holding that the factors established in Cirillo were no longer applicable to alleged waiver of ADEA rightsSummary of this case from Gardoski v. Pats Aircraft, LLC
Argued October 11, 1996
Filed February 6, 1997
Richard Z. Freemann, Jr. (argued), Philadelphia, PA, for Appellant.
Robert M. Goldich (argued), Rachel S. Lieberman, Wolf, Block, Schorr Solis-Cohen, Philadelphia, PA,, for Appellees.
Cathy Ventrell-Monsees, Laurie A. McCann, Melvin Radowitz, American Association of Retired Persons, Washington, DC, for Amicus-Appellant American Association of Retired Persons.
C. Gregory Stewart, General Counsel, Gwendolyn Young Reams, Associate General Counsel, Vincent J. Blackwood, Assistant General Counsel, Jennifer S. Goldstein (argued), Equal Employment Opportunity Commission, Washington, DC, for Amicus-Appellant Equal Employment Opportunity Commission.
Ann E. Reesman, Robert E. Williams, Mary Chlopecki, McGuiness Williams, Washington, DC, for Amicus-Appellee Equal Employment Advisory Council.
Appeal from the United States District Court for the Eastern District of Pennsylvania, (D.C. Civ. No. 95-cv-00141).
OPINION OF THE COURT
Thomas Long, a former Sears employee, appeals an order of the district court granting summary judgment in favor of Sears in an action filed by Long based in part on the Age Discrimination in Employment Act ("ADEA"), 29 U.S.C. §(s) 621 et seq., as amended by Title II of the Older Workers Benefit Protection Act of 1990 ("OWBPA"), 29 U.S.C. §(s) 626(b). The main issue, one of first impression for us, concerns the effect of a release, drafted by Sears and executed by Long, by which Long purported to waive all claims, including those based on age discrimination, associated with his termination. The district court rejected Long's argument that the release was invalid because it failed to meet specific and detailed requirements of the OWBPA. The court declined to consider alleged deficiencies in the release, concluding that the document, even if flawed, was ratified when Long accepted and retained severance benefits paid to him following execution of the release. Reasoning that the ratified release operated to preclude all claims associated with Long's termination, the district court granted summary judgment in favor of Sears.
Because we are convinced that the ratification doctrine should not apply to a waiver of age discrimination claims which is invalid under the OWBPA and that Long should not be required to tender back severance benefits before proceeding with his age discrimination claims, we find that the grant of summary judgment with respect to these claims was inappropriate. We will, therefore, reverse in part the order of the district court relating to the ADEA claim. We will remand the non-ADEA claims for further consideration.
The relevant facts, which relate primarily to Long's employment history with Sears, are undisputed. Long, who was born in 1936, had a twenty year history with Sears where, beginning in 1964, he worked in a variety of sales capacities. From the early 1980s Long was employed in Sears' Home Improvement Products and Services Division (HIPS). As a HIPS employee, Long, at different times, sold heating and air conditioning, siding, windows, and doors, although his primary responsibility was to sell roofing. By all accounts, Long's job performance was excellent and his earnings, based on straight commission, were in the neighborhood of $100,000 per year.
In 1992, Sears analyzed the HIPS division's economic performance and concluded that reorganization was warranted. In January 1993 Sears announced that its HIPS division, with the exception of one unit, would close nation-wide. HIPS employees were told that Sears would discontinue its home improvement services permanently and that it would lay off employees not transferred to other Sears positions by mid-April. Employees allegedly were promised that every effort would be made to place them elsewhere in the Sears organization and were told that placement preference would be given to long-term HIPS employees with satisfactory performance.
In February 1993 Sears offered Long and certain other employees a reorganization package which included severance benefits. In exchange for the package, eligible employees were asked to sign a "General Release and Waiver" which read as follows:
GENERAL RELEASE AND WAIVER I
In consideration of the benefits I will receive under the Sears Closed Unit/Reorganization Severance Allowance Plan as described in the attached Benefit Notification form, I, _________________ hereby release, waive, and forever discharge officers, successors, and assigns from any and all actions, causes of action (INCLUDING, BUT NOT LIMITED TO, ACTIONS UNDER TITLE VII OF THE CIVIL RIGHTS ACT OF 1964, THE AGE DISCRIMINATION IN EMPLOYMENT ACT, STATE CIVIL RIGHTS STATUTES, AND THE AMERICANS WITH DISABILITY ACT), damages or claims of damage of every character whatsoever by reason of my employment with Sears, whether known or hereafter discovered, including, but not limited to, my termination from Sears.
I have read this General Release and Waiver and understand all of its terms. I have signed it voluntarily with full knowledge of its legal significance. I have been given the opportunity to consult with an attorney but have chosen not to do so.
Date: ______________ /s/ ______________________
Written in capital letters across the top of the release was the following: "DO NOT SIGN THIS UNTIL YOU HAVE READ THE ATTACHED NOTICE."
The notice attached bore a heading which read: "IMPORTANT NOTICE: THIS NOTICE IS BEING PROVIDED TO SATISFY THE REQUIREMENTS OF THE OLDER WORKERS BENEFIT PROTECTION ACT." The notice itself provided that: 1) an employee would have up to forty-five days from receipt of the severance package to decide whether to sign the release; 2) the release was revocable for up to seven days following its execution and no severance payments would be made until this seven day period had passed; 3) a list was attached showing the birth dates and job titles of those to whom the package had been offered; and 4) if applicable, a list was attached which included a list of employees deemed ineligible to receive the package. The notice also contained a provision suggesting that the employee consult an attorney prior to signing the release and clarified that rights which might arise in the future would not be waived by signing the release.
Although he now alleges that he did not understand the terms or significance of the release, Long signed the document on March 18, 1993. He contends that he was pressured by his supervisor to sign and did so, in part, based upon his confidence that he would eventually be placed elsewhere in the Sears organization. Although he actively sought a transfer, Long was not offered another position; his last day of work was April 9, 1993.
Because Sears placed Long on a one-year leave of absence, his actual date of termination was April 8, 1994.
On March 8, 1994, after having signed the release and receiving more than $39,000 in severance benefit payments, Long filed charges of discrimination with the Equal Employment Opportunity Commission and the Pennsylvania Human Relations Commission alleging violations of the Age in Discrimination in Employment Act of 1967, 29 U.S.C. §(s) 621-634, and the Pennsylvania Human Relations Act of 1955 ("PHRA"), 43 Pa. Cons. Stat. Ann. Section(s) 951-863. Long claimed that he was discriminated against based on his age because, following his layoff, Sears "retrained younger employees with less seniority" to work in other departments and, contrary to its representation, Sears "continued its Home Improvement Operations." On January 10, 1995, Long filed a complaint in the district court claiming age discrimination under the ADEA and the PHRA. He also alleged that Sears violated Section 510 of the Employment Retirement Security Act of 1974 ("ERISA"), 29 U.S.C. §(s) 1140, by terminating his employment to avoid further accrual and payment of pension benefits, and asserted state common law claims.
It is undisputed that Long was not otherwise entitled to receive these payments from Sears. Unlike the Sears general pension plan, the severance package offered to Long provided for twenty-six weeks of pay in addition to other benefits. Sears allegedly failed to inform Long that he could opt to accept a less generous severance package without executing a release.
In October, 1995, Long amended his complaint to add a claim alleging that Sears discriminated against him on the basis of age after the date of his termination when it failed to rehire him.
In response to Long's complaint, Sears filed an answer raising, among others, the following affirmative defense: "Plaintiff has waived and released all claims against [Sears]" and "Plaintiff ratified his waiver and released all claims against [Sears] by his acceptance of and failure to return his severance payment." Thereafter, Sears filed a motion for summary judgment, alleging that Long's claims were barred by the release. Specifically, Sears argued that the release satisfied the requirements of the OWBPA and that, in any event, Long had ratified the release, making it enforceable despite any statutory deficiencies. Long opposed this motion with facts alleged to demonstrate discrimination and filed a cross-motion for summary judgment contending that the release was invalid for failure to comply with the requirements of the OWBPA and was void and unenforceable because it had been obtained by fraud. Long agreed to credit severance pay received against any damages awarded.
Specifically, Long alleges that the release violated the OWBPA in the following respects: 1) it purported to bar future claims; 2) the release was not accompanied by the required data relating to other Sears employees; and 3) the release materials were incomprehensible to Long and other employees.
On March 1, 1996, the district court granted summary judgment in favor of Sears on all claims without addressing Sears' compliance with the OWBPA. While the court recognized that "[w]hether the Sears release meets the facial requirements of the OWBPA is a question of fact not resolvable here by summary judgment," it concluded that summary judgment was, nonetheless, appropriate. In reaching this conclusion the district court relied on authority holding that releases which fail to conform to the OWBPA are merely voidable and, under traditional principles of contract law, may be ratified by retention of benefits received. Citing the decision in Wamsley v. Champlin Refining and Chemical, Inc., 11 F.3d 534 (5th Cir. 1993), the district court concluded that the ratification doctrine had survived the enactment of the OWBPA and operated to bar Long's claims:
Because Long retained, and did not offer to return more than $39,000 paid in consideration of a Release that he suspected was defective, we hold that he ratified that Release and is precluded from pressing claims arising from his termination by Sears.
Two days prior to trial the district court, in a pre-trial conference, signaled its intent to grant summary judgment in favor of Sears. Understanding that the court's decision turned on his retention of benefits, Long submitted a brief in which he offered to tender the money. This brief and the court's order granting summary judgment were docketed the same day. Apparently, they "crossed in the mail."
