dissenting.
The majority concludes that the disability-retirement plan of the Jefferson County Sheriffs Department, Kentucky Retire*584ment Systems, and the Commonwealth of Kentucky (collectively referred to as “KRS”) amounts to a facial violation of the Age Discrimination in Employment Act (“ADEA”), 29 U.S.C. §§ 621 et seq., as amended by the Older Workers Benefits Protection Act (“OWBPA”), P. Law 101-433, 104 Stat. 978 (1990). I believe that a careful examination of the plan shows that it considers age only in combination with years of service and years to retirement age, and is a non-discriminatory way of providing workers with protection against disability before they have had an opportunity to earn a normal pension at retirement age. It therefore is not illegal under the Supreme Court’s precedent in Hazen Paper Co. v. Biggins, 507 U.S. 604, 113 S.Ct. 1701, 123 L.Ed.2d 338 (1993), and I respectfully dissent.
Congress stated that the purpose of the ADEA is to “promote employment of older persons based on their ability rather than age; to prohibit arbitrary age discrimination in employment; to help employers and workers find ways of meeting problems arising from the impact of age on employment.” 29 U.S.C. § 621(b). 29 U.S.C. § 623(a)(1) prohibits an employer from, inter alia, “discriminatpng] against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s age.”1
The Supreme Court has noted that the ADEA “broadly prohibits arbitrary discrimination in the workplace based on age.” Trans World Airlines, Inc. v. Thurston, 469 U.S. 111, 120, 105 S.Ct. 613, 83 L.Ed.2d 523 (1985) (internal quotation marks and citation omitted). In discussing the ADEA in Hazen Paper, the Supreme Court emphasized the distinction between the disparate treatment and disparate impact theories of employment discrimination. 507 U.S. at 609, 113 S.Ct. 1701. The Court stated that under a disparate treatment theory, “[t]he employer simply treats some people less favorably than others because of their race, color, religion [or other protected characteristics.] Proof of discriminatory motive is critical, although it can in some situations be inferred from the mere fact of differences in treatment. ...” Ibid, (second alteration and ellipsis in original) (internal quotation marks omitted) (quoting Teamsters v. United States, 431 U.S. 324, 335-36 n. 15, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977)).
The Court continued that “[i]n a disparate treatment case, liability depends on whether the protected trait (under the ADEA, age) actually motivated the employer’s decision.” Hazen Paper, 507 U.S. at 610, 113 S.Ct. 1701. The employer “may have relied upon a formal, facially discriminatory policy requiring adverse treatment of employees with that trait,” ibid, (citing Thurston, supra, and Los Angeles Dep’t of Water & Power v. Manhart, 435 U.S. 702, 704-18, 98 S.Ct. 1370, 55 L.Ed.2d 657 (1978)), or “may have been motivated by the protected trait on an ad hoc, informal basis.” Ibid. ‘Whatever the employer’s decisionmaking process,” the Court noted, “a disparate treatment claim cannot succeed unless the employee’s protected trait actually played a role in that process and had a determinative influence on the outcome. ” Ibid, (emphases added).
Defined in that way, the Court added, disparate treatment “captures the essence of what Congress sought to prohibit in the ADEA. It is the very essence of age discrimination for an older employee to be *585fired because the employer believes that productivity and competence decline with old age.” Ibid. “Congress’ promulgation of the ADEA was prompted by its concern that older workers were being deprived of employment on the basis of inaccurate and stigmatizing stereotypes.” Ibid. This type of stereotyping is nowhere found in the plan under consideration today.
In Hazen Paper, the Court addressed the question of whether an employer’s firing an employee whose pension was shortly to vest violated the ADEA where the vesting of the pension was based exclusively on years of service. The Court held that although pension status is typically correlated with age, and although it is perhaps true that “older employees of Hazen Paper are more likely to be ‘close to vesting’ than younger employees,” age and years of service were nevertheless analytically distinct. If the employer fired the employee because of pension status, not because of age, “[t]he prohibited stereotype (‘Older employees are likely to be -’) would not have figured in this decision, and the attendant stigma would not ensue.” Such conduct would not violate the ADEA, the Court held. 507 U.S. at 611-12, 113 S.Ct. 1701. (The Court ultimately remanded for a determination of whether the firing had in fact been based on age rather than on years of service. Id. at 613, 113 S.Ct. 1701.) In short, the Supreme Court held that, under the ADEA, companies could make a decision based solely on the need to save money, even if that decision bore more heavily, on average, on older workers, so long as the factor relied on was only correlated with age, not determined by age. In our case, with much less base motives, the KRS plan is impacted by age only in relation to years of service and years remaining until normal retirement age, as shown by the examples given on pages 16 and 17, infra.
