Richard C. Gaworski, Plaintiff-appellee,equal Employment Opportunity Commission, Intervenor Plaintiff-appellee, v. Itt Commercial Finance Corp., a Nevada Corporation; Ittfinancial Corporation, a Delaware Corporation;itt Corporation, a Delaware Corporation,defendants-appellants,richard C. Gaworski, Plaintiff,equal Employment Opportunity Commission, Intervenorplaintiff-appellant, v. Itt Commercial Finance Corp., a Nevada Corporation; Ittfinancial Corporation, a Delaware Corporation;itt Corporation, a Delaware Corporation,defendants-appellees, 17 F.3d 1104 (8th Cir. 1994)

Richard Gaworski and intervenor E.E.O.C. ("plaintiffs") brought an action under the Age Discrimination in Employment Act, 29 U.S.C. § 621 et seq. (1988) ("ADEA" or "Act"),¹  against ITT Commercial Finance Corp., ITT Financial Corp., and ITT Corp. (collectively, "ITT"). Gaworski had been terminated from his position as Manager of Credit and Operations for ITT Commercial Finance Corporation's Capital Resources Group ("CRG"). At the time of his termination, he was fifty-five years old. Following a four-day trial, a jury returned a verdict finding that age had been a determining factor in ITT's discharge of Gaworski. The district court²  then entered judgment and awarded Gaworski backpay in the amount of $265,892.27. The court denied ITT's post-trial motion for judgment notwithstanding the verdict or a new trial. On appeal, ITT urges that there exists insufficient evidence to support the jury's verdict and that certain additional offsets should have been taken from the backpay award. The E.E.O.C. cross-appeals, challenging the district court's deduction from the backpay award of unemployment compensation that Gaworski received following his termination. We affirm the judgment on the verdict finding that age was a determining factor in Gaworski's discharge, but reverse in part and remand as to the backpay award.

Gaworski came to ITT in 1976 as Comptroller of ITT Industrial Credit. As Comptroller, he was responsible for the company's accounting, budgeting, financial analysis and electronic data processing. He moved to CRG after Industrial Credit merged with Commercial Finance in 1984. Two years later, ITT terminated his employment. At the time of his termination, Gaworski was the oldest and highest paid CRG employee, and he was the only CRG employee eligible to receive a pension. Defendants claim that Gaworski was terminated as part of a "reduction in force" ("RIF") initiated by Michael Guimbarda, the director of the CRG division, and that Gaworski's position was eliminated. Gaworski disputes the employer's reasons for his discharge; he urges that age was the motivating factor in his termination.

At the time of Gaworski's discharge, the number of employees at CRG dropped from seventeen to thirteen. Gaworski argues, however, that the purported RIF was not objectively carried out and that he was included among the employees to be laid off because of his age. To support his claims, Gaworski presented evidence at trial that despite Guimbarda's contention that the RIF was needed, CRG's business was increasing at the time of the layoffs. In addition, Gaworski adduced evidence that in conducting the alleged RIF, Guimbarda failed to comply with a company policy explicitly requiring "advance written notification to both the senior levels of operating management and the director of personnel" before any layoff could occur.

Gaworski also presented evidence that his position was not eliminated, as Guimbarda claimed, but that instead, he was replaced by a younger man. The day after Gaworski's employment was terminated and his position of "Manager of Credit and Operations" was supposedly eliminated, Guimbarda promoted John Olker, a 46-year-old senior investment manager whom Gaworski had supervised, to the "new" position of "Manager of Credit and Administration." Olker received an $8,000 raise, moved into Gaworski's office, and assumed substantially all of Gaworski's former duties. Gaworski claims that this promotion, too, was conducted in violation of company policy. The policy required a performance review before an employee could be promoted, and Olker had not been given such a review. At the time of Olker's previous performance review, seven months earlier, his performance was evaluated as "standard" (average), while Gaworski had generally been rated as "above standard," and it was recommended that Olker not be reassigned or promoted during the ensuing twelve-month period.

Guimbarda testified that he retained Olker over Gaworski because, although they were both good performers, Olker had greater knowledge of the computer system than Gaworski and had more experience performing actual credit analyses. In response, Gaworski elicited testimony from Olker that it had been necessary for Gaworski to understand credit analysis in order to supervise and direct Olker's and others' performance in this area. Gaworski also argued that he gained a thorough understanding of credit analysis while serving as Comptroller of Industrial Credit. Olker, by contrast, had little, if any, experience supervising credit analysts and had performed his first "solo" credit analysis only four months before the alleged RIF and his promotion. As for the purported need for computer skills, Gaworski presented evidence that the personnel director had never been informed of this need and that computer skills were not listed on Olker's job description. Moreover, Olker testified that his use of the computer had actually decreased by 50% after his promotion. Gaworski claims that Guimbarda's purported reasons are not credible and are mere pretext for discrimination on the basis of age.

ITT contends that Gaworski has nowhere demonstrated a link between his age and his termination. Gaworski responds, and the district court agreed, that there is substantial evidence to support the jury's verdict. Moreover, the district court specifically rejected ITT's claim in its motion for j.n.o.v. that Gaworski was terminated as part of a RIF, holding that "sufficient evidence was before the jury that Gaworski was replaced by a younger and less compensated employee." The court thus refused to overturn the jury's verdict, leading to this appeal.

On appeal our task is limited. We must: (1) consider the evidence in the light most favorable to Gaworski; (2) assume that all conflicts in the evidence were resolved by the jury in Gaworski's favor; (3) assume Gaworski proved all the facts his evidence tends to prove; (4) give Gaworski the benefit of all favorable inferences that may reasonably be drawn from the facts proved; and (5) affirm the denial of the motion for judgment as a matter of law if reasonable persons could differ as to the conclusions to be drawn from the evidence. Doyne v. Union Elec. Co., 953 F.2d 447 (8th Cir. 1992). Applying this highly deferential standard, we affirm Judge Magnuson's denial of the motion for judgment as a matter of law.