Our review of the district court's grant of summary judgment is plenary. "[B]ecause the facts are undisputed, we decide [this] appeal as a matter of law." DiBiase v. SmithKline Beecham, 48 F.3d 719 (3d Cir.), cert. denied, 116 S.Ct. 306 (1995). The legal questions before us are straightforward: (1) Are releases of ADEA claims which fail to conform to the requirements of the OWBPA enforceable or can they be rendered enforceable, or ratified, by an employee's acceptance and retention of severance benefits? (2) Where ratification does not apply, does retention of severance benefits operate nonetheless to prevent an employee from pursuing a claim under the ADEA? (3) Where a release of ADEA claims is invalid under the OWBPA and does not, therefore, release the employer from liability for the ADEA claims, is the release nonetheless effective in insulating the employer with respect to non-ADEA claims covered by the release? The district court granted summary judgment in favor of Sears, predicting that we would hold that Long, by retaining severance benefits, ratified the allegedly defective release and, as a result, is precluded from pursuing a claim under the ADEA. This prediction was based, in part, on the decision in Ponzoni v. Kraft General Foods, 774 F. Supp. 299 (D.N.J. 1991), aff'd, 968 F.2d 14 (3d Cir. 1992). In Ponzoni, a pre-OWBPA case, we affirmed without opinion the district court's determination that a release of ADEA claims which was not knowing and voluntary could be enforced, nonetheless, because it had been ratified through retention of benefits. As we will explain, we conclude that the enactment of the OWBPA changed the legal landscape with respect to the release of ADEA claims. In light of the law as it now stands, we conclude that the ratification doctrine does not apply to ADEA releases which fail to comply with the OWBPA. As a result, the district court erred in granting Sears' motion for summary judgment. This error is rooted in the fact that the law with respect to employee releases which fail to comply with the OWBPA is unsettled.
Long also asks that we determine whether the release at issue waived future claims in violation of the OWBPA. Because the district court found it unnecessary to determine whether the release did, in fact, fail to comply with the OWBPA, our consideration of this issue would be premature. Our disposition of this appeal will require that the district court assess the merits of this argument, as well as Long's other allegations of statutory deficiency.
The decision in Ponzoni does not control the outcome of this case as our affirmances, without opinion, of district court decisions are not binding precedent. See Ransom v. Marrazzo, 848 F.2d 398, 411 (3d Cir. 1988).
In order to place the issues raised here in context we turn to the language and legislative history of the OWBPA.
The Older Workers Benefit Protection Act became effective on October 16, 1990, as an amendment to the ADEA. Its purpose was two-fold: to "make clear that discrimination on the basis of age in virtually all forms of employee benefits is unlawful," and to "ensure that older workers are not coerced or manipulated into waiving their rights to seek legal relief under the ADEA." S. Rep. No. 263, 101st Cong., 2d Sess. 2 (1990). Congress' concern over employee waiver of rights under the ADEA was summarized as follows:
[E]arly retirees or employees offered the chance to participate in exit incentive or other group termination programs can effectively be forced to waive their right to file a claim when the employer conditions such participation on the signing of a waiver. The problem is particularly acute in large-scale terminations and lay-offs, where an individual employee would not reasonably be expected to know or suspect that age may have played a role in the employer's decision, or that the program may be designed to remove older workers from the labor force. The preemptive waiver of rights occurs before a dispute has arisen and indeed before an employee is even aware of any potential or actual pattern of discrimination. Such a pre-emptive waiver may also preclude the employee from asserting claims that arise out of subsequent discriminatory conduct by the employer, e.g. hiring younger workers to replace the terminated older workers. These waivers are both unfair and inconsistent with the intent of the ADEA.
This language was adopted by reference in S. Rep. No. 263, 101st Cong. 2d Sess. 15 (1990).
Age discrimination victims typically earn more than the minimum wage, but their average annual income is only $15,000. Moreover, once out of work, these older Americans have less than a 50/50 chance of ever finding new employment. The[y] often have little or no savings, and may not yet be eligible for Social Security. Accordingly, it is reasonable to assume that many employees would be coerced by circumstances into accepting significant compromises. This is especially true where employees are unable even to recognize the potential of their claims because no dispute exists between them and their employer.
Id. In an effort to protect older workers, Title II of the OWBPA defined those circumstances in which ADEA waivers would be permitted.
Prior to the OWBPA, the general approach of the courts was to find waivers permissible subject to a requirement that they be made "knowingly" and "willfully." "Under the [OWBPA,] waivers must [still] be made knowingly and voluntarily. However, they cannot be deemed to be so unless several statutory minima are met." Pellicciotti, Older Workers Benefit Protection Act of 1990: Congress Responds to Betts Decision, 35 Res Gestae 114, 115 (1991). These minimum requirements are set forth at 29 U.S.C. Section(s) 626(f):
(1) An individual may not waive any right or claim under this chapter unless the waiver is knowing and voluntary . . . . [A] waiver may not be considered knowing and voluntary unless at a minimum —
(A) the waiver is part of an agreement between the individual and the employer that is written in a manner calculated to be understood by such individual . . .
(B) the waiver specifically refers to rights or claims arising under this chapter;
(C) the individual does not waive rights or claims that arise after the date the waiver is executed;
(D) the individual waives rights or claims only in exchange for consideration in addition to anything of value to which the individual already is entitled;
(E) the individual is advised in writing to consult with an attorney prior to executing the agreement;
(F) . . . (ii) if a waiver is requested in connection with an exit incentive or other employment termination program offered to a group or class of employees, the individual is given . . . at least 45 days within which to consider the agreement;
(G) the agreement provides that for a period of at least 7 days following the execution of such agreement, the individual may revoke the agreement;
(H) if a waiver is requested in connection with an . . . employment termination program . . . the employer . . . notifies the individual in writing in a manner calculated to be understood . . . as to —
(i) any class, unit or group of individuals covered by such program, any eligibility factors for such program, and any time limits applicable . . .; and
(ii) the job titles and ages of all individuals eligible or selected for the program, and the ages of all individuals in the same job classification or organizational unit who are not eligible or selected. . . .
The statute also specifies that the party asserting the validity of the waiver bears the burden of showing that the waiver was indeed knowing and voluntary. 29 U.S.C. §(s) 626(f)(4).
While Congress carefully defined what must be included in a knowing and voluntary waiver under the ADEA, it did not characterize the legal effect of a release which fails to satisfy the statutory requirements. Courts asked to determine whether deficient waivers should be enforced have purported to rely on "clear" congressional intent. They have reached different conclusions, however, with respect to whether Congress intended that waivers deficient under the OWBPA be declared void or merely voidable.
The view that deficient releases are void was adopted by the Court of Appeals for the Seventh Circuit in Oberg v. Allied Van Lines, Inc., 11 F.3d 679 (7th Cir. 1993). The court in Oberg held that releases which fail to conform to the OWBPA have no legal significance; they cannot be ratified or enforced. Relying on OWBPA language that "[a]n individual may not waive any right or claim . . . unless the waiver is knowing and voluntary," the court concluded that, "unless a waiver contract takes the form required by the statute, an employer and an employee cannot contract to waive the ADEA provisions. . . . No matter how many times parties may try to ratify such a contract, the language of the OWBPA . . . forbids any waiver." Id. at 682. In light of the OWBPA's unambiguous statutory prohibition against waiver where one or more of the statutory requirements are absent, the court concluded that further analysis was unwarranted:
On January 9, 1997, the Court of Appeals for the Sixth Circuit issued its decision in Raczak v. Ameritech Corp., No. 95-1082, 1997 WL 5921 (6th Cir. Jan. 9, 1997). Two panel members, on divergent grounds, agreed that the plaintiffs were not precluded from pursuing an action under the ADEA despite having executed waivers deficient under the OWBPA and retained payments made pursuant to these waivers. One of these panel members adopted the ratification analysis set forth in Oberg while the other concurred in the result only, concluding that "there was not a total failure of consideration when the ADEA claim was instituted without plaintiffs tendering back benefits received . . ." and "reserv[ing] judgment as to whether a release limited to ADEA claims would require a different result." Id. at *15. The remaining panel member adopted an analysis similar to that undertaken by the district court in this case.
This decision, fractured though it is, means that the appellate courts, prior to our opinion, are evenly divided on the issue of tender back.
The common law doctrine of ratification results in the enforcement of "a promise to perform all or part of an antecedent contract of the promisor, previously voidable by him, but not avoided prior to the making of the promise." Restatement (Second) of Contracts Section(s) 85 (1981). Promises that are void cannot be ratified. Void promises are not legally binding, have no legal effect and, therefore, are not contracts. Id. Section(s) 7 cmt. a.
When the plain text of a statute is clear then "courts must presume that a legislature says in a statute what it means and means in a statute what it says. When the words of a statute are unambiguous then, this first canon is also the last: `judicial inquiry is complete.'" Connecticut Nat. Bank. v. Germain, 503 U.S. 249, 253-54 (1992).
Id. at 682 (citations omitted).
Having pronounced non-conforming releases void, the court next considered whether an employee is required to tender back severance benefits in order to maintain suit under the ADEA. The court concluded that employees "were not required to tender back their severance benefits before filing ADEA claims, notwithstanding [the] previously executed waiver of all claims . . . ." Id. at 684.
Because void releases cannot be ratified, the court did not consider whether failure to tender back might amount to ratification.