Thurston, an ADEA case cited by the Supreme Court in Hazen Paper as one in which the employer’s formal, facially discriminatory policy showed intent to discriminate against older employees because of their age, fits comfortably within the “very essence” of the ADEA as discussed by the Court in Hazen Paper. In Thurston, TWA had adopted a policy by which captains who were disqualified from serving in that capacity for reasons other than age were allowed to transfer to the position of flight engineer, and in the process to “bump” less senior flight engineers. Pilots who were going to be disqualified from continuing to serve as captains because they had reached the age of 60, however, had to resort to bidding procedures in order to become a flight engineer, and if the procedures did not result in a flight engineer position, the captain had to retire at 60. 469 U.S. at 115-17, 120, 105 S.Ct. 613. Under the TWA policy, the following results would obtain. Captain A is disqualified from continuing to serve as captain because he is found to be incompetent. Captain B is nearing age 60. They are otherwise similarly situated. Captain A is allowed to transfer to the position of flight engineer and bump less-senior flight engineers in the process, id. at 117, 105 S.Ct. 613; Captain B is not. The Court found that the policy, under which “the method of transfer available to a disqualified captain depends upon his age,” was “discriminatory on its face,” and therefore amounted to direct evidence of age discrimination in violation of the ADEA. Id. at 121-22, 105 S.Ct. 613. The Thurston case is thus an example of a forbidden policy that, in fact, implicates a stigmatizing and inaccurate stereotype-that pilots over age 60 are less capable, or at least less valuable as employees, even than younger workers who have been relieved for incompetence.
TWA’s policy clearly embodied the “essence” of what the ADEA sought to pro*586hibit-that is, arbitrary age discrimination. That is not the case with the KRS disability-retirement plan, nor was it the case with the early-retirement policy in Lyon v. Ohio Education Ass’n and Prof'l Staff Union, 53 F.3d 135 (6th Cir.1995). Here, and in Lyon, no such stereotype is implicated.
The majority contends that there are two ways in which the KRS plan facially discriminates against older employees. First, workers with less than 20 years of service time (for employees in hazardous positions2) who become disabled before a certain age (in the case of employees in hazardous positions, 55) receive additional credit toward retirement, while those over normal retirement age do not receive additional credits (that is, their years to normal retirement age is zero). Second, a disabled worker with fewer years remaining until retirement age may receive fewer additional credits than a worker with more years remaining.3 But the calculations for determining disability benefits for workers who become disabled under age 55 and for normal retirement benefits for those over age 55 are the same, with the exception of the imputed years added to the former’s service time.
The original panel succinctly describes how the KRS plan works:
When workers are disabled after they become eligible for normal retirement, they receive only normal retirement benefits. The amount of the yearly benefits is generally calculated as 2.5% of the employee’s final compensation times the number of years worked. However, for employees who are not yet eligible for normal retirement (i.e., employees under age 55 and with fewer than 20 years of service), additional years are added to the number of years worked for purposes of the calculation. The number of years added is the number of years remaining until the worker would have reached either normal retirement age or twenty years of service, but no more than the number of years already worked.
EEOC v. Jefferson County Sheriff’s Dep’t, 424 F.3d 467, 469 (6th Cir.2005) (footnote omitted), vacated on grant of reh’g en banc, 2006 U.S.App. LEXIS 258 (6th Cir. Jan. 4, 2006). The majority argues that because older employees with, e.g., 10 years of service and final pay of $50,000 receive fewer benefits than younger employees with the same years of service and final pay, the KRS plan is facially discriminatory. Yet the majority misses the point that a 53-year-old employee who becomes disabled is not similarly situated to a 33-year-old employee who becomes disabled, even if they have the same service years and final pay at the time of disability. All else being equal, the non-disabled 33-year-old of course has more years to work and live than does a non-disabled 55-year-old. See Lyon, 53 F.3d at 137, 140-41 (explaining that an early-retirement plan that “ensure[s] early retirees the same benefits that they would have received had they continued to work until their normal retirement date” did not pay older workers *587lower benefits because of age; two employees of different ages but equal years of service are not similarly situated). Here, the number of years of additional work credit lost is a factor related to, but not determined by, age.
In Public Employees Retirement System v. Betts, 492 U.S. 158, 109 S.Ct. 2854, 106 L.Ed.2d 134 (1989), the Supreme Comb confronted a disability-retirement plan that provided age-and-service retirement benefits to those who retired over a certain age (for Betts, the plaintiff, the retirement age was 60) and had a certain number of service years, or to those under the age who had served a set higher number of years. The plan also provided disability benefits to those under the retirement age who had served a certain number of years and suffered a disability. 492 U.S. at 162, 109 S.Ct. 2854. The scheme in Betts provided that disability payments would constitute not less than 30% of the disability retiree’s final average salary; no such floor existed for age- and-service retirees. Betts became disabled at age 61, and therefore was unable to receive disability benefits. She received in age-and-service benefits just over half what she would have received in disability. Id. at 163, 109 S.Ct. 2854.
