Opinion for the court filed by Circuit Judge BUCKLEY.
Dissenting opinion filed by Senior Circuit Judge GARTH.
BUCKLEY, Circuit Judge:The United States Postal Inspection Service appeals from a decision of the district court holding that it discriminated against a class of older postal inspectors in violation of the Age Discrimination in Employment Act. The district court ruled that a single element of the Postal Service’s Career Path Policy constituted illegal age discrimination by mandating that the most senior postal inspectors transfer to any of *996fourteen chronically understaffed offices across the country. Arnold, v. Postmaster General, 667 F.Supp. 6 (D.D.C.1987). We reverse.
I. Background
A. Legal Framework
The Age Discrimination in Employment Act, 29 U.S.C. §§ 621-84 (1982) (“ADEA”), “broadly prohibits arbitrary discrimination in the workplace based on age.” Lorillard v. Pons, 434 U.S. 575, 577, 98 S.Ct. 866, 868, 55 L.Ed.2d 40 (1978). Specifically, it provides:
It shall be unlawful for an employer—
(1) to fail or refuse to hire or to discharge any individual or otherwise discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s age;
(2) to limit, segregate, or classify his employees 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 age.
29 U.S.C. § 623(a) (1982).
In applying the terms of the statute to particular cases, courts have drawn upon Supreme Court decisions detailing the twin methods of proof in race or gender discrimination cases under Title VII of the Civil Rights Act of 1964. Under traditional Title VII analysis, a plaintiff may establish that he was the victim of disparate treatment by introducing sufficient evidence to establish a prima facie case of discrimination. The defendant must then come forward and articulate some legitimate, non-discriminatory reason for his actions. Provided the defendant does so, the plaintiff must establish that the defendant’s proffered reason is a mere pretext for discrimination. Texas Dep’t of Community Affairs v. Burdine, 450 U.S. 248, 252-53, 101 S.Ct. 1089, 1093, 67 L.Ed.2d 207 (1981); McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802-04, 93 S.Ct. 1817, 1824-25, 36 L.Ed.2d 668 (1973). Throughout the entire process, the burden of persuasion remains with the plaintiff and proof of discriminatory intent is critical. Burdine, 450 U.S. at 248, 101 S.Ct. at 1089.
Alternatively, a Title YII plaintiff may recover by establishing that a particular employment practice, while neutral on its face, has a disparate impact on the protected class. Griggs v. Duke Power Co., 401 U.S. 424, 432, 91 S.Ct. 849, 854, 28 L.Ed.2d 158 (1971) (Title VII directed to “consequences of employment practices, not simply the motivation”) (emphasis original). Proof of discriminatory intent is not necessary to prevail on a disparate impact claim. The plaintiff must simply establish, generally through the use of statistical evidence, that the challenged practice has a disparate impact on the protected class. Once this is accomplished, the burden shifts to the defendant to supply evidence that the practice is a “business necessity.” Id. at 432, 91 S.Ct. at 854. The plaintiff may still prevail if he can prove that a non-discriminatory alternative was available.
In Johnson v. Lehman, 679 F.2d 918, 921-22 (D.C.Cir.1982), one of our early forays into ADEA litigation, we expressly held that the disparate treatment mode of analysis is applicable to an age discrimination claim under ADEA. See also Stacey v. Allied Stores Corp., 768 F.2d 402, 407 (D.C.Cir.1985). We have never determined whether a plaintiff bringing suit under ADEA may prevail on a disparate impact theory.
B. The Career Path Policy
The United States Postal Inspection Service employs more than 1,900 postal inspectors across the United States. As the law enforcement branch of the service, it is responsible for “investigatpng] offenses and civil matters relating to the Postal Service.” 39 U.S.C. § 404(a)(7) (1982). The Postal Inspection Service is directed by the Chief Postal Inspector, who in turn reports to the Postmaster General.
In response to a shortage of experienced postal inspectors in fourteen major metropolitan areas (“MMAs”), the Postal Inspection Service instituted the Career Path Poli*997cy (“CPP”) on December 1, 1980. Appendix for Appellant (“App.”) at 82-92. The goal of the policy was to ensure that these areas would be staffed by a constant supply of experienced postal inspectors, and “to guide inspectors along a career path which provides them experience and exposure.” Id. at 88.