The Courts of Appeals for the Fourth and Fifth Circuits have rejected the Oberg approach, holding instead that defective waivers are merely voidable and are, therefore, subject to ratification by an employee's retention of severance benefits. See Wamsley v. Champlin Refining and Chemicals, Inc., 11 F.3d 534 (5th Cir. 1995); and Blistein v. St. John's College, 74 F.3d 1459 (5th Cir. 1996). In Wamsley the court wrote:
The Oberg analysis has been adopted by a majority of district courts outside the Seventh, Fourth and Fifth Circuits. See EEOC v. Sara Lee Corp., 923 F. Supp. 994 (W.D. Mich. 1995) (no ratification; where waiver is deficient under the OWBPA, employer and employee cannot contract to waive ADEA provisions); Elliott v. United Technologies Corp., 94 CV 01577, slip op. (D. Conn. March 24, 1995) (ratification at odds with plain language of the OWBPA); Soliman v. Digital Equipment Corp., 869 F. Supp. 65 (defective release cannot be ratified; tender back will chill bringing of meritorious claims) (D, Mass, 1994); Raczak v. Ameritech Corp., 1994 WL 78099 (E.D.Mich. 1994), rev'd in part, No. 95-1082, 1997 WL 5921 (6th Cir. Jan. 9, 1997), (Oberg approach better reasoned); Carr v. Armstrong Air Conditioning, 817 F. Supp. 54 (N.D. Ohio 1993) (plaintiff waived no rights since severance agreement violated the OWBPA; tender requirement not consistent with purposes of the ADEA); Pierce v. Atchison, Topeka and Santa Fe Railway Co., 1993 WL 18437 (N.D. Ill. 1993) (OWBPA requirements preclude waiver by ratification); Collins v. Outboard Marine Corp., 808 F. Supp. 590, 594 (N.D. Ill. 1992) (scope of defective release did not include claim under ADEA; consideration received need not be returned since it was not paid for relinquishment of ADEA claim); Issacs v. Caterpillar, Inc., 765 F. Supp. 1359 (C.D. Ill. 1991) (ratification and tender-back not applicable to releases deficient under the OWBPA). But see Hodge v. New York College of Podiatric Medicine, 940 F. Supp. 579 (S.D.N.Y. 1996) (Oberg rejected; defective releases subject to ratification); and Rivers v. Northwest Airlines, Inc., 71 Fair Empl. Prac. Cas. (BNA) 1217 (E.D.Mo. 1995) (failure to tender back benefits precluded ADEA claim even where release was deficient under OWBPA).
We do not interpret the language of section 626(f)(1) [of the OWBPA] to mean that a waiver which fails to meet the requirements of subsections (A) through (H) is void of legal effect. Rather, we interpret it to mean that such waivers are not knowing and voluntary and thus are subject to being avoided at the election of the employee . . . .
Therefore, we hold that neither the language nor the purpose of the OWBPA indicates a congressional desire to deprive an employee of the ability to ratify a waiver that fails to meet the requirements of the OWBPA. When [the employees] chose to retain and not tender back to [the employer] the benefits paid them in consideration for their promise not to sue . . ., they manifested their intention to be bound by the waivers and thus, made a new promise to abide by their terms.
Id. at 539-40. In part, the court's conclusions were drawn from the legislative history of the OWBPA:
The legislative history indicates that the fundamental purpose of the OWBPA waiver provisions is to ensure that an older worker who is asked to sign an ADEA waiver does so in the absence of fraud, duress, coercion, or mistake of material facts. S. Rep. No. 101-263, 101st Cong., 2d Sess. (1990) . . . . The circumstances against which these provisions were designed to protect are the same circumstances that have traditionally given rise to grounds upon which a party can avoid contractual obligations. That the Committee enumerated several of the traditional grounds of avoidance is significant. Also significant is the absence of any language in the statute and any statement in the legislative history indicating that a waiver executed in contravention of the OWBPA requirements is void . . . and cannot be ratified . . . .
Id. The court also cited section 626(f)(1)(G) of the Act which provides that an employee may revoke a waiver for any reason within seven days of its execution and that the agreement is not enforceable during that seven-day period. "If non-compliance with the other subparts of section 626(f)(1) rendered the agreement void, there would be no need for subpart (G)." Id. at 539. According to the court, declaring defective waivers void:
would be inconsistent with one of the expressed purposes of the ADEA: "to help employers and workers find ways of meeting problems arising from the impact of age on employment." 29 U.S.C. § 621(b). The simplest and easiest way to further this purpose is to give effect to private agreements which resolve age-related employment problems without the inevitable delays and costs associated with litigation.
Finally, the court held that "justice and equity require the employee who seeks to avoid the obligations to which he agreed under the settlement agreement to return the considerations which he received for his promise not to sue." Id. at 541.
While the district court in the case before us adopted the reasoning underlying the decision in Wamsley, on balance we find the Oberg approach to tender back to be more consistent with the purposes underlying the OWBPA.
While we, like the court in Oberg, reject the argument that Long is precluded from maintaining an ADEA claim because he retained severance benefits, we reach this end via a different route. In both Oberg and Wamsley the courts focused first upon whether a defective release should be characterized as void or voidable. It was only after this question was resolved that the courts considered the import of the plaintiff's failure to return severance benefits. In Wamsley, the court concluded that the allegedly defective release was voidable. The plaintiff's retention of benefits served to ratify the defective release and, therefore, suit was barred. The court in Oberg reasoned that defective releases were void and that ratification could not apply. The court found it necessary, however, to address the issue of retention of severance benefits in that tender back of benefits has been viewed by some courts as a prerequisite to suit. The court in Oberg relied on Supreme Court precedent and other policy arguments to hold that tender back should not be a prerequisite to an ADEA suit where the release at issue is defective under the OWBPA.
Although the analysis in Oberg and Wamsley might suggest otherwise, resolution of the void/voidable issue does not control our disposition of this appeal. No matter how we characterize a release which fails to comply with the OWBPA — void or voidable — we must, under either theory, decide whether Long's retention of severance benefits should prevent him from pursuing his ADEA claim. Because we conclude that neither ratification nor tender back was meant to apply in the ADEA context, we need not address the void/voidable distinction.
Indeed, to resolve that question here is premature. The district court has yet to determine whether the release at issue is, in fact, defective under the OWBPA. While this case does not turn upon whether an allegedly defective release is void or voidable, we recognize that it might be necessary to reach this question under a different set of facts. For example, holding a release of ADEA claims void could have implications for non-ADEA claims also covered by the release.
The court in Oberg declared that releases which fail to conform to the OWBPA are void and, consequently, found it unnecessary to consider the ratification doctrine. Our analysis, however, requires that we consider its application. The language of the OWBPA and its legislative history convince us Congress did not intend that the ratification doctrine be invoked to enforce the terms of a deficient release.
Prior to the OWBPA, it was generally recognized that employees could waive federal ADEA rights in private settlements with their employees so long as the employees' consent to the settlement was knowing and voluntary. Even during this pre-OWBPA period, however, there was disagreement over the standard to be applied in assessing whether a waiver was, in fact, knowing and voluntary. The Courts of Appeals for the Fourth, Sixth and Eighth Circuits held that ordinary state contract law principles controlled.
See, e.g., O'Shea v. Commercial Credit Corp., 930 F.2d 358, 362 (4th Cir.), cert. denied, 502 U.S. 859 (1991); Lancaster v. Buerkle Buick Honda Co., 809 F.2d 539 (8th Cir.), cert. denied, 482 U.S. 928 (1987); and Runyan v. National Cash Register Co., 787 F.2d 1039, 1045 (6th Cir. 1986) (en banc).
We, however, elected to apply a more stringent federal "totality of the circumstances test," reasoning that "in recognition of the important interests involved . . . careful evaluation of the release form itself as well as the complete circumstances in which it was executed [is] warranted." Cirillo v. Arco Chemical Co., 862 F.2d 448, 450 (3d Cir. 1988). Factors relevant in assessing the totality of the circumstances included, but were not limited to, the following:
The totality of the circumstances test was also adopted by the Court of Appeals for the Fifth Circuit in O'Hare v. Global Natural Resources, Inc., 898 F.2d 1015, 1017 (5th Cir. 1990); and by the Court of Appeals for the Second Circuit in Bormann v. ATT Communications, Inc., 875 F.2d 399, 403 (2d Cir. 1989).
It is significant that even prior to enactment of the OWBPA, we concluded that Congress' ordinary contract principles were not sufficient to vindicate the protective purpose of the ADEA. Cirillo, 862 F.2d at 450. It is difficult, therefore, to understand the dissent's "urging" that we retreat to traditional common law principles in order to decide this case.
(1) the clarity and specificity of the release language; (2) the plaintiff's education and business experience; (3) the amount of time plaintiff had for deliberation about the release before signing it; (4) whether plaintiff knew or should have known his rights upon execution of the release; (5) whether plaintiff was encouraged to seek, or in fact received benefit of counsel; (6) whether there was an opportunity for negotiation of the terms of the Agreement; and (7) whether the consideration given in exchange for the waiver and accepted by the employee exceed[ed] the benefits to which the employee was already entitled by contract or law.
Id. at 451.
In enacting the OWBPA, Congress resolved this dispute regarding the standard to be applied in determining whether a release is "knowing and voluntary." Congress rejected the applicability of common law contract principles and declined to embrace even the more demanding "totality of the circumstances" test:
Even the decisions that have followed the more protective "totality of the circumstances" approach . . . have not held that certain protective factors must be present . . . . The [OWBPA], by contrast, will limit unsupervised waivers to certain situations and then spell out clear and ascertainable standards to govern those situations.
