dissenting.
The majority affirms the district court’s grant of summary judgment on the apparent theory that Hazen Paper Co. v. Biggins, — U.S. —, 113 S.Ct. 1701, 123 L.Ed.2d 338 (1993), precludes the use of the disparate impact theory of liability under the ADEA. The disparate impact theory as applied here would concern the school’s employment policy in refusing to depart from its practice of raising salaries for years of experience even when an older and more experienced teacher is willing to work for less. Such a policy disproportionately burdens older workers. I say “apparent theory” because the majority stops ever so slightly short of announcing this conclusion with perfect clarity. But that is the unmistakable import of the majority approach.
The majority’s analysis begins with the premise that the decision not to hire Johnson was not based on misperceptions about the competence of older workers. If this characterization is accurate, then the majority is correct in saying that Hazen Paper precludes ADEA liability. My difficulty with the majority’s approach is that its analysis begins with its conclusion: that the decision to pass Johnson by had nothing to do with stereotypical views of older workers. But, as I understand the use of disparate impact analysis in the age discrimination context, one of its important purposes is to answer this very question. Hence, disparate impact analysis should be allowed to proceed to determine whether the refusal to hire did really arise from stereotypical views of older workers. Not to do so is to say that “overqualified” (i.e. overage) music teachers need not apply.
I.
Metz v. Transit Mix, Inc., 828 F.2d 1202 (7th Cir.1987), held that employers violate the ADEA when they make employment decisions based on factors that correlate with age in an obvious manner, like salary, the proximity of pension benefits’ vesting or gray hair. The Supreme Court’s recent opinion in *1079Hazen Paper, — U.S. —, 113 S.Ct. 1701, rejects part of that framework. Hazen Paper focuses the scrutiny of the ADEA on discrimination on the basis of age qua age. The ADEA’s purpose under Hazen Paper is to prohibit discrimination against older workers “on the basis of inaccurate and stigmatizing stereotypes.” Id. at —, 113 S.Ct. at 1706. An “employer cannot rely on age as a proxy for an employee’s remaining characteristics, such as productivity, but must instead focus on those factors directly.” Id.
So where an employer is in fact motivated by a desire to reduce salaries, it is permissible to fire higher-paid older workers and replace them with younger ones who will work for less. And perhaps if the employer actually has an aversion to gray hair, she can make her employment decisions accordingly. But, even so, Hazen Paper does not spell the end of the ADEA. It remains unlawful to invoke pretextually an ostensibly neutral factor that tends to correlate with age, while actually laboring under forbidden “inaccurate and stigmatizing stereotypes.” Id. at— —, 113 S.Ct. at 1706-07 (“We do not preclude the possibility that an employer who targets employees with a particular pension status on the assumption that these employees are likely to be older thereby engages in age discrimination.” Id. at --, 113 S.Ct. at 1707).
A1 of this tells us about the underlying theory of ADEA liability, not the method of proof. Even before Hazen Paper there was substantial disagreement about the permissibility of premising ADEA liability on a showing of disparate impact. See Markham v.
Geller, 451 U.S. 945, 101 S.Ct. 2028, 68 L.Ed.2d 332 (1981) (Rehnquist, J., dissenting from denial of certiorari).1 And while the three concurring justices in Hazen Paper believed that “there are substantial arguments that it is improper to carry over disparate impact analysis from Title VII to the
ADEA,” — U.S. at-, 113 S.Ct. at 1710
(Kennedy, J., concurring), there is no suggestion that Hazen Paper itself answers the question.2
So the question presented by this appeal is whether a plaintiff may endeavor to prove that an employer discriminated on the basis of age qua age by implementing a particular employment practice that disproportionately burdens older workers.
II.
The answer to that question depends mostly on what one thinks are the. purposes of disparate impact liability. Implicit in the majority’s approach is the view that employers are held liable under a disparate impact theory even where the practice or policy they have implemented isn’t really discriminatory. As I have indicated, the employment policy at issue in this case is the school’s refusal to depart from its policy of escalating salaries in relation to years of experience, where an older and more experienced teacher is willing to work for less. .
The majority says that “there is no allegation that Francis Parker’s salary system is a subterfuge for its belief that older teachers are less effective than younger teachers.” But it would not appear to me that such a conscious and invidious scheme would need *1080to be alleged in order to state a claim under the ADEA.
The basic prohibition of the ADEA makes it unlawful to “discriminate against any individual ... because of such individual’s age.” ADEA § 4(a)(1), 29 U.S.C. § 623(a). This language mirrors the anti-discrimination provision of Title VII.3 In the Title VII context, the disparate impact method of proving “discrimination ... because of [membership in a protected class]” is well-established (codified, in fact, by the Civil Rights Act of 1991). See Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971).
The Griggs disparate impact method recognizes that not all discrimination is apparent and overt. It is sometimes subtle and hidden. It is at times hidden even from the decisionmaker herself, reflecting perhaps subconscious predilections and stereotypes. See Charles Lawrence. The disparate impact method therefore requires employers to determine which of their employment practices and policies burden a protected class in a disproportionate way.