Generally, postal inspectors are categorized according to certain experience levels, ranging from level 17 to level 24. Postal inspectors ordinarily spend one year as a level 17 inspector, one year at level 19, and two years at level 21 before being promoted to level 23. By the time they reach level 23, postal inspectors will have had experience in each of the four basic investigation-areas. Level 24 inspectors are Team Leaders, who supervise groups of inspectors from each of the other service levels. The Inspection Service is authorized by statute to transfer any postal inspector, from any level, at any time. 39 U.S.C. § 1001(e)(2) (1982).
The Career Path Policy was established as a complement to this basic framework. A central objective of the CPP, made explicit in 1984, is to ensure that all level 23 inspectors spend five years in an MMA during the course of their service. App. at 88 (“It is Inspection Service policy that normally all Inspectors will serve at least 5 years in a major metropolitan area as a Level 23/24.”).
The CPP establishes three methods for achieving that objective. First, level 23 postal, inspectors are encouraged to apply for lateral reassignments to vacancies in an MMA. If more than one level 23 postal inspector bid for a given vacancy, the first to bid is awarded the assignment. Second, if a vacancy is not filled by a voluntary level 23 lateral assignment, the Postal Service advertises the vacancy among level 21 postal inspectors, who may volunteer for lateral reassignment as level 23 inspectors. Finally, if the vacancy is not filled through either of the first two methods, the Postal Inspection Service will refer to a list of current level 23 inspectors and choose the most senior, in terms of service, for mandatory reassignment. Certain postal inspectors are expressly exempted from this final category, including those eligible to retire within five years, those who relocated within the past two years, and those who already served for two years in an MMA. After five years in an MMA, any inspector may ask to be reassigned.
C. The Suit
In September 1982, two years after the CPP was first unveiled, a postal inspector named Charles Netherton filed a class action on behalf of himself and other similarly situated inspectors in the United States District Court for the Middle District of Florida. See generally Arnold v. Postmaster General, 667 F.Supp. 6, 12-13 (D.D.C.1987) (“Arnold II”) (outlining procedural history). Netherton claimed that the CPP’s provision for the directed transfer of senior postal inspectors had a disparate impact on inspectors aged forty and over in violation of the ADEA.
In August 1985, two additional groups of postal inspectors filed suit in the District of Columbia. The cases were consolidated on September 16, 1986, after the Netherton suit had been transferred to the District of Columbia. In December 1986, the United States District Court for the District of Columbia held that the plaintiff class could proceed under a theory of disparate treatment as well as disparate impact and denied cross-motions for summary judgment. Arnold v. United States Postal Service, 649 F.Supp. 676 (D.D.C.1986) (“Arnold I”). Finally, by separate order, the district court granted the Inspection Service’s motion to bifurcate the trial into separate liability and damages phases.
Following a three-day bench trial, the court issued its opinion in favor of the postal inspectors. First, the district court held that the third aspect of the CPP— which it termed the “senior-first rule”— had a disparate impact on level 23 employees over the age of forty. Arnold II, 667 F.Supp. at 19-26. It accepted appellees’ statistical proof that the senior-first rule “falls more harshly” on this protected class. Id. at 19. After making this initial finding, the court concluded that appellant *998had failed to demonstrate that the discriminatory rule could be justified as a business necessity because it “failed to carry its burden of persuading the Court that it had no other choice but to transfer the senior and older postal inspectorsId. at 24 (emphasis original). Finally, the court concluded that even if the defendant had a business necessity to transfer the most senior level 23 inspectors, it could have accomplished this purpose without discriminating against older employees by using a random transfer system. Id. at 25-26.
Having concluded that the senior-first rule had a disparate impact on older postal inspectors, the court further found that the Postal Service knew of the correlation between age and seniority, and had therefore intentionally discriminated against level 23 postal inspectors over the age of forty. Id. at 26-28. Specifically, the court held that: (1) appellees’ statistical proof and anecdotal evidence raised “far more than an inference of age discrimination,” id. at 27; (2) the Postal Inspection Service did not sustain its burden of producing evidence of a legitimate non-discriminatory reason for the senior-first rule; and (3) any non-discriminatory reason could only have been a pretext for age discrimination because the Postal Inspection Service could instead have used a random transfer system.