S. Rep. No. 79, 101st Cong., 1st Sess. 17 (1989) (emphasis added).
The requirements established in order for releases to be "knowing and voluntary" under the OWBPA clearly exceed the protections available under the common law. The right to seek counsel, the 45-day consideration period, the seven-day right of revocation, and the provision of detailed information about those affected by group terminations are all protections which were unavailable under the "knowing and voluntary" standard of the common law. In addition, the OWBPA establishes that waivers must be in writing, must specifically refer to rights under the ADEA, and are subject to "average person" standard; whether a particular plaintiff understood the implications of the release is irrelevant. The OWBPA also effected a shift in the burden of proof applicable in judicial proceedings. Where, under the common law, the employee challenging a waiver was required to show that the waiver was not "knowing and voluntary," the OWBPA shifts this burden to the employer. See 29 U.S.C. §(s) 626(f)(3). These requirements, too, overrode pre-OWBPA decisions applying common law principles. Most importantly, Congress, after grappling with the question of whether to permit ADEA waivers at all, stated unequivocally that unless the enumerated requirements are met, an individual "may not waive" ADEA rights. 29 U.S.C. Section(s) 626(f)(1) (emphasis added).
See Taylor v. Gordon Flesch Co., 793 F.2d 858, 862 (7th Cir. 1986) (oral waivers enforceable); Runyan v. National Cash Register Corp., 787 F.2d 1039 (6th Cir. 1986) (en banc), 479 U.S. 850 (1986) (if plaintiff understood release, whether anyone else would have understood it was not controlling); Lancaster v. Buerkle Buick Honda Co., 809 F.2d 539 (8th Cir. 1987) (release enforced despite failure to refer to ADEA); and Harrison v. Arlington Ind. School Dist., 717 F. Supp. 453, 455 (N.D. Tex.), aff'd without op., 891 F.2d 904 (5th Cir. 1989) (under common law, employee challenging waiver on "knowing and voluntary" grounds carried burden of proof).
We are convinced that in enacting the OWBPA, Congress intended to occupy the area of ADEA releases and, in doing so, to supplant the common law; the OWBPA was enacted to "establish a floor, not a ceiling." Soliman v. Digital Equipment, 869 F. Supp. 65, 68 n. 12 (D. Mass. 1994). Enforceability of a waiver is made contingent upon the presence of certain enumerated factors.
Given the clear and specific goals of the OWBPA, we cannot accept that Congress intended that the common law doctrine of ratification be applied to releases invalid under the OWBPA. The common law fiction of a "new promise" forged from retention of benefits has no place in this statutory scheme. To conclude otherwise would be to say that Congress only intended that the OWBPA requirements apply to the "first" waiver. Such a result is inconsistent with the aims of the Act and works a hardship on employees who could not have known that by retaining severance pay they were, in effect, declining the protection of the OWBPA. This conclusion applies as well to the common law concept of tender back of benefits as a pre-requisite to suit under the ADEA.
It is significant that neither the text of the OWBPA nor its legislative history discusses ratification as a bar to suit; there is no mention of the ratification doctrine or reference to caselaw invoking that doctrine. In fact, at the time of the OWBPA's enactment, not a single court of appeals had held that a waiver which was not knowing and voluntary could, nonetheless, be enforced pursuant to a ratification theory. Many courts, however, including ours, had resolved waiver questions without reference to the doctrine. See O'Hare v. Global Natural Resources, Inc., 89 F.2d 1015 [ 898 F.2d 1015], 1017 (5th Cir. 1990); Bormann v. ATT Communications, Inc., 875 F.2d 399, 403 (2d Cir. 1989); Cirillo v. Arco Chem. Co., 862 F.2d 448, 451-55 (3d Cir. 1988); Coventry v. United States Steel Corp., 856 F.2d 514, 523 (3d Cir. 1988); Lancaster v. Buerkle Buick Honda Co., 809 F.2d 539, 541 (8th Cir. 1987); Runyan v. National Cash Register Corp., 787 F.2d 1039, 1044 (6th Cir. 1986) (en banc). Several of these cases are cited in the OWBPA's legislative history. See H.R. Rep. No. 664, 101st Cong., 2d Sess. 26-27 (1990). Furthermore, we believe that it would be wrong to conclude, as does the dissent, that simply because ratification is not rejected in the text or legislative history of the OWBPA that the common law is unchanged. The Act establishes that validity and enforceability are inextricably linked; adherence to the terms of the Act is fundamental to enforcement. The critical point is not that the Act fails specifically to do away with ratification. It is instead that the ratification doctrine is logically inconsistent with the specific terms of the OWBPA.
We are not persuaded by the dissent's argument that "the policy of the OWBPA requires that an employee should be able to ratify a defective release" (Typescript at 4), nor do we believe that the dissent's "dramatic" example purporting to illustrate this conclusion is apposite. The OWBPA was written to govern waivers of the right to file a claim. This "pre-emptive waiver of rights occurs before a dispute has arisen and indeed before an employee is even aware of any potential or actual pattern of discrimination." S. Rep. No. 79, 101st Cong., 1st Sess. 9 (1989). We certainly have not suggested, nor, as far as we know, is there any authority for the proposition that the OWBPA might apply to the terms of a settlement agreement forged after the taking of evidence in a civil trial.
In any event, it is a given that even well-designed remedial legislation may result in unintended consequences. Suffice it to say that we are not aware, nor was counsel for the EEOC when questioned at oral argument, of any case where an employer has sought to avoid its obligations under a severance agreement by invoking its own failure to comply with the OWBPA. That case, should it ever arise, is a matter for a different day; we need not decide this case on the basis of what might happen in a hypothetical case which might come before us at some point in the future.
"States that require a tender to challenge a release sometimes use the language `condition precedent to suit' and sometimes use the language of `ratification.' But there is no meaningful difference between the two." Issacs v. Caterpillar, Inc., 765 F. Supp. 1359, 1372 (C.D. Ill. 1991).
In any event, the legislative history indicates that Congress was aware of the benefit retention issue and chose not to include in the OWBPA a proviso prohibiting this retention. The minority members of the Committee on Education and Labor referred to a letter from IBM which mentioned the possibility that former employees might retain benefits while pursuing an ADEA claim. H.R. Rep. No. 664 at 81 proposed a substitute bill containing the following provision: "If a waiver is set aside for any reason, any damages received through a discrimination action shall be offset by the consideration received for the waiver." Id. at 200. This proposal did not mention ratification or tender back but was a moderate response to the possibility of retention of benefits. It was, nonetheless, rejected. Id. at 29.
Although our rejection of the ratification and tender back theories could rest alone on the language and legislative history of the OWBPA, we find additional support for our position in Supreme Court precedent and the caselaw interpreting that precedent.
In the leading case rejecting a tender back requirement under the ADEA, Oberg, the Court of Appeals for the Seventh Circuit rejected the employer's argument that "Plaintiffs must tender back the consideration received for executing . . . Severance Agreements" in order to maintain a claim under the ADEA. 11 F.3d 679, 683. Recognizing that the requirement might initially seem "appealing under common law notions of fairness," the court nonetheless held that the Supreme Court's decision in Hogue v. Southern R. Co., 390 U.S. 516 (1968), compelled a different result. Id.
The dissent argues that the holding in Oberg was undermined by the decision in Fleming v. United States Postal Service AMF O'Hare, 27 F.3d 259 (7th Cir. 1994). In Fleming the court of appeals refused to extend the reasoning of Hogue to a Title VII case, limiting Hogue to the "context of a federal statute that regulates releases, displacing common law rules." Id. at 261-62. The OWBPA is precisely such a statute.
In Hogue, a brief per curiam opinion, the Supreme Court considered the tender back requirement in a case arising under the Federal Employer's Liability Act ("FELA"), 45 U.S.C. Section(s) 51 et seq. (1939). Stating that the tender back question was to be resolved under federal rather than state law, the Court concluded that requiring a refund would be "wholly incongruous with the general policy of the [FELA]." Id. at 517 (citation omitted). The court in Oberg summarized the Hogue decision as follows:
In Hogue, the Court rejected any notion that state common law principles could help resolve the tender back question in FELA cases. The Court stated that "[t]he question whether a tender back of the consideration was a prerequisite to the bringing of the suit is to be determined by federal rather than state law." The Court went on to hold that an employee, who previously executed an employer release, need not, as a precondition to bringing suit under FELA tender back to his employer any of the consideration he received for executing the release. The Court did, however, state that the benefits paid should be deducted from any award to the employee.
Id. at 683-84 (citations omitted). The Court in Oberg was "convinced that . . . analogizing the policy of [the] ADEA to that of [the] FELA, and thus applying Hogue, [was] correct." Id. at 684. We agree, based on a number of factors.