It is axiomatic that employment discrimination need not be proved by direct evidence, and indeed, that doing so is often impossible, because as the Supreme Court has said, " [t]here will seldom be 'eyewitness' testimony as to the employer's mental processes." United States Postal Serv. Bd. of Governors v. Aikens, 460 U.S. 711, 716, 103 S. Ct. 1478, 1482, 75 L. Ed. 2d 403 (1983). In disparate treatment cases based on circumstantial evidence, courts apply the analytical framework of shifting burdens developed in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S. Ct. 1817, 36 L. Ed. 2d 668 (1973), and its progeny.³  Under this framework, the plaintiff has the burden of establishing a prima facie case of discrimination. Texas Dep't. of Community Affairs v. Burdine, 450 U.S. 248, 252-53, 101 S. Ct. 1089, 1093-94, 67 L. Ed. 2d 207 (1981). Once established, the prima facie case raises a legal presumption of discrimination in the plaintiff's favor, requiring the defendant to produce legitimate, nondiscriminatory reasons for its actions. Id. If such reasons are put forth, the plaintiff, who at all times retains the burden of proving discrimination, may attempt to demonstrate that the proffered reasons are pretextual. Id. This framework is designed as a "sensible, orderly way to evaluate the evidence in light of common experience as it bears on the critical question of discrimination." Furnco Constr. Corp. v. Waters, 438 U.S. 567, 577, 98 S. Ct. 2943, 2949, 57 L. Ed. 2d 957 (1978). On appeal, our focus is on the "critical question," not on the sufficiency of a party's showing at any particular stage of the McDonnell Douglas analysis. Morgan v. Arkansas Gazette, 897 F.2d 945, 948 (8th Cir. 1990). We are guided, however, by the Supreme Court's most recent refinement of the McDonnell Douglas test in St. Mary's Honor Center v. Hicks, --- U.S. ----, 113 S. Ct. 2742, 125 L. Ed. 2d 407 (1993), which clarifies when the question of discrimination is to be left to the finder of fact and when it may be decided as a matter of law.

In Hicks, the Supreme Court held that a panel of this Court had erred in concluding that a finding that the employer's proffered nondiscriminatory reasons for its actions were pretextual compelled a judgment as a matter of law for the plaintiff. --- U.S. at ----, 113 S. Ct. at 2749. The Court stated that the legal presumption that would justify a judgment as a matter of law based on the plaintiff's prima facie case "simply drops out of the picture" when the defendant articulates a nondiscriminatory explanation. Id. From that point on, whatever the persuasive effect of the defendant's "production," the ultimate question of discrimination is for the finder of fact to decide. Id.

Most relevant to this appeal, the Court in Hicks described the sufficiency of the evidence required to support a factfinder's determination of discrimination:

The factfinder's disbelief of the reasons put forward by the defendant (particularly if disbelief is accompanied by a suspicion of mendacity) may, together with the elements of the prima facie case, suffice to show intentional discrimination. Thus, rejection of the defendant's proffered reasons, will permit the trier of fact to infer the ultimate fact of intentional discrimination, and the Court of Appeals was correct when it noted that, upon such rejection, ' [n]o additional proof of discrimination is required,' [Hicks v. St. Mary's Honor Center,] 970 F.2d, at 493 [(8th Cir. 1992)] (emphasis added).

--- U.S. at ----, 113 S. Ct. at 2749 (footnote omitted) (first alteration in original). Thus, if (1) the elements of a prima facie case are present, and (2) there exists sufficient evidence for a reasonable jury to reject the defendant's proffered reasons for its actions, then the evidence is sufficient to allow the jury to determine whether intentional discrimination has occurred, and we are without power to reverse the jury's finding. Id.

In this case, it is clear that the elements of a prima facie case of age discrimination are present. "The burden of establishing a prima facie case of disparate treatment is not onerous." Burdine, 450 U.S. at 253, 101 S. Ct. at 1093. A plaintiff establishes a prima facie case of age discrimination if she can show that she is a member of the protected age class, that she was performing adequately in her job, that she was fired, and that she was replaced by a younger person after her dismissal. Lowe v. J.B. Hunt Transp., Inc., 963 F.2d 173 (8th Cir. 1992). Here, plaintiff Gaworski clearly was within the ADEA's "protected class" of those over forty years old. See 29 U.S.C. § 631(a) (1988). His performance reviews and Guimbarda's admission that he was a good worker indicate that he was performing more than adequately. He was terminated from his employment. And contrary to ITT's claims, the evidence supports a finding that he was replaced by a younger employee.

Defendants argue that Gaworski was terminated as part of a reduction in force and that, therefore, under this Court's decision in Holley v. Sanyo Manufacturing, Inc., 771 F.2d 1161 (8th Cir. 1985), he must make an "additional showing" of age-related animus in order to establish his prima facie case. We agree with the district court, however, that the evidence supports a finding that Gaworski's position was filled by a younger employee rather than eliminated after Gaworski was laid off. Viewed in the light most favorable to Gaworski, the evidence shows that the younger Olker was given Gaworski's office, substantially all of Gaworski's former duties, a job description that Olker admits was "fairly comparable" to Gaworski's, and a title almost identical to the one Gaworski had had. Moreover, we have previously stated that RIF plans "generally include objective criteria by which to determine which jobs will be eliminated and often include objective evidence of a business decline." Hillebrand v. M-Tron Indus., Inc., 827 F.2d 363, 367 (8th Cir. 1987), cert. denied, 488 U.S. 1004, 109 S. Ct. 782, 102 L. Ed. 2d 774 (1989). Neither were present here. In fact, Gaworski presented evidence that at the time ITT sought to cut its workforce, business was increasing--of the $30 million of new loans booked in the year preceding the RIF, $20 million were booked in the final four months before the RIF took place. On this basis, reasonable minds could conclude that Gaworski had, in fact, been replaced and that ITT's purported RIF was a pretextual explanation for his discharge.

Evidence exists to allow reasonable minds to conclude that each of ITT's proffered nondiscriminatory explanations was pretextual. ITT claimed that it retained Olker over Gaworski because Gaworski lacked a substantive understanding of credit analysis, that his skills were, as Guimbarda described at trial, "more administrative in nature." Tr. at 81. Gaworski presented evidence, however, that in December of 1985, Guimbarda had signed a performance review stating that one of Gaworski's weaknesses was that he was too detail oriented and that he should delegate and manage more. Moreover, Gaworski established that in addition to his experience and highly-rated performance at CRG, as Comptroller of ITT Industrial Credit, he had served with the company president and other top-level administrators on ITT's Senior Credit Committee, the "final arbiter" assessing credit analyses conducted by other Industrial Credit employees. Olker, by contrast, had little or no experience supervising analysts, as he was required to do after his promotion. Olker himself testified that it was necessary for Gaworski to understand credit analysis in order to supervise Olker's activities.

