 to 17% of the applicants for these jobs were nonwhite and that nonwhites made up a lower percentage of the relevant qualified labor market, it is hard to see how respondents, without more, cf. Connecticut v. Teal, 457 U. S. 440 (1982), would have made out a prima facie case of disparate impact. Yet, under the Court of Appeals’ theory, simply because nonwhites comprise 52% of the cannery workers at the cannery in question, see App. to Brief for Respondents B-l, respondents would be successful in establishing a prima facie case of racial discrimination under Title VII.
Such a result cannot be squared with our cases or with the goals behind the statute. The Court of Appeals’ theory, at the very least, would mean that any employer who had a segment of his work force that was — for some reason — racially imbalanced, could be haled into court and forced to engage in the expensive and time-consuming task of defending the “business necessity” of the methods used to select the other members of his work force. The only practicable option for many employers would be to adopt racial quotas, insuring that no portion of their work forces deviated in racial composition from the other portions thereof; this is a result that Congress expressly rejected in drafting Title VII. See 42 U. S. C. §2000e-2(j); see also Watson v. Fort Worth Bank & Trust, 487 U. S. at 922-994, and n. 2 (opinion of O’Connor, J.). The Court of Appeals’ theory would “leave the employer little choice . . . but to engage in a subjective quota system of employment selection. This, of course, is far from the intent of Title VII.” Albemarle Paper Co. v. Moody, *653422 U. S. 405, 449 (1975) (Blackmun, J., concurring in judgment).
The Court of Appeals also erred with respect to the unskilled noncannery positions. Racial imbalance in one segment of an employer’s work force does not, without more, establish a prima facie case of disparate impact with respect to the selection of workers for the employer’s other positions, even where workers for the different positions may have somewhat fungible skills (as is arguably the case for cannery and unskilled noncannery workers). As long as there are no barriers or practices deterring qualified nonwhites from applying for noncannery positions, see n. 6, supra, if the percentage of selected applicants who are nonwhite is not significantly less than the percentage of qualified applicants who are nonwhite, the employer’s selection mechanism probably does not operate with a disparate impact on minorities.8 Where this is the case, the percentage of nonwhite workers found in other positions in the employer’s labor force is irrelevant to the question of a prima facie statistical case of disparate impact. As noted above, a contrary ruling on this point would almost inexorably lead to the use of numerical quotas in the workplace, a result that Congress and this Court have rejected repeatedly in the past.
Moreover, isolating the cannery workers as the potential “labor force” for unskilled noncannery positions is at once both too broad and too narrow in its focus. It is too broad because the vast majority of these cannery workers did not *654seek jobs in unskilled noneannery positions; there is no showing that many of them would have done so even if none of the arguably “deterring” practices existed. Thus, the pool of cannery workers cannot be used as a surrogate for the class of qualified job applicants because it contains many persons who have not (and would not) be noncannery job applicants. Conversely, if respondents propose to use the cannery workers for comparison purposes because they represent the “qualified labor population” generally, the group is too narrow because there are obviously many qualified persons in the labor market for noncannery jobs who are not cannery workers.
The peculiar facts of this case further illustrate why a comparison between the percentage of nonwhite cannery workers and nonwhite noncannery workers is an improper basis for making out a claim of disparate impact. Here, the District Court found that nonwhites were “overrepresent[ed]” among cannery workers because petitioners had contracted with a predominantly nonwhite union (local 37) to fill these positions. See 34 EPD ¶33,437, p. 33,829. As a result, if petitioners (for some permissible reason) ceased using local 37 as its hiring channel for cannery positions, it appears (according to the District Court’s findings) that the racial stratification between the cannery and noncannery workers might diminish to statistical insignificance. Under the Court of Appeals’ approach, therefore, it is possible that with no change whatsoever in their hiring practices for noncannery workers — the jobs at issue in this lawsuit — petitioners could make respondents’ prima facie case of disparate impact “disappear. ” But ¿/‘there would be no prima facie case of disparate impact in the selection of noncannery workers absent petitioners’ use of local 37 to hire cannery workers, surely petitioners’ reliance on the union to fill the cannery jobs not at issue here (and its resulting “overrepresentation” of nonwhites in those positions) does not — standing alone — make out a prima facie case of disparate impact. Yet it is precisely *655such an ironic result that the Court of Appeals reached below.
Consequently, we reverse the Court of Appeals’ ruling that a comparison between the percentage of cannery workers who are nonwhite and the percentage of noncannery workers who are nonwhite makes out a prima facie case of disparate impact. Of course, this leaves unresolved whether the record made in the District Court will support a conclusion that a prima facie case of disparate impact has been established on some basis other than the racial disparity between cannery and noncannery workers. This is an issue that the Court of Appeals or the District Court should address in the first instance.