First, we note that courts have regularly applied the analysis in Hogue to reject tender requirements in lawsuits brought under a variety of federal remedial statutes. It is impossible to view the ADEA as anything other than a federal remedial statute. The ADEA was enacted in order to further the dual goals of compensating discrimination victims and deterring employers from practicing discrimination. As the Supreme Court wrote in McKennon v. Nashville Banner Publishing Co., 115 S.Ct 879, 884 (1995): "The private litigant who seeks redress for his or her injuries vindicates both the deterrence and compensation objectives of the ADEA." In light of this clear recognition of the purpose of the ADEA, we are confident that the tender back rule rejected in suits under the FELA should be rejected in suits under the ADEA as well. Imposing a tender back rule in the ADEA context would almost certainly compromise the purposes underlying that statute. The concerns expressed by our sister court in the pre-OWBPA case of Forbus v. Sears Roebuck Company, 958 F.2d 1036, 1041 (11th Cir. 1992), apply with equal force here:
See Botefur v. City of Eagle Point, 7 F.3d 152, 156 (9th Cir. 1993) (Hogue generalizable to other federal compensatory statutes including Title VII); Home Box Office, Inc. v. Spectrum Electronics, Inc., 100 F.R.D. 379, 382 n.1 (E.D.Pa. 1991) (not citing Hogue but holding that under antitrust law benefits available under federal law cannot be defeated by state common law rules; no ratification where plaintiffs failed to tender back); Washner v. American Motors Sales Corp., 597 F. Supp. 991 (E.D.Pa. 1984) (not citing Hogue but finding Pennsylvania law with respect to ratification releases incongruous with Automobile Dealers' Day in Court Act where Act was intended to provide redress for the very activity alleged; amount retained was to be set off against any damages); Smith v. Pinell, 597 F.2d 994, 996 (5th Cir. 1979) (plaintiff allowed to proceed under Jones Act despite having signed release and received settlement; Jones Act analogized to FELA); Taxin v. Food Fair Stores, Inc., 287 F.2d 448 (3d Cir. 1961) (cases regarding necessity for tender are in hopeless confusion; in case involving Sherman Act, plaintiffs not required to tender back consideration received for release).
In reaching this conclusion, we are mindful of the fact that other courts have rejected Hogue's applicability to ADEA releases. In Wamsley, for example, the court of appeals wrote:
[Hogue] is founded on the recognition that a "tender back" requirement would be "wholly incongruous" with the right of recovery provided under the FELA and inconsistent with the objectives of the act. The right of recovery under the FELA, however, is unique in that it advances a congressional intention of facilitating recovery by injured railroad workers against their employers.
11 F.3d 534, 540. The Court concluded that by eliminating the tender requirement in connection with FELA releases, "Congress advanced the FELA's purpose of providing liberal recovery to injured rail workers . . . . No such purposes underlie the ADEA." Id. at 542.
There is no question that the FELA and ADEA are not identical in purpose. Nor, in our view, need they be in order for the rule in Hogue to apply. The mandate of Hogue is that tender back requirements imposed in connection with the release of federal rights be evaluated in light of the general policy of the statute in question. That the ADEA as amended by the OWBPA serves a purpose distinct from that underlying the FELA does not change the fact that a tender back requirement is "wholly incongruous" with the general policies of the ADEA and the OWBPA. In enacting the OWBPA, Congress specifically regulated ADEA releases in order to provide employees with protection not available at common law. To strip them of this protection through application of the common law principle of tender back would be anomalous indeed. "When federal law limits a class of releases, . . . the common law requiring tender . . . may have to give way." Fleming v. U.S. Postal Service, 27 F.3d 259, 260 (7th Cir. 1994).
The Court in Hogue found that a tender requirement would deter meritorious challenges to releases in FELA lawsuits. The same deterrence factor applies to ADEA claims. Forcing older employees to tender back their severance benefits in order to attempt to regain their jobs would have a crippling effect [sic] on the ability of such employees to challenge releases obtained by misrepresentation or duress. Such a rule would . . . encourage egregious behavior on the part of employers in forcing certain employees into early retirement for the economic benefit of the company. The ADEA was specifically designed to prevent such conduct, and we reject a tender requirement as a prerequisite to instituting a challenge to a release in an ADEA case.
Second, any doubt about whether Hogue's rejection of the tender back rule should apply to ADEA claims was resolved with the enactment of the OWBPA. The very specific requirements of this Act, considered against the background of its legislative history, demonstrate that Congress intended to provide protections unavailable at common law; congressional focus extended beyond ensuring that ADEA releases were untainted by fraud, duress, or some other defect recognized at common law. Congress explicitly stated that it intended to protect employees who might waive rights under the ADEA before they were "even aware of any potential or actual pattern of discrimination." H.R. Rep. No. 664 at 23. See also S. Rep. No. 263, 101st Cong. 2d Sess. 32 (1990) (recognizing "need for adequate information before waivers are signed."). Imposing a tender requirement where a release is defective under the OWBPA would effectively eviscerate that act.
The Court of Appeals for the Fifth Circuit in Wamsley, 11 F.3d at 1539, rested its conclusion that common law contract principles should apply to ADEA releases on one portion of the OWBPA's legislative history which provides that the primary purpose underlying the OWBPA was to ensure that workers executing ADEA waivers do so "in the absence of fraud, duress, coercion, or mistake of material facts. S. Rep. No. 263, 101st Cong., 2d Sess. (1990), reprinted in 1990 U.S.C.C.A.N. 1509, 1539." The court reasoned that these concerns were the same as those "that have traditionally given rise to grounds upon which a party can avoid contractual obligations." Id. The court's reliance on this single statement fails to take into account the fact that both the provisions of the statute and other statements in the legislative history clearly establish that Congress rejected and intended to move beyond application of common law principles.
Through the OWBPA Congress sought to insure that employees faced with deciding whether to sign an ADEA waiver and forego an ADEA claim be provided with sufficient information to allow them to evaluate the merits of that claim. Applying a ratification-tender back rule would require employees in Long's position, who arguably did not receive the required information, to make a similarly uninformed choice. These employees would be forced by the ratification and tender back doctrines to decide, in the continued absence of information, whether to surrender severance pay or waive all claims under the OWBPA. Employees whose releases are defective under the OWBPA would be no better off than before the OWBPA was enacted; they could be forced to make critical decisions without information deemed essential by Congress.
The choice which the Wamsley approach, adopted by the district court, places before older employees amounts to no choice at all: pursue your claim at the risk of your livelihood. Testimony before Congress established that older workers facing termination "could not afford" to do without separation benefits. Age Discrimination in Employment Waiver Protection Act of 1989: Hearing on 5.54 Before the Subcomm. on Labor of the Senate Comm. on Labor and Human Resources, 101st Cong., 1st Sess. 61 (1989) (Testimony of Robert Patterson). Other excerpts from the OWBPA's legislative history emphasize the economic plight of older workers and Congress' intention to better equip them to make choices directly affecting their rights and livelihood. Applying ratification or a tender back requirement in the circumstances presented here would mean that:
[n]o matter how egregiously releases might violate the requirements of the [OWBPA], employees would be precluded from challenging them unless they somehow . . . come up with the money they were given when allegedly forced into retirement.
Issacs v. Caterpillar, Inc., 765 F. Supp. 1359, 1367 (C.D. Ill. 1991).
Courts which have applied tender-ratification principles to ADEA releases which fail to conform to the OWBPA have rendered the OWBPA meaningless. The gist of these holdings is, as Long argues, "to uphold flawed ADEA releases [to] block discrimination claims [rather] than . . . require employers to comply fully with federal law." (Long. br. at 38).
Application of the tender-ratification doctrines and rejection of Hogue stem, we believe, from an incomplete analysis of the equities involved in allowing an employee to retain severance benefits while pursuing ADEA claims. The court in Wamsley expressed concern that were ratification and tender back held not to apply, employees would face "continued litigation with opponents who could use, and very possibly already have used, to finance their suit, the very funds paid as consideration to avoid litigation." 11 F.2d at 539. Amicus, Equal Opportunity Advisory Council, frames the concern even more starkly, warning that abrogation of the tender-ratification doctrines, "improperly encourages nefarious plaintiffs to secure the benefits awarded for signing a release and then press for more gains through a legal challenge." (Br. at 7). While this argument, although stated in extreme terms, is not without merit, it misperceives the economic reality for many older workers. The congressional assumptions underlying the OWBPA posit that most covered employees need severance benefits to fund living rather than legal expenses. Employees with baseless claims have strong financial incentives to keep severance payments rather than risk them in prolonged litigation.
The Equal Opportunity Advisory Council, in an amicus brief filed on behalf of Sears, identifies itself as an organization comprised of nearly 300 major U.S. corporations and several industry associations. The organization exists "to promote sound approaches to the elimination of employment discrimination."
Neither will rejecting ratification and tender back principles mean that employees will receive a "double recovery" by first accepting a severance payment and later winning a judgment. An employer found liable will be entitled to a set-off of any severance benefits paid. Oberg, 11 F.3d at 684. In any event, the windfall argument cuts two ways. Presumably, employers offer severance packages, in part, in exchange for the employees' release of claims. Where an employee must tender severance benefits prior to suit, it is very difficult to return that employee to his pre-release position. He is not restored to employment, the employer may still assert the release as an affirmative defense, and there is no guarantee that the employee will receive the information to which he was entitled under the OWBPA. "Such an exchange would arguably unjustly enrich the employer." Issacs, 765 F. Supp. 1367.
The equities associated with applying the logic of Hogue to eliminate the tender-ratification rule in the ADEA context are at least in equipoise. In this circumstance, we will give effect to what we believe Congress intended: the OWBPA was designed to protect employees negotiating with employers, not to protect employers from overreaching plaintiffs. Employers are, by far, in a better position to protect their own interests than are older employees. Employers should not need the ratification doctrine in order to ensure that their releases are effective; they need to comply with the OWBPA. Most OWBPA requirements are clear and specific, and, once these requirements are met, waivers executed by employees will be valid and enforceable. If prodded by necessity into following the mandates of the OWBPA, employers will have purchased a valid affirmative defense against suit.