ITT also claimed that Olker was more valuable to the company than Gaworski because of his greater computer skills. Guimbarda testified that the "computer experience to me was critical." Tr. at 104. Again, Gaworski presented evidence to the contrary. He demonstrated that the need for computer skills was not mentioned in Olker's new job description, that the personnel director had never been informed of this need, and that Olker was in fact applying his computer skills much less in the new position than he had before. Moreover, Gaworski elicited testimony from Guimbarda that Guimbarda had never considered teaching Gaworski about the workings of the computer system, even though Gaworski's performance review spoke highly of his ability to grasp new concepts and Guimbarda himself admitted that "it's something I guess that Dick [Gaworski] could have done." Tr. at 83.

" [T]he ADEA is not intended to be used as a means of reviewing the propriety of a business decision...." Jorgensen v. Modern Woodmen of Am., 761 F.2d 502, 505 (8th Cir. 1985). The materially conflicting evidence in this case, however, raises a question of fact as to the believability, not the propriety, of ITT's purported reasons for discharging Gaworski. Viewing the evidence in the light most favorable to Gaworski, we find that the evidence is sufficient to allow a reasonable jury to conclude that ITT's proffered non-discriminatory reasons were "unworthy of credence," see Hazen Paper Co. v. Biggins, --- U.S. ----, ----, 113 S. Ct. 1701, 1708, 123 L. Ed. 2d 338 (1993). A reasonable jury could find that ITT acted for reasons other than what it claimed.

The elements of the plaintiff's prima facie case are thus present and the evidence is sufficient to allow a reasonable jury to reject the defendant's non-discriminatory explanations. The "ultimate question" of discrimination must therefore be left to the trier of fact to decide. See Hicks, --- U.S. at ----, 113 S. Ct. at 2749. The jury in this case could reasonably have accepted or rejected the defendant's proffered explanations. By its verdict, it presumably rejected them. Hicks establishes that " [n]o additional proof of discrimination is required." Id. (emphasis added) (citation omitted). Based upon the elements of the plaintiff's prima facie case and the jury's rejection of the defendant's explanations, the jury could infer that discrimination had occurred. Id.

The jury's finding that ITT intentionally discriminated against Gaworski on the basis of age was within its purview as the finder of fact. We therefore affirm.

In its calculation of backpay to be awarded to Gaworski, the district court relied on the jury's special verdict to calculate the pay Gaworski would have received had he not been wrongfully terminated.⁴  From an initial award of $387,523.27, the district court then deducted $109,530.00 in income Gaworski had earned from two other employers in the interim, as well as $12,101.00 Gaworski had collected in unemployment compensation. The court refused to deduct $7,345.00 Gaworski had earned doing free-lance consulting work during the backpay period, determining that this was part-time moonlighting income that he could have earned even if he had remained employed with ITT. The court also refused to deduct $18,061.80 that Gaworski had received from his pension fund during the backpay period, concluding that those payments were from a "collateral source." Finally, the court added $11,625.71 in lost matching contributions to Gaworski's 401(k) plan, $4,151.50 in job search expenses, and $4,467.00 in replacement insurance premiums, to reach a grand total of $286,136.48.⁵ 

Defendants argue on appeal that the court should not have included the 401(k) contributions and the insurance expenses and should have deducted the consulting income and pension payments. Cross-appellant E.E.O.C. argues that the unemployment expenses should not have been deducted. We agree with defendants that the pension payments should have been deducted, and with the E.E.O.C. that the unemployment compensation should not have been deducted. We affirm as to the other possible offsets and additions.

1. Insurance Expenses and 401(k) Contributions

ITT urges that the district court should not have added insurance expenses or 401(k) contributions to Gaworski's backpay award, because the claimed losses were not supported by evidence admitted at trial. Our review of the record, however, shows that Gaworski did testify that he paid "about $4,000" to replace the life and disability insurance that had previously been paid by ITT. Tr. at 278-79. More important, the record shows that the parties understood that the record would be left open after the jury trial for submission to the court of material relevant to the determination of damages.⁶  Based on the agreement, Gaworski submitted to the court copies of the checks he wrote to pay his insurance premiums. He also made a representation to the court, not disputed by ITT, that ITT had regularly made matching contributions of 3% of Gaworski's salary to his 401(k) savings plan. On this basis, it was not clearly erroneous for the court to add the $4,467 in insurance costs, see Tolan v. Levi Strauss & Co., 867 F.2d 467 (8th Cir. 1989), or the $11,625.71 in lost 401(k) contributions to Gaworski's backpay award. Both amounts serve to restore Gaworski to the position he would have been in but for his discriminatory discharge.

Second, ITT challenges the district court's refusal to deduct the $7,345 Gaworski earned from his self-employment as a consultant, claiming that it is mitigation income that must be offset against an award of backpay. The district court found that the money was "moonlighting" income that Gaworski could have earned even if he had remained employed with ITT. It is not error for a court to refuse to deduct moonlighting income from an award of backpay. Behlar v. Smith, 719 F.2d 950, 954 (8th Cir. 1983) (per curiam), cert. denied, 466 U.S. 958, 104 S. Ct. 2169, 80 L. Ed. 2d 552 (1984). ITT had the burden of proving that Gaworski could not have earned the income had he not been terminated. See DiSalvo v. Chamber of Commerce, 568 F.2d 593, 598 (8th Cir. 1978). Our review of the record fails to convince us that the district court's finding was clearly erroneous.

Finally, ITT argues that the district court should have deducted $18,061.80 that Gaworski collected from his pension fund after his termination. Gaworski and ITT stipulated at trial that Gaworski would not oppose deducting the pension payments from the backpay and ITT would agree to calculate Gaworski's future pension payments as if he had continued working at ITT until 1991 and had never withdrawn any of the earlier amounts. Tr. at 236-37. Gaworski's position changed only in a letter brief regarding damages issues written to the district court after the jury trial. In his appellate brief, Gaworski does not mention the pension payments, and the E.E.O.C., relying on Gaworski's earlier stipulation, does not oppose ITT on this point.