Ill
Since the statistical disparity relied on by the Court of Appeals did not suffice to make out a prima facie case, any inquiry by us into whether the specific challenged employment practices of petitioners caused that disparity is pretermitted, as is any inquiry into whether the disparate impact that any employment practice may have had was justified by business considerations.9 Because we remand for further proceedings, however, on whether a prima facie case of disparate impact has been made in defensible fashion in this case, we address two other challenges petitioners have made to the decision of the Court of Appeals.
*656A
First is the question of causation in a disparate-impact case. The law in this respect was correctly stated by Justice O’Connor’s opinion last Term in Watson v. Fort Worth Bank & Trust, 487 U. S., at 994:
“[W]e note that the plaintiff’s burden in establishing a prima facie case goes beyond the need to show that there are statistical disparities in the employer’s work force. The plaintiff must begin by identifying the specific employment practice that is challenged. . . . Especially in cases where an employer combines subjective criteria with the use of more rigid standardized rules or tests, the plaintiff is in our view responsible for isolating and identifying the specific employment practices that are allegedly responsible for any observed statistical disparities.”
Cf. also id., at 1000 (Blackmun, J., concurring in part and concurring in judgment).
Indeed, even the Court of Appeals — whose decision petitioners assault on this score — noted that “it is . . . essential that the practices identified by the cannery workers be linked causally with the demonstrated adverse impact.” 827 F. 2d, at 445. Notwithstanding the Court, of Appeals’ apparent adherence to the proper inquiry, petitioners contend that that court erred by permitting respondents to make out their case by offering “only [one] set of cumulative comparative statistics as evidence of the disparate impact of each and all of [petitioners’ hiring] practices.” Brief for Petitioners 31.
Our disparate-impact cases have always focused on the impact of particular hiring practices on employment opportunities for minorities. Just as an employer cannot escape liability under Title VII by demonstrating that, “at the bottom line,” his work force is racially balanced (where particular hiring practices may operate to deprive minorities of employment opportunities), see Connecticut v. Teal, 457 U. S., at *657450, a Title VII plaintiff does not make out a case of disparate impact simply by showing that, “at the bottom line,” there is racial imbalance in the work force. As a general matter, a plaintiff must demonstrate that it is the application of a specific or particular employment practice that has created the disparate impact under attack. Such a showing is an integral part of the plaintiff’s prima facie case in a disparate-impact suit under Title VII.
Here, respondents have alleged that several “objective” employment practices (e. g., nepotism, separate hiring channels, rehire preferences), as well as the use of “subjective decision making” to select noncannery workers, have had a disparate impact on nonwhites. Respondents base this claim on statistics that allegedly show a disproportionately low percentage of nonwhites in the at-issue positions. However, even if on remand respondents can show that nonwhites are underrepresented in the at-issue jobs in a manner that is acceptable under the standards set forth in Part II, supra, this alone will not suffice to make out a prima facie case of disparate impact. Respondents will also have to demonstrate that the disparity they complain of is the result of one or more of the employment practices that they are attacking here, specifically showing that each challenged practice has a significantly disparate impact on employment opportunities for whites and nonwhites. To hold otherwise would result in employers being potentially liable for “the myriad of innocent causes that may lead to statistical imbalances in the composition of their work forces.” Watson v. Fort Worth Bank & Trust, supra, at 992.
Some will complain that this specific causation requirement is unduly burdensome on Title VII plaintiffs. But liberal civil discovery rules give plaintiffs broad access to employers’ records in an effort to document their claims. Also, employers falling within the scope of the Uniform Guidelines on Employee Selection Procedures, 29 CFR §1607.1 et seq. (1988), *658are required to “maintain . . . records or other information which will disclose the impact which its tests and other selection procedures have upon employment opportunities of persons by identifiable race, sex, or ethnic group[s].” See § 1607.4(A). This includes records concerning “the individual components of the selection process” where there is a significant disparity in the selection rates of whites and nonwhites. See § 1607.4(C). Plaintiffs as a general matter will have the benefit of these tools to meet their burden of showing a causal link between challenged employment practices and racial imbalances in the work force; respondents presumably took full advantage of these opportunities to build their case before the trial in the District Court was held.10
Consequently, on remand, the courts below are instructed to require, as part of respondents’ prima facie case, a demonstration that specific elements of the petitioners’ hiring process have a significantly disparate impact on non whites.
B
If, on remand, respondents meet the proof burdens outlined above, and establish a prima facie case of disparate impact with respect to any of petitioners’ employment practices, the case will shift to any business justification petitioners offer for their use of these practices. This phase of the disparate-impact case contains two components: first, a consideration of the justifications an employer offers for his use of these practices; and second, the availability of alternative practices to achieve the same business ends, with less racial impact. See, e. g., Albemarle Paper Co. v. Moody, 422 U. S., at 425. We consider these two components in turn.