The Supreme Court stated that “[o]n its face, the ... scheme renders covered employees ineligible for disability retirement once they have attained age 60.” Yet the Court found that the scheme did not violate the ADEA because it fit under the then-existing provision of 29 U.S.C. § 623(f)(2) that exempted age-based decisions taken pursuant to the terms of “any bona fide employee benefit plan such as a retirement, pension, or insurance plan, which is not a subterfuge to evade the purposes of’ the ADEA. Id. at 161-62, 166, 109 S.Ct. 2854. A year after the Supreme Court’s decision in Betts, Congress passed the OWBPA, amending the ADEA. The OWBPA included a finding that “as a result of [Betts], legislative action is necessary to restore the original congressional intent in passing and amending the [ADEA], which was to prohibit discrimination against older workers in all employment benefits except when age-based reductions in employee benefit plans are justified by significant cost considerations.” OWBPA, Pub.L. No. 101-433, § 101, 104 Stat. 978 (1990) (codified at 29 U.S.C. § 621 note).
In contradistinction to the plan at issue in Betts, the case the OWBPA was designed to overturn, the KRS plan does not provide younger workers with a specific benefit unavailable to older workers. In Betts, the crucial distinction was that a “disability” retirement gave a worker a guaranteed minimum income of 30% of the worker’s final salary. Betts was disqualified from receiving that benefit solely on the basis of age, though she was qualified in every other respect. Betts, 492 U.S. at 162-63, 109 S.Ct. 2854.
Under the KRS plan, a “disability” retirement is intrinsically no different from the “normal” retirement pension available to every worker. Benefits are based on years of credited service (augmented to a maximum of 20 for disabled workers ineligible for normal retirement), multiplied by a factor related to final salary. If “normal” benefits are greater than the augmented benefits for disability, of course, the greater benefits are provided. The plan simply provides that a worker who is disabled before reaching eligibility for normal retirement benefits has a way of receiving a retirement benefit equal to (or closer to) what he would have received had he not become disabled before reaching the normal retirement age or 20 years of service.
*588This is a very reasonable benefit, and one that a younger worker, in particular, would be more likely to value more highly than a newly-hired older worker, who would be more likely to have acquired other benefits from more extensive earlier employment. The extensive questioning by this court at oral argument, attempting to probe for a way to provide this sensible result under the appellants’ reasoning, shows that we should not be too quick to assume that Congress intended, or commanded by the language it wrote, that such a reasonable plan was being outlawed.
It is worth repeating that the plaintiff here, Mr. Lickteig, is already eligible for “normal” retirement benefits, based on his 17 years of service and his having attained the normal retirement age (55 for employees in hazardous jobs). What he (or rather, what the EEOC on his behalf) is attempting to do in this suit is to prevent employees who have been granted a form of insurance against disability at a pre-retirement age from obtaining that benefit.
Several examples show how the KRS plan does not differentiate based on age, but on age only in relation to years of service.
1) Take two employees of the same age. One is 48 with 10 years of service, the other is 48 with 15 years of service. They both become disabled: The first employee gets 7 years of extra credit, which takes him to age 55, and the second gets 5 years of extra credit, which takes him to 20 years of service, but in each case it is as if the employee had worked until eligible for normal retirement.
2) Take two employees with the same service. One is 50 with 15 years of service. The second is 35 with 15 years of service. They both become disabled. They each get exactly the same additional credit — 5 years — despite the difference in their ages, and with that additional credit, it is as if each had worked until eligible for normal retirement.
3)Take two employees with differing ages and years of service. The first is 45 with 10 years of service. The second is 40 with 17 years of service. They both become disabled. This time, the older worker actually gets a greater benefit because he will get credit for all of the 10 years that would bring him to the retirement age, whereas the younger worker will “max out” at the full 20 years of credit with only an additional 3 years of credit.
These examples demonstrate starkly that age is not a controlling variable in the operation of the KRS plan. They also show that the plan provides in practice exactly what was claimed for it by appel-lees’ counsel in the argument before us: a way to insure against disability that is sensitive to the greatest loss caused by disability — the inability to continue earning credits toward retirement at a normal retirement age.