We agree with ITT that Gaworski should be bound by his earlier stipulation. In Consolidated Grain & Barge Co. v. Archway Fleeting & Harbor Service, Inc., 712 F.2d 1287 (8th Cir. 1983) (per curiam), we held that the district court had erred in failing to follow the parties' stipulation regarding interest to be awarded in the event liability were found. There, as here, there was "no showing that either party sought relief from the stipulation in district court, or that the district court on its own motion considered and rejected the stipulation." Id. at 1290. Indeed, the district court here explicitly relied on part of the stipulation to order that the defendants pay the pension benefits Gaworski would have received had he not been terminated, including those benefits that would have accrued had he been employed until 1991. The court erred in not enforcing both sides of the parties' bargain. Consequently, Gaworski's backpay award should be reduced by the sum of $18,061.80.⁷ 

B. The Cross-Appeal: Unemployment Compensation

The district court deducted $12,101.00 Gaworski received in unemployment compensation after his termination because, in the court's view, " [t]o allow Gaworski to receive unemployment compensation as well as the back pay would provide double recovery." We reverse.

It is fundamental that "an employer can not set up in mitigation of damages in a tort action by an injured employee indemnity from a collateral source, such as insurance or compensation or benefits under a Workmen's Compensation Act, even where the defendant has contributed to the fund." Chicago Great W. Ry. v. Peeler, 140 F.2d 865, 868 (8th Cir. 1944). This "collateral source rule" has a long pedigree, but it remains applicable and gains in significance in the context of employment discrimination claims. As we noted in Beshears v. Asbill, 930 F.2d 1348 (8th Cir. 1991), because of the collateral source rule, " ' [m]ost courts have refused to deduct such benefits as social security and unemployment compensation from ADEA awards,' " id. at 1355 n. 11 (quoting Guthrie v. J.C. Penney Co., 803 F.2d 202, 209 (5th Cir. 1986)). Such a refusal is sensible and in accordance with the purposes of the ADEA.

Backpay awards in discrimination cases serve two functions: they make victimized employees whole for the injuries suffered as a result of the past discrimination, and they deter future discrimination. Albemarle Paper Co. v. Moody, 422 U.S. 405, 421, 95 S. Ct. 2362, 2373, 45 L. Ed. 2d 280 (1975); see Gibson v. Mohawk Rubber Co., 695 F.2d 1093, 1097 (8th Cir. 1982) (noting that ADEA remedies aim "to eliminate the unlawful practices and to restore aggrieved persons to the position where they would have been if the illegal discrimination had not occurred"). The Supreme Court in Albemarle emphasized the importance of the deterrence function, saying:

It is the reasonably certain prospect of a backpay award that "provide [s] the spur or catalyst which causes employers and unions to self-examine and to self-evaluate their employment practices and to endeavor to eliminate, so far as possible, the last vestiges of an unfortunate and ignominious page in this country's history."

422 U.S. at 417-18, 95 S. Ct. at 2371-72 (quoting United States v. N.L. Indus., Inc., 479 F.2d 354, 379 (8th Cir. 1973)).⁸  Reducing a backpay award by unemployment benefits paid to the employee, not by the employer, but by a state agency, see NLRB v. Gullett Gin Co., 340 U.S. 361, 364, 71 S. Ct. 337, 339, 95 L. Ed. 337 (1951), makes it less costly for the employer to wrongfully terminate a protected employee and thus dilutes the prophylactic purposes of a backpay award. See Craig v. Y & Y Snacks, Inc., 721 F.2d 77, 84 (3d Cir. 1983); E.E.O.C. v. Ford Motor Co., 645 F.2d 183, 196 (4th Cir. 1981), rev'd & remanded on other grounds, 458 U.S. 219, 102 S. Ct. 3057, 73 L. Ed. 2d 721 (1982). Indeed, it leads to a windfall to the employer who committed the illegal discrimination. Hunter v. Allis-Chalmers Corp., 797 F.2d 1417, 1429 (7th Cir. 1986); E.E.O.C. v. Sandia Corp., 639 F.2d 600, 626 (10th Cir. 1980). By virtue of state aid provided "to carry out a policy of social betterment for the benefit of the entire state," and not "to discharge any liability or obligation" of the employer, Gullett Gin, 340 U.S. at 364, 71 S. Ct. at 339, the employer winds up paying less to the employee than it would have had it not illegally terminated him.⁹ 

Based on these considerations, no circuit that has considered the matter has determined that unemployment benefits should, as a general rule, be deducted from backpay awards in discrimination cases. Circuits have split, however, over whether deducting unemployment benefits should be prohibited or should be left to the discretion of the trial court. The majority have held that, as a matter of law, unemployment benefits should not be deducted from backpay awards. See Craig, 721 F.2d at 85; Rasimas v. Michigan Dep't of Mental Health, 714 F.2d 614, 627-28 (6th Cir. 1983), cert. denied, 466 U.S. 950, 104 S. Ct. 2151, 80 L. Ed. 2d 537 (1984); Kauffman v. Sidereal Corp., 695 F.2d 343, 346-47 (9th Cir. 1982) (per curiam); E.E.O.C. v. Ford Motor Co., 645 F.2d 183, 196 (4th Cir. 1981), rev'd & remanded on other grounds, 458 U.S. 219, 102 S. Ct. 3057, 73 L. Ed. 2d 721 (1982), original position adhered to on remand, 688 F.2d 951, 952 (4th Cir. 1982) (per curiam); Brown v. A.J. Gerrard Mfg. Co., 715 F.2d 1549, 1550-51 (11th Cir. 1983) (en banc) (per curiam). Three circuits have adopted a minority position that deducting unemployment benefits lies within the discretion of the trial court. See Cooper v. Asplundh Tree Expert Co., 836 F.2d 1544, 1555 (10th Cir. 1988); Hunter, 797 F.2d at 1429 (Posner, J., acknowledging discretion as Seventh Circuit rule but stating that it "may be unduly favorable to defendants"); Marshall v. Goodyear Tire & Rubber Co., 554 F.2d 730, 736 (5th Cir. 1977). The Second Circuit has in the past affirmed a deduction of unemployment benefits as discretionary, see E.E.O.C. v. Enterprise Assoc. Steamfitters Local 638 of U.A., 542 F.2d 579, 592 (2d Cir. 1976), cert. denied, 430 U.S. 911, 97 S. Ct. 1186, 51 L. Ed. 2d 588 (1977), but more recently indicated that the circuit's rule remains unsettled, see Promisel v. First Am. Artificial Flowers, 943 F.2d 251, 258 (2d Cir. 1991), cert. denied, --- U.S. ----, 112 S. Ct. 939, 117 L. Ed. 2d 110 (1992).