Leaving the question of damages to a separate proceeding, the district court enjoined appellant from further reliance on the senior-first aspect of the CPP. Id. at 29.
II. DISCUSSION
A. Disparate Impact
1. Disparate Impact Under the ADEA
In Griggs, 401 U.S. at 430, 91 S.Ct. at 853, the Supreme Court first held that a Title VII plaintiff can establish race discrimination by demonstrating that a neutral employment practice had a disparate racial impact. Although the Court “has never held that proof of discriminatory impact can establish a violation of the ADEA,” Markham v. Geller, 451 U.S. 945, 948, 101 S.Ct. 2028, 2030, 68 L.Ed.2d 332 (1981) (Rehnquist, J., dissenting from the denial of certiorari), the district court followed the lead of a number of our sister circuits and held that the disparate impact theory is applicable to this case. See, e.g., EEOC v. Borden’s, Inc., 724 F.2d 1390, 1394-95 (9th Cir.1984); Leftwich v. Harris-Stowe State College, 702 F.2d 686, 690 (8th Cir.1983); Geller v. Markham, 635 F.2d 1027, 1032 (2d Cir.1980), cert. denied, 451 U.S. 945, 101 S.Ct. 2028, 68 L.Ed.2d 332 (1981). Cf. Metz v. Transit Mix, Inc., 828 F.2d 1202, 1216-22 (7th Cir.1987) (Easter-brook, J., dissenting).
We do not decide whether the test is applicable to ADEA cases because we conclude that the CPP does not have a disparate impact on postal inspectors who are forty years of age and older.
2. Impact of the CPP on Older Inspectors
The central element of the district court’s disparate impact holding was its decision to focus solely on the “senior-first” aspect of the CPP, and to exclude from consideration those postal inspectors who were transferred under the two complementary aspects of the program. The court explained:
Plaintiffs, however, have made it clear since this lawsuit was filed that they challenge the discriminatory senior-first rule of the Career Path Policy. It is defendant who has tried to mischaracterize plaintiffs’ challenge to encompass the entire Career Path Policy in order to dilute the adverse impact that the senior-first rule has on plaintiffs. Plaintiffs have never challenged the aspects of the Policy permitting postal inspectors to volunteer for lateral reassignment.
Arnold II, 667 F.Supp. at 19-20 (citations omitted).
By so limiting the scope of its scrutiny, the district court effectively determined the outcome of the case. Appellant does not dispute the district court’s finding that the senior-first transfer rule, considered alone, had a disproportionate impact on individuals over forty. Instead, appellant argues that the impact of the CPP must be viewed *999in all its facets, including the voluntary transfer provisions. Considered in this light, appellant contends that the CPP did not have a disparate impact on the protected class. According to statistics presented by appellant’s expert, inspectors aged forty and above constituted 34.4 percent of all level 23 inspectors eligible for transfer into an MMA, and constituted 33.8 percent of all inspectors actually transferred, either voluntarily or involuntarily. App. at 101, 113. These statistics illustrate the distortions that will result if an analysis of the CPP’s impact is limited to just one of its elements.
Appellees argue that the narrow focus adopted by the district court is mandated by Connecticut v. Teal, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982). That case involved a Connecticut state agency’s requirement that candidates for the position of Welfare Eligibility Supervisor pass a written examination that was not shown to be job related. The Court held that because a disproportionate number of blacks had failed the test, it could not be used even though more than a proportionate number of blacks had actually been promoted to that position as a result of the agency’s application of an affirmative action program. The Court wrote that such a focus on “the bottom line” is improper because Title VII protects individual employees. Id. at 455, 102 S.Ct. at 2534-35. The employer could not use the non-discriminatory later stages of the promotion process to temper the effects of the written examination at the first stage, because those who failed the examination never received consideration at the later stages.