Many life insurance policies have a feature that if premiums are paid from the inception of the policy to age 65, no further premium payments are required. Some also have a feature called “disability waiver of premiums”: if the policy holder becomes disabled, the insurance company will no longer collect premiums, in effect crediting the policyholder as though those premiums are being paid. Yet, on average, such a disability feature (which no one would contend constitutes age discrimination) works exactly like the KRS plan. The waiver feature is worth more to the person who is disabled at 40 than one who is disabled at 60, because, on average, the waiver is in effect for many more years. And one who is disabled after 65 receives no benefit at all, as the premium is already fully paid up.
*589It is undisputed, and patently obvious from the nature of the KRS plan, that both the rationale and the effect of that plan is to insure that all employees have a reasonable prospect of employer-sponsored insurance against a disability that occurs before retirement income becomes available. On average, if employee A is younger than employee B and both suffer the same disability at the same time, A 1) will have had less time to earn money and retirement credit, and 2) will have more years to live. The KRS plan is meant to, and does, ameliorate exactly the ways in which A and B are not similarly situated, by providing A with a “bonus” for the circumstances of the disability. That “bonus” depends on years of service and years to normal retirement age. The bonus will often, but not always, be larger for a younger worker than an older one. As the examples show, the bonus provided to employees who become disabled can vary from no additional credit to as much as 10 years’ credit, in a fashion that may be correlated (just as pension status in Hazen Paper was correlated) with age, but far from perfectly.
The KRS plan has nothing whatever to do with “inaccurate and stigmatizing stereotypes” surrounding the relative ability of older employees to do the job-that is to say, Congress’s reason for enacting the ADEA, as the discussion above and the majority’s own analysis of the legislative history indicate (Maj.Op. pp. 576-77). There is no intimation in any part of the KRS plan of any demeaning or inappropriate stereotyping of older workers. All employees who are considered disabled under the KRS plan are equally unable to do their jobs-because of disability. The only factor resembling consideration of the nature of aging that counsel for appellants could adduce at oral argument was that some older employees might wish to work beyond normal retirement age (55 for hazardous employees, 65 for nonhazardous employees). But that does not differentiate between workers who are now more advanced in years and those younger. All such workers may wish to (and, had they not become disabled, might have been able to) work beyond that normal retirement age. But any retirement plan must have criteria for qualification, and using age as one of the criteria has never been thought to violate the ADEA. If KRS changed the plan to have a later retirement age, the mathematics of some of the computations would change, but the aspect being attacked here would remain. Thus, it is clear that impermissible stereotyping had nothing to do with the age-correlated features of the plan that are involved here.
Under the KRS plan, the employee’s age in relation to his years of service with the employer factors into retirement benefits calculations. The majority, by dubiously labeling that plan as facially discriminatory under the ADEA, uses that statute to invalidate a policy that lies far from the “essence” of the ADEA-and, in fact, does not implicate that essence at all. Age in relation to years of service performed for this employer is not the same as age qua age. See Hazen Paper, 507 U.S. at 612, 113 S.Ct. 1701 (the ADEA “requires the employer to ignore an employee’s age (absent a statutory exemption or defense); it does not specify further characteristics that an employer must also ignore.”).
There is a further problem with the majority’s analysis. As the Lyon court noted, in reasoning that applies at least as much to this case:
[The employer’s] very willingness to ignore ageist stereotypes and hire workers of any age actually appears to have exacerbated plaintiffs “problem.” Those most “disadvantaged” are those who were oldest at the time of hiring. It would be contrary to the letter, as *590well as the spirit, of the ADEA to penalize the employer for the incidental ramifications of objectivity.
53 F.3d at 140 n. 6. Here, the older workers “discriminated against” are sometimes exactly those who were hired later in life, and thus had not accumulated as many years with this employer as a younger worker (although, as example 3 shows (supra, at 588), sometimes exactly this factor will give the later-hired older worker an advantage). Thus, I believe that neither the intent nor the letter of the ADEA bars the reasonable KRS retirement plan, and I respectfully dissent.
. " '[CJompensation, terms, conditions, or privileges of employment1 encompasses all employee benefits, including such benefits provided pursuant to a bona fide employee benefit plan.” 29 U.S.C. § 630(1).
. I follow the convention of the parties, the original panel, and the majority of this en banc court, by referring to the KRS plan for employees in hazardous positions, although the analysis applies to the plan for employees in nonhazardous positions as well.
. “May receive” for two reasons: first, as shown below at pages 16 and 17, a younger worker may only need the same number of years for maximum benefits as an older worker (e.g., in the case of a 40-year-old with 15 years of service and a 50-year-old with 15 years of service), and second, additional credit is limited to the number of years already worked. Thus, a 49-year-old with six service years and a 40-year-old with six service years will each get the same benefit-six additional years of credit.