Because we believe the majority rule is the more sound position, we hold that unemployment benefits should not be deducted from awards of backpay under the ADEA. Unemployment benefits are collateral source payments that cannot be termed even "partial consideration" for employment. Gullett Gin, 340 U.S. at 364, 71 S. Ct. at 339. We believe that no employer should benefit because of the state's beneficence in providing for one of the employer's illegally discharged employees. If Gaworski's award is to be reduced, it should be the state that seeks to recoup the benefits it paid out.¹⁰  Accordingly, we reverse this aspect of the district court's decision and find that Gaworski's backpay award should be increased by $12,101.00.

Based on the foregoing analysis, the judgment on the jury verdict finding the defendants liable for violating the ADEA is affirmed. The backpay award of $265,892.27 is reduced by $18,061.80 to reflect the parties' stipulation regarding pension payments and increased by $12,101.00 received by Gaworski as unemployment compensation. We remand for entry of a new award consistent with this opinion.

JOHN R. GIBSON, Senior Circuit Judge, dissenting.

I respectfully dissent because I conclude that no reasonable jury could have determined that ITT terminated Gaworski because of his age, and therefore, I would reverse the judgment of the district court.

The court today concludes that it must affirm the jury's verdict because Gaworski established a prima facie case of discrimination and there is sufficient evidence from which the jury could reject ITT's explanation for Gaworski's termination. I disagree with the court's analysis of the evidence and its interpretation of the Supreme Court's most recent decision dealing with discrimination, St. Mary's Honor Center v. Hicks, --- U.S. ----, 113 S. Ct. 2742 (1993). The reasoning of both Justice Scalia's opinion for the Court and Justice Souter's dissent in Hicks revolved around the Supreme Court's earlier decisions in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S. Ct. 1817, 36 L. Ed. 2d 668 (1973), and Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 101 S. Ct. 1089, 67 L. Ed. 2d 207 (1981). The Court particularly focused on the allocation of the burdens of production required by those cases. The Court stated that once a plaintiff establishes a prima facie case of discrimination, the defendant's failure to introduce evidence of a non-discriminatory reason will cause judgment to go against it "unless the plaintiff's prima facie case is held to be inadequate in law or fails to convince the factfinder." Hicks, --- U.S. at ----, 113 S. Ct. at 2749 n. 3 (emphasis in original). This practical coercion "causes the McDonnell Douglas presumption to function as a means of 'arranging the presentation of evidence.' " Id. (quoting Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 986, 108 S. Ct. 2777, 2784, 101 L. Ed. 2d 827 (1988)). Both Justices Scalia and Souter acknowledged that at all times the plaintiff has the burden to demonstrate the ultimate fact of intentional discrimination. --- U.S. at ----, 113 S. Ct. at 2747, 2760.

Justice Scalia's opinion for the Court makes clear that the burden of production "precedes the credibility-assessment stage." Id. --- U.S. at ----, 113 S. Ct. at 2748 (emphasis in original). The Court states: "At the close of the defendant's case, the court is asked to decide whether an issue of fact remains for the trier of fact to determine." Id. The Court explained that it would be required to award judgment to the employee as a matter of law if the employee established a prima facie case of discrimination and the employer failed to meet its burden of producing a non-discriminatory reason for its action. Id. The Court explained further that if the employer failed to sustain its burden of producing a non-discriminatory reason for its action, but reasonable minds could differ as to whether the employee established a prima facie case, a question of fact remained for determination. Id.

The Court then considered the situation in which an employer had carried its burden of production by articulating a non-discriminatory justification for its action. Id. --- U.S. at ----, 113 S. Ct. at 2749. The Court stated: "The McDonnell Douglas framework--with its presumptions and burdens--is no longer relevant," and " [t]he presumption, having fulfilled its role of forcing the defendant to come forward with some response, simply drops out of the picture." Id. --- U.S. at ----, 113 S. Ct. at 2749 (citing Burdine, 450 U.S. at 255, 101 S. Ct. at 1094).¹¹  Hicks then looked to United States Postal Service Board of Governors v. Aikens, 460 U.S. 711, 715-16, 103 S. Ct. 1478, 1482, 75 L. Ed. 2d 403 (1983), and concluded that after the close of evidence, the district court's inquiry is limited to considering the ultimate issue of whether the defendant intentionally discriminated against the plaintiff, rather than considering the several steps in the shifting burdens. Hicks, --- U.S. at ----, 113 S. Ct. at 2754.

Thus, the lesson from Hicks is that whether there is a factual issue for determination must be decided at the close of the defendant's case. At that point, the district court must decide whether the plaintiff has presented sufficient evidence to establish the ultimate fact of intentional discrimination.

This is the approach we have taken in a number of our earlier decisions. We have held that our task on appeal is limited, and that we will not assess the adequacy of a party's showing at any particular stage of the McDonnell Douglas analysis. Aikens, 460 U.S. at 714-15, 103 S. Ct. at 1482; Morgan v. Arkansas Gazette, 897 F.2d 945, 948 (8th Cir. 1990); Kientzy v. McDonnell Douglas Corp., 990 F.2d 1051, 1056 (8th Cir. 1993). See Maness v. Star-Kist Foods, Inc., 7 F.3d 704, 707-08 (8th Cir. 1993) (following Aikens and Hicks and holding the Burdine analysis irrelevant). Thus, we focus our attention on "the ultimate factual issue of whether [the employer] intentionally discriminated against [the employee]," and study the record to determine whether the evidence was sufficient to support the verdict. Aikens, 460 U.S. at 714-15, 103 S. Ct. at 1481-82; Morgan, 897 F.2d at 948-49; Kientzy, 990 F.2d at 1056.