We are not swayed by the argument that rejection of the tender-ratification theory will discourage employers from offering severance packages and will encourage plaintiffs and their counsel knowingly to sign releases that do not comply with the OWBPA in order to receive benefits to which they are not entitled even though they have no intention of honoring the release agreement. The ratification-tender theory has been applied only by Courts of Appeals in the Fourth and Fifth circuits. There is no empirical evidence to show that in other areas of the country severance plans are offered less often or that there is a greater volume of litigation turning on allegedly inadequate ADEA releases.
A final factor favoring Hogue's rejection of a tender back rule in the ADEA context is practicality. "[I]mposing a `tender' requirement for challenges to ADEA releases would frequently create insoluble practical problems." Issacs, 765 F. Supp. 1367.
A tender requirement in such cases would . . . create a conundrum as to how much [consideration] should be tendered to restore the pre-release status quo. There is no available method of forcing the parties to agree on what an appropriate amount would be, since typically the employer does not specify how much of the consideration paid to the employee is for the retirement and how much is for the release.
Another practical problem identified by the district court in Issacs is the fact that "ordinary contract principles" governing tender back are not uniform:
Some states impose no tender requirement for law suits that challenge releases. Some state impose [a] tender requirement for certain kinds of challenges to releases, but not for others, becoming enmeshed in the technicalities of "void" versus "voidable" contracts. Some states impose universal tender requirements. Some states cannot make up their minds from decision to decision.
765 F. Supp. at 1372. The court in Issacs concluded that, "[t]he confusion . . . on this issue is reason enough for this Court to decide this [question] not on the basis of state-law contract doctrines, but — as Hogue commands . . . on the simple basis of what rule best serves the purposes of the ADEA." Id.
To require tender of the full amount of a severance payment would force an employee to return a sum that typically incorporates consideration for multiple factors not challenged in an age case: waivers for other violations of law or contract, rolled-in vacation and sick time, and a public relations benefit to the employer that itself may deter other litigation. This approach "would appear to leave the employer better off and the employee worse off than they were under the status quo." Id. at 1370. We are convinced that "[i]t would not serve the purposes of the ADEA to impose a tender requirement that creates such disputes and inequities." Id. at 1368.
Our decision in DiBiase v. Smith Kline Beecham Corp., 48 F.3d 719 (3d Cir.), cert. denied, 116 S.Ct. 306 (1995), does not resolve this difficulty as that case did not address tender back or the effect of OWBPA section 26(f)(1)(D). We do not read DiBiase to suggest that all severance pay received must be allocated to ADEA claims for purposes of tender back.
Having examined the language of the OWBPA, the purposes underlying its enactment, and the caselaw bearing on its application, we hold that where a release of ADEA claims fails to comply with the provisions of the OWBPA, the common law doctrines of ratification and tender back should not be applied to bar an employee's ability to pursue claims under the ADEA. In light of this holding, the district court's entry of summary judgment in favor of Sears on claims brought pursuant to the ADEA was erroneous. Accordingly, we will reverse that portion of the order of the district court granting summary judgment in favor of Sears with respect to the ADEA claims.
Contrary to the position taken in the dissent, we do not read our decision in McNemar v. The Disney Store, Inc., 91 F.3d 610 (3d Cir. 1996), petition for cert. filed Dec. 16, 1996, as having any bearing on this case. In McNemar we held that a plaintiff was precluded from pursuing a claim under the ADA based on the fact that he had made assertions inconsistent with his right to recover in proceedings before the Social Security Administration and two state agencies. The dissent argues that because "our ultimate conclusion . . . was that conduct which the ADA never addressed barred McNemar's action . . . . Similarly, the OWBPA does not set forth the controlling law in this case as it does not address the ratification issue." (Typescript at 11). The point seems to be that a plaintiff may be barred from pursuing a claim by factors lying outside the statute pursuant to which the claim is brought. We agree with McNemar's basic premise. Nevertheless, McNemar is distinguishable from this case both legally and factually. McNemar has nothing to do with what Congress intended in enacting the OWBPA nor with the analysis of those factors outside the statute which might bear on how the OWBPA should be applied in a particular circumstance. We have discussed these policy underpinnings at length and conclude that the issues here have little in common with the invocation of judicial estoppel in McNemar.
One final matter remains. Our rejection of the void/voidable distinction in reaching this result has implications for the non-ADEA claims asserted by Long. This case has, from its inception, centered on the question of whether the release, as a whole, was void or voidable. The parties have consistently framed and briefed the issues in terms of this distinction and, indeed, the district court rested its grant of summary judgment as to all claims on its finding that the release as a whole was voidable and had been ratified. Because Long approached this case by arguing that the entire release — including non-ADEA claims — was void we believe that these non-ADEA claims were adequately preserved for consideration on appeal.
Our holding, confined as it is to ADEA releases invalid under OWBPA, does not automatically dispose of the remainder of Long's claims as might be the case if we had rested our decision on the void/voidable distinction. Therefore, in order to ensure that the parties have an opportunity to analyze the remaining claims in terms of our holding and to present that analysis to the district court, we will vacate the district court's entry of summary judgment as to the non-ADEA claims and remand for further consideration.
The majority succinctly sets forth its primary conclusion at the outset of the opinion: "the ratification doctrine should not apply to a waiver of age discrimination claims which is invalid under the OWBPA and [consequently] Long should not be required to tender back severance benefits before proceeding with his age discrimination claims." Typescript at 3. The majority reaches this conclusion even though Long executed a broad form of release and waiver of his claims for which Sears paid Long over $39,000 which he retains. Consequently, the majority reverses the district court's summary judgment in favor of Sears on Long's ADEA claim. While the majority understandably seeks to protect the rights of older workers in accordance with the Older Workers Benefit Protection Act of 1990, 29 U.S.C. §(s) 626(f), and surely Congress did intend to protect older workers in that act, well-established principles of law lead me to a different conclusion. Thus, I dissent with respect to Long's ADEA claim. The majority vacates the summary judgment on Long's non-ADEA claims, but I dissent on this aspect of the case as well, as Long has not preserved his appeal with respect to these claims.
The majority sets forth the first legal question to be decided as follows: can an employee render a release enforceable which fails to conform to the requirements of the OWBPA with respect to ADEA claims by ratifying the agreement by acceptance and retention of severance benefits? And next: if ratification does not apply, does the employee's retention of severance benefits operate nonetheless to prevent an employee from pursuing a claim under the ADEA? Typescript at 9. After further discussion, the majority compares Oberg v. Allied Van Lines, Inc., 11 F.3d 679 (7th Cir. 1993), cert. denied, 114 S.Ct. 2104 (1994), with Wamsley v. Champlin Ref. and Chems., Inc., 11 F.3d 534 (5th Cir. 1993), cert. denied, 115 S.Ct. 1403 (1995), and Blistein v. St. John's College, 74 F.3d 1459 (4th Cir. 1996). In Oberg, the Court of Appeals for the Seventh Circuit held that releases not conforming with the OWBPA are void and cannot be ratified or enforced. On the other hand, the Courts of Appeals for the Fifth and Fourth Circuits in Wamsley and Blistein held that defective releases were voidable and that employees could ratify them. Thus, while Oberg did not require an employee to tender back the consideration paid for a release before bringing an ADEA suit, Wamsley and Blistein reached the opposite result and held that by retaining the consideration the employees ratified the releases, thereby barring their ADEA actions.
I am aware that the OWBPA speaks of an "individual" waiving rights and thus does not use the terms "employee" or "release." As a matter of convenience I will use the terms "employee" and "release," as Long was an employee and the terms "waiver" and "release" have the same meaning in the context of this case.
The majority next rejects the methodology of both Oberg and Wamsley (and thus Blistein, as well), though not the result in Oberg as, unlike the Oberg and Wamsley courts, the majority holds that it need not decide whether a release not conforming with the OWBPA is void or voidable. Indeed, the majority also rejects the approach of the parties to this appeal; Long explains in his brief that the legal principle governing the enforceability of his release depends on the answer to the following question: "Is a release obtained in violation of OWBPA and by fraud void or merely voidable?" Br. at 18-19. Sears argues the case on the same basis, and the Equal Employment Opportunity Commission agrees that we must decide whether a release not conforming with the OWBPA is void or voidable.
Instead, the majority finds that it need not determine whether a release which does not conform to the OWBPA is void or voidable because, without regard to the answer to that question, "neither ratification nor tender back was meant to apply in the ADEA context." Typescript at 20. It reaches that result because the "language of the OWBPA and its legislative history convince [it that] Congress did not intend that the ratification doctrine be invoked to enforce the terms of a deficient release." Id. After a discussion of the circumstances leading to the enactment of the OWBPA, the majority indicates that:
Congress intended to occupy the area of ADEA releases and, in doing so, to supplant the common law; the OWBPA was enacted to `establish a floor, not a ceiling.' Enforceability of a waiver is made contingent upon the presence of certain enumerated factors. Given the clear and specific goals of the OWBPA, we cannot accept that Congress intended that the common law doctrine of ratification be applied to releases invalid under the OWBPA. Id. at, 23-24 (citation omitted).
I respectfully state that we should not hold that the "doctrine of ratification" cannot be applicable to a release which is invalid under the OWBPA. Certainly nothing in the OWBPA states that an employee cannot ratify an invalid release. Rather, the act merely provides that an "individual may not waive any right or claim under the [ADEA] unless the waiver is knowing and voluntary." The act then provides minimum requirements for a waiver to be knowing and voluntary.