We think that Teal is inapposite, and we agree with appellant that the district court erred by focusing solely on the senior-first rule. The central distinction between Teal and the current case is that, unlike the pass-fail test at issue in Teal, the senior-first rule was not a free-standing element of an employment program. Under the CPP, every level 23 employee was expected to spend five years in an MMA. Thus each level 23 employee can fulfill this uniform service requirement through either an early voluntary or a later involuntary transfer. Those individuals who are subject to involuntary transfers, and protected by the ADEA, were not excluded from the voluntary transfer stage. While the impact of an unfair test on the individual plaintiffs in Teal has nothing to do with the fact that other blacks were ultimately promoted, the impact of the CPP on senior postal inspectors can be fairly measured only by taking into account those who, subject to an identical obligation, elected to fulfill their responsibility by volunteering for transfer. Our decision to view the CPP transfer plan as a whole does not involve using the nondiscriminatory effects of voluntary transfers to answer the complaint of one who did not have the opportunity to undertake such a transfer. Rather it reflects a realistic assessment of the transfer options available alike to each and every level 23 employee.
Because the district court failed to consider the impact of the senior-first transfer rule within the context of the CPP as a whole, its finding that that rule had a disparate impact on older employees is legally irrelevant. This is so because the court based its factual finding on an erroneous understanding of the perspective appropriate to a disparate impact analysis in this case, assuming such an analysis is applicable to ADEA cases. Thus notwithstanding the court’s contrary finding, we conclude that appellees failed to establish that the CPP, taken as a whole, had a disparate impact on older postal inspectors.
B. Disparate Treatment
Noting that we have previously held that a plaintiff may establish a prima facie case of discrimination with statistics alone, see Palmer v. Schultz, 815 F.2d 84, 90 (D.C.Cir.1987), the district court relied on its statistical analysis of the senior-first rule to establish the requisite inference of age discrimination under a disparate treatment analysis. Arnold II, 667 F.Supp. at 27.
The district court then found that appellant had not only failed to proffer a non-discriminatory reason for the senior-first rule, but that the rule was “a pretext to mask its *1000discriminatory animus and motive.” Id. at 27-28. In so finding, the court relied primarily on anecdotal evidence that appellant was aware of the correlation between age and seniority and had rejected a system of random selection as an alternative even though it had been advised that the senior-first rule might be found in violation of the ADEA.
We disagree. There may well be cases in which seniority is simply a code word for age discrimination. Because the senior-first rule does not stand alone, however, it is neither fair nor reasonable to judge the motives of the Inspection Service by uncoupling the rule from the other aspects of what was clearly intended to be an integrated program. We emphasize again: The CPP established that all eligible level 23 inspectors, young or old, should expect to serve five years in an MMA. Viewed in this context, the purpose of the senior-first rule was not to require that the oldest postal inspectors be transferred, but to establish a rational and orderly system to ensure that all level 23 inspectors serve a tour of duty in an MMA prior to retirement.
In this context, we find the Inspection Service’s refusal to substitute a system of random selection for its senior-first rule to be both rational and devoid of any implication of animus. Reliance on a lottery would have reduced incentives for voluntary transfers by encouraging chance-takers to gamble on the possibility that they could reach retirement without ever having contributed the five years of MMA service expected of all level 23 inspectors. We reject the district court’s conclusion that the refusal or the lottery alternative proves a discriminatory intent, 667 F.Supp. at 28, because based, as it is, on an incorrect legal premise, it is clearly erroneous. The implications of the employer’s awareness of the senior-first plan’s effects can only be assessed within the context of the larger, comprehensive program of which it is an integral part.
Viewed from this perspective, and applying the three-pronged analytical framework that governs our disposition of disparate treatment cases, we conclude that appellees failed to establish a violation of the ADEA., First, the evidence supplied by appellees did not establish a prima facie case of discrimination. While we have held that statistical proof may be sufficient to establish discriminatory intent, this is only true where the statistics focus on the appropriate labor pool. Palmer, 815 F.2d at 91 n. 6. Second, appellant provided ample evidence of the non-discriminatory reasons that led to the design of the three elements of the CPP. The district court erred in concluding that appellant had failed to rebut the inference of discrimination because of its focus on only the third of those elements. Finally, appellees did not successfully establish that, in this case, appellant’s reliance on seniority as a method of selection was simply a pretext for age discrimination. While the appellant apparently realized that the senior-first rule would result in the mandatory transfer of older employees, it relied on the rule as an appropriate mechanism for enforcing an obligation that was both universal and neutral.
III. Conclusion
Appellees failed to establish that the Postal Inspection Service discriminated against older postal inspectors in violation of the ADEA. Accordingly, the judgment of the district court is
REVERSED.