The court's analysis starts with language from Hicks that the factfinder's disbelief of the reasons put forward by the employer may, together with the elements of the prima facie case, suffice to show intentional discrimination and that rejection of the employer's proffered reason for discharge will permit the trier of fact to infer the ultimate fact of intentional discrimination, and that no additional proof of discrimination is required. Supra at 7 (citing Hicks, --- U.S. at ----, 113 S. Ct. at 2749). The shortcoming of the court's analysis is that here we have no clear expression of disbelief or rejection of ITT's proffered reasons by the jury, but at most, evidence upon which the jury could have rejected ITT's reasons for discharge.

This case points up the inexact fit of the McDonnell Douglas Burdine standard as restated by Hicks to a case submitted to a jury. The court instructed the jury in this case as to the elements of the prima facie case, and that it must find that age was a determining factor in ITT's decision to discharge Gaworski. The court instructed the jury most generally that ITT denied that it discriminated against Gaworski because of his age. After explaining to the jury that Gaworski could prove age discrimination in a "variety of ways," the court instructed that Gaworski could prove age discrimination by meeting his burden of proving that the reasons offered by ITT for his discharge were "arranged to accomplish discrimination" against him or designed as a coverup. The instruction did not state ITT's proffered reasons for Gaworski's discharge or require that the jury reject ITT's reasons. The court submitted the case to the jury using a special verdict form, asking the jury only whether it found that age was the determining factor in Gaworski's termination, and if so, what percentage in compensation Gaworski would have received if he had continued his employment.

These most general findings do not show that the jury rejected or disbelieved ITT's proffered reasons. Indeed, the instructions did not require that the jury reject or disbelieve ITT's proffered reasons in order to find age discrimination. Thus, in place of the rejection or disbelief required by Hicks to create an inference of age discrimination, we have only the court's inference of such a finding. From this inference, the court then assumes the ultimate inference of age discrimination. It is a fundamental principle that an inference will not support an inference. The court's analysis simply fails at this point.

It is also significant that two of the recent opinions of courts of appeals, LeBlanc v. Great American Insurance Company, 6 F.3d 836 (1st Cir. 1993), and Anderson v. Baxter Healthcare Corporation, 13 F.3d 1120 (7th Cir. 1994), both start their analysis with the same passage from Hicks that the court relies on today. LeBlanc, 6 F.3d at 842-43; Anderson, 13 F.3d at 1122-1124. Both courts sustained summary judgments in favor of the employer. The court in LeBlanc concluded that there was no direct or circumstantial evidence of discriminatory animus on the part of the employer. 6 F.3d at 845. LeBlanc stressed the fact that once the employer articulated non-discriminatory reasons for its action, the presumption of age discrimination raised by the prima facie case vanished. Id. "Left to be decided is whether the evidence, in its entirety, would permit a reasonable fact-finder to infer that Great American's decision to terminate LeBlanc was inspired by age animus." Id. The Seventh Circuit in Anderson engaged in a similar, but more detailed, analysis of the evidence put forward to demonstrate the falsity of the employer's proffered reasons for discharge, and ultimately determined that the evidence was inadequate and that the employee failed to meet the ultimate burden of proving intentional age discrimination. Anderson, 13 F.3d at 1124-26.

Both LeBlanc and Anderson demonstrate the validity of this court's approach in determining sufficiency of the evidence to support verdicts in age discrimination cases by considering, as the Supreme Court instructed in Aikens, the entire record to determine if the employee has met his burden of proving the ultimate issue of age discrimination.¹²  The shortcoming of the court's approach today is simply that it attempts to elevate the inferences to a greater position than that accorded them, particularly by the explication of the McDonnell Douglas and Burdine standards in Hicks. The court today simply carries the inference arising from the prima facie case a step too far. As Hicks makes clear, the inference of discrimination falls from the case once the employer articulates a non-discriminatory reason for the discharge. The court's creation of an inference that the jury must have rejected ITT's explanation as a basis for continuing the inference does not square with Hicks.

The essence of the court's reasoning is that an inference of discrimination arises from Gaworski's prima facie case, separate and distinct from the mandatory rebuttable presumption. Although the court concedes that the mandatory rebuttable presumption drops from the case, the court insists that the inference of discrimination remains and is sufficient to take Gaworski's case to the jury. Hicks, McDonnell Douglas, and Burdine do not support this reasoning.

The court's reliance on the force of the inference arising from the prima facie case is best answered by the following footnote from Burdine:¹³ 

The phrase 'prima facie case' not only may denote the establishment of a legally mandatory, rebuttable presumption, but also may be used by courts to describe the plaintiff's burden of producing enough evidence to permit the trier of the fact to infer the fact at issue. 9 J. Wigmore, Evidence Sec. 2494 (3d ed. 1940). McDonnell Douglas should have made it apparent that in the Title VII context we use 'prima facie case' in the former sense.

450 U.S. at 254 n. 7, 101 S. Ct. at 1094 n. 7.

The Court made clear in Burdine that if the employer meets its burden of producing a non-discriminatory reason for its action, the mandatory presumption is rebutted and drops from the case. Both the majority and dissenting opinion in Hicks clearly recognize this. Justice Scalia comments that at this point the McDonnell Douglas framework--with its presumptions and burdens--is no longer relevant and to resurrect it would be contradictory to Burdine. Hicks, --- U.S. at ----, 113 S. Ct. at 2749. In his dissent, Justice Souter states that everyone agrees that the mandatory presumption is not resurrected. Id. at ----, 113 S. Ct. 2759 n. 2.

Thus, the issue before us is whether there was sufficient evidence to support an inference of age discrimination. I reject the court's proposition that the inference from the prima facie case has continuing preemptive force. There is no doubt that once the presumption drops from the case, the evidence introduced to establish a prima facie case and "inferences properly drawn therefrom," may be considered in determining whether the employer's explanation is pretextual. Burdine, 450 U.S. at 255 n. 10, 101 S. Ct. at 1095 n. 10. Justice Souter specifically referred to this note in his dissent, Hicks, --- U.S. at ----, 113 S. Ct. at 2759, but it is evident that both Justice Souter in his dissent and the Court in Burdine are discussing the consideration of evidence, not the continuing existence of an independent inference of discrimination. This discussion will not carry the baggage the court attempts to place upon it.