Furthermore, the holding that an employee cannot ratify an invalid release sometimes will lead to an outcome directly contrary to the policy of the OWBPA to protect older workers. In this case, of course, the employee, not the employer, is seeking to avoid the settlement agreement. But, as the majority seems to recognize, an employer also might seek to avoid its obligation to pay severance benefits. Typescript at 24-25 n. 18. It seems clear that inasmuch as the OWBPA was enacted to protect employees' rights, an employee should be able to ratify a defective release and hold a recalcitrant employer to its bargain.
I will demonstrate with a particular example why the policy of the OWBPA requires that an employee should be able to ratify a defective release. Under the OWBPA, a waiver of ADEA rights and claims in a settlement of an action in court must comply with certain of the minimum requirements of the OWBPA to be knowing and voluntary. 29 U.S.C. § 626(f)(2). One of these requirements is that the employee "is advised in writing to consult with an attorney prior to executing the agreement." 29 U.S.C. §(s) 626(f)(1)(E).
Consider the following situation. An employee represented by experienced and competent counsel brings an ADEA action. At trial, at the end of the presentation of evidence, the parties settle the case on the record in open court with the employee waiving his ADEA claims in return for the promise of a cash payment. Subsequently, however, the employer reneges on the settlement and refuses to make the payment, contending that the settlement agreement cannot be enforced because the employee was not "advised in writing to consult with an attorney prior to executing the agreement." In my view, in these circumstances it would be contrary to the intent of Congress in enacting the OWBPA if a court refused to enforce the agreement on the employee's motion, thus requiring the employee to try the case again. Indeed, it would be amazing if a court reached that result, as the settlement would have been enforceable if Congress had not adopted the OWBPA. Thus, unless the settlement could be enforced, the OWBPA would have the exact opposite effect to that which Congress intended. The OWBPA would prejudice the employee. Furthermore, a holding that the employee could not enforce the settlement would not protect any legitimate interest of the employer, as 29 U.S.C. §(s) 626(f)(1)(E) was enacted for the benefit of employees.
The majority dismisses my example by indicating that "[w]e certainly have not suggested, nor, as far as we know, is there any authority for the proposition that the OWBPA might apply to the terms of a settlement agreement forged after the taking of evidence in a civil trial." Typescript at 24, n. 18. I am at a total loss to understand how the majority can make this statement as the OWBPA provides that a "waiver in settlement of a charge filed with the Equal Employment Opportunity Commission, or an action filed in court by the individual or the individual's representative, alleging age discrimination . . . may not be considered knowing and voluntary unless at a minimum — (A) subparagraphs (A) through (E) of paragraph (i) have been met." 29 U.S.C. §(s) 626(f)(2) (emphasis added). I reiterate that subparagraph (E) provides "the individual is advised in writing to consult with an attorney prior to executing the agreement." While the majority cites legislative history indicating congressional concern with preemptive waiver of an employee's rights before a dispute has arisen, the OWBPA as written simply is not limited to such situations. The majority thus is confining the application of the OWBPA in a way Congress did not.
I have given a dramatic example demonstrating that the majority's holding that an employee cannot ratify a release "invalid under the OWBPA" in some cases will frustrate the policy of the OWBPA. Typescript at 23-24. Yet in other circumstances, as well, employers might want to repudiate a release even though delivered at an earlier stage of litigation or not given in settlement of an action in court. In my view, the employer should not be able to repudiate a release because of its own failure to comply with the OWBPA. Nevertheless, unless an employee can ratify a release which is invalid under the OWBPA, the employer will be able to do exactly that.
Actually, notwithstanding its holding that a defective release cannot be ratified, the majority will not foreclose the possibility that an employee may ratify a defective release. The majority does, after all, leave open the possibility that in some other context, i.e., when it is in the employee's interest to enforce a settlement, he or she may be able to do so for it indicates that "[t]hat case, should it ever arise, is a matter for a different day; we need not decide this case on the basis of what might happen in a hypothetical case which might come before us at some point in the future." Typescript at 25, n. 18. While I agree that we cannot decide cases not before us, yet, if the majority is correct in holding that "neither ratification nor tender back was meant to apply in the ADEA context" then the outcome of the "hypothetical case" is preordained. Thus, the majority is unwilling to accept the consequences of its own holding. In my view, we should consider the consequences of a holding with respect to future cases. After all, I thought that we should avoid construing a statute to reach an absurd result. See, e.g., Government of the Virgin Islands v. Berry, 604 F.2d 221, 225 (3d cir. 1979).
Accordingly, the issue in this case should not be whether an employee can ratify a release not complying with the OWBPA, but what conduct of the employee constitutes a ratification of an invalid release? In particular, does the employee's withholding of the consideration the employer paid for a release ratify the release? Inasmuch as the OWBPA does not indicate what conduct constitutes ratification, we must refer to the common law for guidance on the point. After all, where else can we look? I realize that the majority indicates that the OWBPA "supplant[s] the common law." Typescript at 23. But I cannot understand how that can be true. The OWBPA provides that an individual may not waive any right or claim under the ADEA except by a knowing and voluntary waiver and sets forth "minimum" requirements for a waiver to be knowing and voluntary. 29 U.S.C. §(s) 626(f)(1). Conspicuously absent from the list are the basic prerequisites to an agreement being knowing and voluntary, i.e., that the employee waiving the rights have at least a certain level of mental competency and that the employee not sign the waiver as a consequence of unlawful threats. Nevertheless, while the OWBPA does not say so, there can be no doubt that if the employer threatens the employee with bodily harm if the employee does not sign the waiver, the waiver is not "knowing and voluntary" and thus is not enforceable.
What then is the source of the requirements beyond those enumerated in 29 U.S.C. §(s) 626(f)(1) for a waiver to be knowing and voluntary? I reiterate that there can be only one source, the common law. Accordingly, I do not doubt that in determining whether a waiver is knowing and voluntary, a court should consider common law principles with respect to such traditional factors relating to the validity of contracts as competency and duress which are absent from the OWBPA.
Further, it is evident that the OWBPA could not have supplanted the common law with respect to ADEA releases for still another reason: the OWBPA does not even purport to occupy the entire area regarding enforceability of ADEA releases. For example, under 29 U.S.C. §(s) 626(f)(1)(D), a knowing and voluntary waiver must be "in exchange for consideration in addition to anything of value to which the individual already is entitled." The OWBPA, however, does not address the possibility of the failure of consideration, i.e., the employer does not fulfill its obligations under the agreement. Is the employee bound by the release if the employer does not pay the consideration it promised for the release? As far as I am concerned, it is clear that the OWBPA did not "supplant" the common law with respect to enforceability of ADEA releases. Instead, it has supplemented the common law and, in this case, as in other cases dealing with a federal statute, we must develop a federal common law to accompany the statute. See, e.g., Ryan v. Federal Express Corp., 78 F.3d 123, 126-27 (3d Cir. 1996); United States v. Alcan Aluminum Corp., 964 F.2d 252, 268 (3d Cir. 1992). Thus, in urging that we apply federal common law in this case, I am suggesting nothing unusual. Rather, I am proposing that we use our ordinary methodology in applying a statute which does not address a problem which arises under it.
Where does application of the common law lead us? The answer is obvious. The Court of Appeals for the Seventh Circuit in Fleming v. United States Postal Serv., 27 F.3d 259, 260-61 (7th Cir. 1994), cert. denied, 115 S.Ct. 741 (1995), pointed out that it is "one of the most elementary principles of contract law . . . that a party may not rescind a contract without returning to the other party any consideration received under it . . . . The principle that a release can be rescinded only upon a tender of any consideration received . . . would surely be a component of any federal common law of releases." Thus, in Fleming the court held that an employee could not avoid a release settling Title VII and Rehabilitation Act claims without tendering back the consideration for the release. We should apply that settled principle in this case and affirm the order for summary judgment.
Long argues that the release is invalid as purporting to "waive rights or claims that may arise after the date the waiver is executed." 29 U.S.C. §(s) 626(f)(1)(C). The majority does not need to reach this point and does not do so. To affirm we would have to reach this issue and I would do so and reject it. However, in view of my dissenting position in this case, I have no need to explain my reasoning on this point.
Our treatment of an analogous issue in our recent opinion in McNemar v. The Disney Store, Inc., 91 F.3d 610 (3d Cir. 1996), petition for cert. filed Dec. 16, 1996, surely points to the result we should reach here. In that case, McNemar, who was HIV-positive, brought an action under the Americans with Disabilities Act saying that Disney violated the ADA when it discharged him. Of course, to recover McNemar had to show that he was discharged because of his disability and that with or without reasonable accommodations he could perform the essential functions of the job.
In McNemar, we held that McNemar could not recover because he had asserted to the Social Security Administration and to two state agencies that he was totally disabled and unable to work. Yet the ADA does not provide that a plaintiff, by making such assertions, forfeits his or her right to recover under the ADA. Nevertheless, through an application of judicial estoppel we would not allow McNemar to pursue his ADA claims. Thus, even though McNemar might have been able to demonstrate that he could establish a cause of action under the ADA, we held that he had deprived himself of that opportunity. Our ultimate conclusion, therefore, was that conduct which the ADA never addressed barred McNemar's action. While we could have said that we would look solely to the ADA for the governing law with respect to McNemar's right of recovery, as McNemar urged, and thus rejected Disney's judicial estoppel arguments, we did not do so. Similarly, the OWBPA simply does not set forth the controlling law in this case as it does not address the ratification issue. In fact, Long's position is weaker than McNemar's because the ADA far more comprehensively regulates actions under it than the OWBPA governs releases under the ADEA.