*659(1)
Though we have phrased the query differently in different cases, it is generally well established that at the justification stage of such a disparate-impact case, the dispositive issue is whether a challenged practice serves, in a significant way, the legitimate employment goals of the employer. See, e. g., Watson v. Fort Worth Bank & Trust, 487 U. S., at 997-999; New York City Transit Authority v. Beazer, 440 U. S., at 587, n. 31; Griggs v. Duke Power Co., 401 U. S., at 432. The touchstone of this inquiry is a reasoned review of the employer’s justification for his use of the challenged practice. A mere insubstantial justification in this regard will not suffice, because such a low standard of review would permit discrimination to be practiced through the use of spurious, seemingly neutral employment practices. At the same time, though, there is no requirement that the challenged practice be “essential” or “indispensable” to the employer’s business for it to pass muster: this degree of scrutiny would be almost impossible for most employers to meet, and would result in a host of evils we have identified above. See supra, at 652-653.
In this phase, the employer carries the burden of producing evidence of a business justification for his employment practice. The burden of persuasion, however, remains with the disparate-impact plaintiff. To the extent that the Ninth Circuit held otherwise in its en banc decision in this case, see 810 F. 2d, at 1485-1486, or in the panel’s decision on remand, see 827 F. 2d, at 445, 447 — suggesting that the persuasion burden should shift to petitioners once respondents established a prima facie case of disparate impact — its decisions were erroneous. “[T]he ultimate burden of proving that discrimination against a protected group has been caused by a specific employment practice remains with the plaintiff at all times.'” Watson, supra, at 997 (O’Con-nor, J.) (emphasis added). This rule conforms with the usual method for allocating persuasion and production bur*660dens in the federal courts, see Fed. Rule Evid. 301, and more specifically, it conforms to the rule in disparate-treatment cases that the plaintiff bears the burden of disproving an employer’s assertion that the adverse employment action or practice was based solely on a legitimate neutral consideration. See Texas Dept. of Community Affairs v. Burdine, 450 U. S. 248, 256-258 (1981). We acknowledge that some of our earlier decisions can be read as suggesting otherwise. See Watson, supra, at 1006-1008 (Blackmun, J., concurring in part and concurring in judgment). But to the extent that those cases speak of an employer’s “burden of proof” with respect to a legitimate business justification defense, see, e. g., Dothard v. Rawlinson, 433 U. S. 321, 329 (1977), they should have been understood to mean an employer’s production— but not persuasion — burden. Cf., e. g., NLRB v. Transportation Management Corp., 462 U. S. 393, 404, n. 7 (1983). The persuasion burden here must remain with the plaintiff, for it is he who must prove that it was “because of such individual’s race, color,” etc., that he was denied a desired employment opportunity. See 42 U. S. C. §2000e-2(a).
(2)
Finally, if on remand the case reaches this point, and respondents cannot persuade the trier of fact on the question of petitioners’ business necessity defense, respondents may still be able to prevail. To do so, respondents will have to persuade the factfinder that “other tests or selection devices, without a similarly undesirable racial effect, would also serve the employer’s legitimate [hiring] interest[s]”; by so demonstrating, respondents would prove that “[petitioners were] using [their] tests merely as a ‘pretext’ for discrimination.” Albemarle Paper Co., supra, at 425; see also Watson, 487 U. S., at 998 (O’Connor, J.); id., at 1005-1006 (Blackmun, J., concurring in part and concurring in judgment). If respondents, having established a prima facie case, come forward with alternatives to petitioners’ hiring practices that *661reduce the racially disparate impact of practices currently being used, and petitioners refuse to adopt these alternatives, such a refusal would belie a claim by petitioners that their incumbent practices are being employed for nondiscriminatory reasons.
Of course, any alternative practices which respondents offer up in this respect must be equally effective as petitioners’ chosen hiring procedures in achieving petitioners’ legitimate employment goals. Moreover, “[fjactors such as the cost or other burdens of proposed alternative selection devices are relevant in determining whether they would be equally as effective as the challenged practice in serving the employer’s legitimate business goals.” Watson, supra, at 998 (O’Connor, J.). “Courts are generally less competent than employers to restructure business practices,” Furnco Construction Corp. v. Waters, 438 U. S. 567, 578 (1978); consequently, the judiciary should proceed with care before mandating that an employer must adopt a plaintiff’s alternative selection or hiring practice in response to a Title VII suit.
IV
For the reasons given above, the judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion.
It is so ordered.
Title 42 U. S. C. §2000e-2(a), provides:
“(a) It shall be an unlawful employment practice for an employer—
“(1) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s race, color, religion, sex, or national origin; or
“(2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin.”