But such practices need not necessarily be abandoned.' They are nonetheless permissible, despite their disparate impact, where they are supported by a “business necessity.” The point of that defense is to rebut the inference of discrimination (even unconscious discrimination) that the disparate impact tends to demonstrate. If business necessity is shown, we can assume that the practice in question was established because of that necessity, not merely as a product of stereotyping.
Of course, in the equal protection context the requirements for proving discrimination are far stricter. There, Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976), requires precisely the sort of smoking gun evidence of conscious discriminatory motive and purpose that Griggs’ more layered approach obviates.4 Today’s majority essentially holds ADEA plaintiffs to a Washington v. Davis standard. I see no reason for such a departure.
III.
The majority offers two reasons for its result. The first comes from ADEA § 4(f)(1), 29 U.S.C. § 623(0(1). That provision provides an affirmative defense to employers for actions taken “where ... the differentiation is based on reasonable factors other than age.” Judge Easterbrook, dissenting in Metz, suggested that this language precluded the availability of disparate impact liability in ADEA cases. Metz, 828 F.2d at 1212-13. And today’s majority agrees.
But it seems to me that the disparate impact method already provides a defense for factors that are determined to be based on reasonable factors other than age — the business necessity defense. While as a general matter, it is true that we ought to interpret statutes to avoid rendering language superfluous, it seems clear to me that § 4(f)(1) simply codifies the business necessity defense. It does not preclude the availability of disparate impact liability.
The second reason the majority offers to support its conclusion is the Supreme Court’s decision in Hazen Paper. But I believe that the disparate impact theory of liability is designed as a means to detect employment decisions that reflect “inaccurate and stigmatizing stereotypes,” — U.S. at —, 113 *1081S.Ct. at 1706. This is precisely the determination that Hazen Paper says the ADEA is intended to outlaw. I do not believe its use is in any way incompatible with ADEA liability. The basic practical difficulty with the majority's result is that it provides an opportunity for employers to exclude older applicants from lower-level jobs simply by declaring the applicants “overage” (i.e. entitled to earn an excessive salary for the job they seek.).
. "The courts and the EEOC have indicated that' the impact analysis of Griggs v. Duke Power Co. is applicable to the ADEA.” Mack A. Player, Employment Discrimination Law, § 6.10, at 525 (1988). See Geller v. Markham, 635 F.2d 1027 (2d Cir. 1980), cert. denied, 451 U.S. 945, 101 S.Ct. 2028, 68 L.Ed.2d 332 (1981); Maresco v. Evans Chemetics, 964 F.2d 106, 115 (2d Cir. 1992); EEOC v. Westinghouse Electric Corp., 725 F.2d 211 (3d Cir. 1983), cert. denied, 469 U.S. 820, 105 S.Ct. 92, 83 L.Ed.2d 38 (1984); Abbott v. Federal Forge, Inc., 912 F.2d 867, 872 (6th Cir. 1990); Wooden v. Board of Educ., 931 F.2d 376, 379 (6th Cir.1991); Leftwich v. Harris-Stowe State College, 702 F.2d 686 (8th Cir.1983); EEOC v. Borden’s Inc., 724 F.2d 1390 (9th Cir. 1984); Shutt v. Sandoz Crop Protection Corp., 934 F.2d 186, 188 (9th Cir.1991); Allison v. Western Union Tel. Co., 680 F.2d 1318 (11th Cir.1982); Mac-Pherson v. University of Montevallo, 922 F.2d 766, 770-71 (11th Cir.1991).
. Commentators on the subject have also offered divergent views. See Note, Age Discrimination and the Disparate Impact Doctrine, 34 Stan. L.Rev. 837 (1982) (theory unavailable); Note, Disparate Impact and the Age Discrimination in Employment Act, 68 Minn.L.Rev. 1038 (1984) (theory available); Steven J. Kaminshine, The Cost of Older Workers, Disparate Impact, and the Age Discrimination in Employment Act, 42 Fla. L.Rev. 229 (1990) (purporting to offer "a more balanced approach”).
. The majority takes issue with the suggested parallel between Title VII and the ADEA. As noted in footnotes 1 and 2, this parallel cannot be characterized as novel. Title VII’s appeal as persuasive authority in ADEA cases is not altered by the fact that the present case deals with "applicants” for employment. Whether or not subsection (2) of the unlawful practices section excludes applicants for employment is not as self-evident as the majority maintains. 29 U.S.C. § 623(a)(2). See Getter, 635 F.2d 1027 (2d Cir. 1980) (recognizing a disparate impact claim under tire ADEA for an applicant in a position parallel to that of Johnson). There is also a good argument for the availability of disparate impact analysis under subsection (1). 29 U.S.C. § 623(a)(1). See Colby v. J.C. Penney Co., Inc., 811 F.2d 1119, 1127 (7th Cir.1987) (holding that plaintiff had stated a valid disparate impact claim under subsection (1) of Title VII’s unlawful practices section, 42 U.S.C. § 2000-e-2).
. An exception, however, is recognized for jury selection challenges. The "impact-inference" standard applicable in that setting, see Casteneda v. Partida, is functionally indistinguishable from the type of disparate impact analysis set out in Griggs.