The court concludes that the jury could have inferred age discrimination from evidence that: (1) ITT presented no evidence of the need for a reduction in workforce; (2) ITT's policies did not support Gaworski's termination or Olker's promotion; (3) Gaworski was the oldest, highest paid, and only "pension eligible" employee at Capital Resources; and (4) ITT's reasons for retaining Olker, a younger employee, were pretextual--Gaworski's job was not really eliminated, Olker replaced him, and Gaworski was more qualified for the position.

None of these facts support an inference of age discrimination. The court today accepts Gaworski's argument that it may infer age discrimination based on ITT's lack of evidence about a need for a reduction in force or any objective plan for implementing a reduction. Whether Guimbarda's decision to reduce Capital Resources' workforce was a sound business decision, however, is irrelevant to our question of whether Gaworski was terminated because of his age. It is well-established that "the ADEA is not intended to be used as a means of reviewing the propriety" of an employer's business decision. Jorgensen v. Modern Woodmen of Am., 761 F.2d 502, 505 (8th Cir. 1985) (discussing a company's decision to split an employee's sales territory); see Neufeld v. Searle Labs., 884 F.2d 335, 340 (8th Cir. 1989); Bell v. Gas Serv. Co., 778 F.2d 512, 515 (8th Cir. 1985); see generally Walker v. AT & T Technologies, 995 F.2d 846 (8th Cir. 1993).

Although we have sometimes found that an employer's "reduction in force" may have been a ruse to mask intentional age discrimination, see, e.g., Hillebrand v. M-Tron Indus., Inc., 827 F.2d 363 (8th Cir. 1987),¹⁴  cert. denied, 488 U.S. 1004, 109 S. Ct. 782, 102 L. Ed. 2d 774 (1989), Gaworski's own testimony shows that is not the case here. First, Gaworski testified that he was not accusing Guimbarda of laying off five people just to make sure he was laid off, and Gaworski's counsel's opening statement and closing argument emphasized that there was no question that Capital Resources' business had taken a downward turn. Second, Guimbarda laid off four out of seventeen total employees in his division for a twenty-three percent reduction, and Gaworski stipulated that Capital Resources never again employed more individuals than it had remaining after the reduction in force. Moreover, there was unrefuted testimony that although Capital Resources had increased profits preceding the layoff, these profits were driven by the liquidation of loans,¹⁵  and that without new business volume there was a decreased prospect for future revenues.¹⁶  There was also unrefuted testimony that there was reduced loan processing prior to the layoff, and that reduced loan processing meant reduced demand for loan processing jobs like Gaworski's. Gaworski admitted that the total number of loans generating income had been shrinking in the year preceding his termination. Moreover, if the court creates an inference of discrimination from evidence that ITT's business was actually increasing, the court must address ITT's complaint that the district court erred in refusing evidence that Capital Resources eventually closed, and that all employees, including Olker, were laid off. The court's statement that "reasonable minds could conclude that Gaworski had, in fact, been replaced and that ITT's purported reduction in force was a pretextual explanation for his discharge" contradicts the evidence in this case, including Gaworski's own admission. I conclude that no reasonable juror could have found that Guimbarda reduced the workforce as a means to discriminate against Gaworski because of his age.

Along this same line, the court accepts Gaworski's argument that Guimbarda's failure to follow ITT's written policies regarding the layoff and promotion of Olker is evidence from which a jury could infer age discrimination. The ITT termination policy provides that no layoff may occur without advance written notification to senior levels of operating management and the director of personnel. Robert Tate, the director of personnel, testified that Guimbarda "technically" violated the policy by giving Tate verbal, instead of written, notification. Guimbarda's failure to comply with an administrative requirement, however, is not enough to infer that age was a determining factor in his decision to terminate Gaworski. Guimbarda notified Tate of the reduction, indeed, Tate traveled to Guimbarda's division to assist with the reduction in force. As we observed earlier, even Gaworski stopped short of accusing Guimbarda of laying off five people just so he could terminate Gaworski because of his age. Likewise, although certain procedures may apply in normal promotional decisions, the failure to "formally" review Olker's performance before adding most of Gaworski's former duties to Olker's job is not evidence that age influenced Guimbarda's decision to terminate Gaworski.

The court also concludes that the jury could have inferred age discrimination because Gaworski was the oldest, highest paid, and only "pension eligible" employee at Capital Resources. A unanimous Supreme Court recently held that an employer does not violate the ADEA by acting on the basis of a factor, such as an employee's pension status or seniority, that is empirically correlated with age. Hazen Paper Co. v. Biggins, --- U.S. ----, ----, 113 S. Ct. 1701, 1705, 123 L. Ed. 2d 338 (1993).¹⁷ 

The Court reiterated that liability in a disparate treatment case "depends on whether the protected trait (under the ADEA, age) actually motivated the employer's decision." Id. at ----, 113 S. Ct. at 1706; see Aikens, 460 U.S. at 716, 103 S. Ct. at 1482. The Court observed that an employer may be liable when it relies on a formal, facially discriminatory policy requiring adverse treatment of employees with the protected trait, or when it is motivated by the protected trait on an informal, ad hoc basis. Hazen Paper, --- U.S. at ----, 113 S. Ct. at 1706. Nevertheless, the Court emphasized that " [w]hatever the employer's decisionmaking process, 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." Id. Thus, the Court concluded that "an employer does not violate the ADEA just by interfering with an older employee's pension benefits that would have vested by virtue of the employee's years of service." Id. at ----, 113 S. Ct. at 1707-08. The Court cautioned, however, that an employer who "targets employees" with a particular pension status on the assumption that those employees are likely to be older engages in age discrimination. Id. In that situation, pension status (or in a similar situation, another empirically correlated factor) may be a "proxy" for age in the sense that an employer may suppose a correlation between the two factors and act accordingly, but not in the sense that the ADEA makes the two factors equivalent. Id. (rejecting Metz v. Transit Mix, Inc., 828 F.2d 1202, 1208 (7th Cir. 1987), which interpreted "proxy" to mean statutory equivalence).