“Independent fishermen catch the salmon and turn them over to company-owned boats called ‘tenders,’ which transport the fish from the fishing grounds to the canneries. Once at the cannery, the fish are eviscerated, the eggs pulled, and they are cleaned. Then, operating at a rate of approximately four cans per second, the salmon are filled into cans. Next, the canned salmon are cooked under precise time-temperature requirements established by the FDA, and the cans are inspected to ensure *647that proper seals are maintained on the top, bottom and sides.” 768 F. 2d 1120, 1123 (CA9), vacated, 787 F. 2d 462 (1985).
The noncannery jobs were described as follows by the Court of Appeals: “Machinists and engineers are hired to maintain the smooth and continuous operation of the canning equipment. Quality control personnel conduct the FDA-required inspections and recordkeeping. Tenders are staffed with a crew necessary to operate the vessel. A variety of support personnel are employed to operate the entire cannery community, including, for example, cooks, carpenters, store-keepers, bookkeepers, beach gangs for dock yard labor and construction, etc.” 768 F. 2d, at 1123.
The fact that neither the District Court, nor the Ninth Circuit en banc, nor the subsequent Court of Appeals panel ruled for respondents on their disparate-treatment claims— i. e., their allegations of intentional racial discrimination-warrants particular attention in light of the dissents’ comment that the canneries “bear an unsettling resemblance to aspects of a plantation economy.” Post, at 664, n. 4 (Stevens, J., dissenting); post, at 662 (Blackmun, J., dissenting).
Whatever the “resemblance,” the unanimous view of the lower courts in this litigation has been that respondents did not prove that the canneries practice intentional racial discrimination. Consequently, Justice Black-mun’s hyperbolic allegation that our decision in this case indicates that this Court no longer “believes that race discrimination . . . against nonwhites ... is a problem in our society,” ibid., is inapt. Of course, it is unfortunately true that race discrimination exists in our country. That does not mean, however, that it exists at the canneries — or more precisely, that it has been proved to exist at the canneries.
Indeed, Justice Stevens concedes that respondents did not press before us the legal theories under which the aspects of cannery life that he finds to most resemble a “plantation economy” might be unlawful. Post, at 664, n. 4. Thus, the question here is not whether we “approve” of petitioners’ employment practices or the society that exists at the canneries, but, rather, whether respondents have properly established that these practices violate Title VII.
The parties dispute the extent to which there is a discrepancy between the percentage of nonwhites employed as cannery workers and those employed in noncannery positions. Compare, e. g., Brief for Petitioners 4-9 with Brief for Respondents 4-6. The District Court made no precise numerical findings in this regard, but simply noted that there were “significant disparities between the at-issue jobs [i. <?., noncannery jobs] and the total workforce at the canneries” which were explained by the fact that “nearly all employed in the ‘cannery worker’ department are non-white.” See 34 EPD ¶ 34,437, pp. 33,841, 33,829 (WD Wash. 1983).
For reasons explained below, the degree of disparity between these groups is not relevant to our decision here.
In fact, where “figures for the general population might. . . accurately reflect the pool of qualified job applicants,” cf. Teamsters v. United States, 431 U. S. 324, 340, n. 20 (1977), we have even permitted plaintiffs to rest their prima facie cases on such statistics as well. See, e. g., Dothard v. Rawlinson, 433 U. S. 321, 329-330 (1977).
Obviously, the analysis would be different if it were found that the dearth of qualified nonwhite applicants was due to practices on petitioners’ part which — expressly or implicitly — deterred minority group members from applying for noncannery positions. See, e. g.. Teamsters v. United States, supra, at 365.
We qualify this conclusion — observing that it is only “probable” that there has been no disparate impact on minorities in such circumstances — because bottom-line racial balance is not a defense under Title VII. See Connecticut v. Teal, 457 U. S. 440 (1982). Thus, even if petitioners could show that the percentage of selected applicants who are nonwhite is not significantly less than the percentage of qualified applicants who are nonwhite, respondents would still have a ease under Title VII, if they could prove that some particular hiring practice has a disparate impact on minorities, notwithstanding the bottom-line racial balance in petitioners’ work force. See Teal, supra, at 450.
As we understand the opinions below, the specific employment practices were challenged only insofar as they were claimed to have been responsible for the overall disparity between the number of minority cannery and noncannery workers. The Court of Appeals did not purport to hold that any specified employment practice produced its own disparate impact that was actionable under Title VII. This is not to say that a specific practice, such as nepotism, if it were proved to exist, could not itself be subject to challenge if it had a disparate impact on minorities. Nor is it to say that segregated dormitories and eating facilities in the workplace may not be challenged under 42 U. S. C. § 2000e — 2(a)(2) without showing a disparate impact on hiring or promotion.
Of course, petitioners' obligation to collect or retain any of these data may be limited by the Guidelines themselves. See 29 CFR § 1602.14(b) (1988) (exempting “seasonal” jobs from certain recordkeepingrequirements).