While I reach my conclusion as the result of independent analysis, I point out that my conclusion accords with the weight of appellate authority and not merely because Oberg is one case and Wamsley and Blistein are two. Rather, it is also for the reason that the Court of Appeals for the Seventh Circuit, which decided Oberg, later questioned Oberg by diplomatically describing its reasoning in that case as "a little obscure." Fleming, 27 F.3d at 261. Thus, the court which supplies the only appellate support under the OWBPA for the majority's approach has cast doubt on its own opinion. The Blistein court made this point when it indicated that the Court of Appeals for the Seventh Circuit itself, in Fleming, "questioned" Oberg. Blistein, 74 F.3d at 1466.
The Court of Appeals for the Fifth Circuit adhered to Wamsley in Blakeney v. Lomas Information Sys., Inc., 65 F.3d 482, 484-85 (5th Cir. 1995), cert. denied, 116 S.Ct. 1042 (1996). In the recent case of Raczak v. Ameritech Corp., F.3d 1997 WL 5921 (6th Cir. Jan. 9, 1997), the panel was too fractured in its approach to render a decision of much precential value on the ratification issue.
Hogue v. Southern R. Co., 390 U.S. 516, 88 S.Ct. 1150 (1968), a case involving a FELA settlement on which the majority partially relies, is not controlling. Hogue was based on a different statutory scheme and, as a footnote in Hogue makes clear, the Supreme Court in part based its opinion on a statute making agreements to exempt the employer from liability "void." Id. at 518, 88 S.Ct. at 1152. In this case the majority does not hold that a release not in conformity with the OWBPA is void. Indeed, it does not reach that issue. I, of course, would hold that the release is not void. Furthermore, the OWBPA, in its terms, simply does not provide that a release not in conformity with the OWBPA is "void." Thus, the OWBPA differs critically from the FELA statute at issue in Hogue. Of course, as Oberg, Wamsley, and Blistein make clear, void contracts cannot be ratified. Thus, Hogue is distinguishable. Indeed, if the OWBPA provided that releases not conforming with its terms are void, I would not be dissenting with respect to the ADEA aspects of the case. I, however, will not extend my discussion of Hogue, for Wamsley discusses Hogue at length and demonstrates that it is not controlling in the OWBPA context and I adopt Wamsley's analysis. Wamsley, 11 F.3d at 540-42.
The legislative history of the OWBPA is of some help in this case but it does not support the majority's result. The majority points out that "Congress was aware of the benefit retention issue and chose not to include in the OWBPA a proviso prohibiting this retention." Typescript at 25 n. 19. Yet Congress also chose not to include a proviso authorizing an individual to challenge a release while retaining the consideration for the release. I would think that if Congress was aware that parties might seek to apply a common law doctrine under a statute and it did not intend that they could do so, it would have addressed the point. We indicated in United States v. Alcan Aluminum Corp., 964 F.2d at 268, in determining the allocation of damages under the Comprehensive Environmental Response, Compensation and Liability Act, that "Congress' deletion of joint and several liability from the final version of the statute signaled its intent to have the courts determine, in accordance with traditional common law principles, whether such liability is proper under the circumstances." A similar type of analysis here demonstrates that, if anything, the legislative history supports my conclusion because, as in Alcan Aluminum, Congress was aware of a problem which could arise under a statute it was enacting and left the issue to the courts to resolve. As in Alcan Aluminum, how could the courts decide the issue except by applying the common law? I will mention briefly some other OWBPA issues. First, I recognize that the employee might be unable to return the consideration paid by the employer for the defective OWBPA release. After all, the employee could have spent the money. That circumstance, however, would not be legally germane because it is not unique to cases in which a party seeks to rescind a release of ADEA claims. Yet, as Fleming explains, to rescind a party must return the consideration the party obtained under the contract. Second, allowing ratification will not encourage employers to obtain invalid releases, for the employer obtaining such a release would run the risk that the employee would tender back the consideration and then sue under the ADEA while seeking to avoid the release. On the other hand, if the employee can retain the consideration and sue on the merits if the release is invalid, the employee will obtain an undeserved windfall.
Third, I note but reject Long's contention that allowing ratification of a defective release may result in inconsistent adjudications from state to state because the law regarding ratification is not uniform in all states. Clearly, a federal common law should be consistent throughout the country, though I acknowledge that in cases involving ratification courts of appeals may reach different conclusions until the Supreme Court settles the law. But the possibility of inconsistent adjudications among the circuits exists throughout the law, so there is nothing unique about that possibility in the context of ratification of ADEA releases. In fact, with this opinion there will be a two/two conflict among the circuits on whether an employee ratifies a defective OWBPA release by retaining the consideration the employer paid for it.
As I indicated at the outset, this appeal involves more than an ADEA claim because the majority vacates the summary judgment awarded to Sears on Long's ERISA, Pennsylvania Human Relation Act, and common law claims. I see no basis for it to do so. There is no argument in Long's brief supporting the contention that the summary judgment in favor of Sears on those claims should be reversed. Rather, Long directs his arguments solely to his ADEA claim. If anyone doubts me on this point, I suggest that the doubter read Long's brief. In fact, with one possible exception, Long never specifically mentions his non-ADEA claims after he describes them in the statement of the case on the second page of his 48-page brief. This possible exception is that Long argues that he had not ratified the release by "undue delay" in declaring his position repudiating the release, because he filed a PHRA charge of discrimination in July 1993, only four months after signing the release. Br. at 43-44.
Long, however, does not raise the PHRA issue in connection with an argument that the summary judgment on the PHRA claim should be reversed. Rather, he makes the point in the context of an argument that even if common law ratification principles are applicable to ADEA releases, he has not ratified the release. Br. at 39. Thus, Sears's brief is correct when it points out that Long has not "asserted that the District Court erred in dismissing Long's non-ADEA claims to which the OWBPA does not apply." Br. at 4.
In response to Sears's observation, Long argues in his reply brief why the summary judgment on the non-ADEA claims should be reversed, contending that the release is invalid under the OWBPA and is not severable with respect to Long's claims so that it is "unenforceable in all respects." Reply br. at 10. That frivolous argument, however, comes too late. We explained the procedural rule in the Republic of the Philippines v. Westinghouse Elec. Corp., 43 F.3d 65, 71 n.5 (3d Cir. 1994), as follows:
Although the Republic stated in its initial briefing that it `strongly disagrees' with the district court's findings `that Philippine government officials engaged in retaliation against or harassment of witnesses in this case' (Appellant's Br. at 19), it did not squarely challenge those findings as clearly erroneous. See Burns and Roe Br. at 14 (noting Republic's failure to challenge findings). The Republic did raise the issue of clear error in its reply brief (Appellant's Reply Br. at 10-16), but that was one brief too late: we have often instructed that `appellants are required to set forth the issues raised on appeal and to present an argument in support of those issues in their opening brief.' Kost v. Kozakiewicz, 1 F.3d 176, 182 (3d Cir. 1993).
Thus, Long has waived his action insofar as he bases it on ERISA, the PHRA, and the common law, as he has not properly challenged the district court's ruling that he has ratified the release on those claims. Circuit procedural precedent requires that we reject Long's challenge to the summary judgment on the non-ADEA claims.
The majority nevertheless finds that "[b]ecause Long approached this case by arguing that the entire release — including non-ADEA claims — was void we believe that these non-ADEA claims were adequately preserved for consideration of appeal." Typescript at 37-38. While the majority correctly recognizes that it is deciding the case on a basis Long does not advance, the fact remains that Long's argument was that the release was void under the OWBPA. He makes no argument in his opening brief explaining why the release could not be enforced with respect to his non-ADEA claims nor does he advance on any basis in that brief to reverse the summary judgment on those claims. Of course, it does not follow automatically that if a release cannot be enforced with respect to ADEA claims, it cannot be enforced with respect to non-ADEA claims. Thus, Long had to have made that contention to preserve it for appeal. Yet he simply did not make that contention until his reply brief when it was too late. I reiterate my suggestion that anyone who doubts what I say should read Long's brief. I am certain that the reader will agree that Long presents no argument on any basis for reversing the judgment on the non-ADEA claims.
While I would not consider the challenge to the release on the non-ADEA claims on the merits, I note that even if I agreed with the majority's holding on the ADEA claim, I could conceive of no way in which the release would be invalid with respect to the non-ADEA claims. Is a court to read the OWBPA requirements into settlement of ERISA, PHRA, and common law claims? The implications that somehow the release in this case might not be effective with respect to the non-ADEA claims are so far-reaching that I hesitate even to state them. Of course, these implications will not be lost on attorneys who represent employees who have signed releases of non-ADEA claims in connection with employment terminations. What I do state is the obvious conclusion that if Long's release is valid as to the non-ADEA claims, it would not have to be ratified with respect to them to be enforceable.
In conclusion, I will sum up my views. I believe that we should follow the weight of appellate authority, our well-established practice of applying federal common law to statutes not addressing issues arising under them, and our recent opinion in McNemar dealing with a situation analogous to that here. Thus, I would hold that an employee may ratify a release which is invalid under the OWBPA, and that Long has ratified the release. I would not entertain the appeal from the summary judgment on the non-ADEA claims. Consequently, I would affirm the order for summary judgment entered by the district court in its entirety and I thus respectfully dissent.