The court today also finds evidence of age discrimination from the fact that Gaworski was the oldest employee. Although the ADEA prohibits age-based employment decisions, it does not confer special rights on an employee who happens to be the "oldest" in a company or division. Thus, we should not infer age discrimination from the simple fact that Gaworski happened to be the oldest in the division. See Hicks, --- U.S. at ----, 113 S. Ct. at 2747. Without more than evidence of status in the protected class, the ADEA does not impose liability. The evidence here does not suggest that Guimbarda "suppose [d] a correlation" between Gaworski's age and his status as the "oldest" employee and acted accordingly. See Hazen Paper, --- U.S. at ----, 113 S. Ct. at 1707. Indeed, none of the other laid off employees was within the protected class, and Olker (age 46), was also in the protected class.

Next, even assuming Guimbarda laid off Gaworski because he had the highest salary, there is still no evidence that Gaworski's age prompted that decision. An analytical framework similar to that used in Hazen Paper can be applied to salary levels: A younger employee, outside the protected class, may have spent his entire career with one employer and worked his way up the ranks to obtain a higher salary than a newly hired older worker. The Supreme Court rejected the Seventh Circuit's decision in Metz, 828 F.2d at 1208, which held that terminating an employee for having a higher salary may be a "proxy" for age discrimination, meaning the statutory equivalent of age discrimination.¹⁸  Hazen Paper, --- U.S. at ----, 113 S. Ct. at 1707. Interestingly, if Guimbarda had used seniority to distinguish Gaworski and Olker, Gaworski still would have been laid off because although both worked at Capital Resources for about the same amount of time, Olker had worked for ITT approximately eight years longer than Gaworski. Again, the evidence here does not suggest that Guimbarda "suppose [d] a correlation" between Gaworski's age and his status as the "highest paid" employee and acted accordingly. See id.; Anderson, 13 F.3d at 1126 (employee could not prove age discrimination even if he was fired simply because employer desired to reduce its salary costs by discharging him).

Finally, the mere fact that Gaworski was "pension eligible" is not enough to infer age discrimination. Contrary to the implication of the court, the decision to lay off Gaworski did not deprive him from his pension because he had already qualified to receive his base pension benefits. Although the decision may have deprived him of the possibility to qualify for a higher pension (thirty to forty percent more) with added years of service, company policy required him to work for ITT for approximately ten more years. Such an assumption requires too much speculation. Indeed, Gaworski's argument might be more persuasive if ITT had terminated him immediately before he became eligible for an increase in pension benefits as eligibility was based on meeting the preliminary age requirement of fifty-five as well as years of service. Cf. Hazen Paper, --- U.S. at ----, 113 S. Ct. at 1707-08 (limiting holding to plans with only a years-of-service eligibility requirement).

The court also concludes that age discrimination is established because ITT's reasons for retaining Olker instead of Gaworski were pretextual. The court concludes that a reasonable jury could conclude that Gaworski's job was not "eliminated" as ITT claims because Olker replaced him, and that Gaworski was more qualified than Olker for the position, and that these facts are sufficient to support an inference of age discrimination.

To ultimately prevail, however, Gaworski must show not only that ITT's reasons were pretextual, but that age was the real reason for firing him. Hicks, --- U.S. at ----, 113 S. Ct. at 2747. This he failed to do. Guimbarda testified that in downsizing the division, he needed credit analysis and credit investigation skills more than he needed supervisory skills. He explained that he chose to retain Olker over Gaworski because Olker had more computer experience, had about the same amount of time in the division as Gaworski, and had more experience doing actual credit write ups and analysis. Guimbarda testified that Olker's computer experience¹⁹  was the critical difference because Olker could fix the computer system if it went down. Gaworski admitted that Olker had more computer experience and had been involved in developing the Capital Resources' software. Although Gaworski believed another employee had more familiarity with how the system worked, he did not dispute Olker's capabilities. Instead, Gaworski attacks ITT's need for Olker's computer skills in the "manager of credit and administration" position. Gaworski points out that the "manager of credit and administration" job description neither lists computer work as a responsibility, nor requires computer skills in its description of qualifications. Nevertheless, Gaworski conceded that ITT's job descriptions are not rigidly followed. Gaworski's own job description described his title as "manager of credit and operations," while Gaworski testified that his title was "vice president and investment director." Gaworski also admitted that when he hired Olker for a senior investment manager position, Olker did not meet the minimum listed qualifications. Gaworski also testified that he did not remember whether he had considered the job description when he recommended Olker for the position, and although Olker did not meet the listed qualifications, he did not think he was discriminating against anyone. Thus, viewing the evidence in the light most favorable to Gaworski, I believe that there is insufficient evidence to show that Gaworski's age, not Olker's computer skills or some other reason, was the real reason ITT retained Olker instead of Gaworski. See id. at ----, 113 S. Ct. at 2752-56. Again, this court cannot second guess a business' decision that computer skills are valuable to a division that has lost five employees. Jorgensen, 761 F.2d at 505; see Neufeld, 884 F.2d at 340; Bell, 778 F.2d at 515; see generally Walker, 995 F.2d at 846.

Even assuming that all of Guimbarda's stated reasons were false, that does not answer the ultimate question. Merely proving that Guimbarda's reasons are false is not enough to prove unlawful age discrimination--Gaworski must prove that age discrimination was the real reason for his termination. Id. --- U.S. at ----, 113 S. Ct. at 2755-56. There is insufficient evidence on this record to link any of Guimbarda's decisions to Gaworski's age.

Finally, although direct evidence is not required to prove age discrimination, Gaworski presented no direct evidence or pattern and practice evidence from which discrimination might be inferred. Gaworski testified that he did not sense any hostility from Guimbarda regarding his age, no one at ITT had ever referred to his age in a disparaging way, he had never been treated badly because of his age, and no documents referred disparagingly to his age. The Capital Resources group hired Gaworski when he was 53. Nor was there evidence of any negative comments about older people generally. Guimbarda had hired several employees in the protected class, and other than Gaworski, never fired any other employees who were in the protected class. Indeed, none of the other laid-off employees was in the protected age group. The evidence is simply not enough to create a jury question.

For the same reasons, I also conclude that Gaworski failed to make a prima facie case under Holley v. Sanyo Manufacturing Inc., 771 F.2d at 1161, 1166-68 (8th Cir. 1985),²⁰  by demonstrating "additional evidence" that age was a factor in his termination.

After carefully examining the evidence, I conclude that no reasonable jury could have found that Guimbarda fired Gaworski because of his age. Accordingly, I dissent and would reverse the district court's order denying ITT's motion for judgment notwithstanding the verdict.