Full opinion text
MEMORANDUM OF DECISION SQUATRITO, District Judge. This is an action for damages in which nineteen Plaintiffs allege that Defendant Northeast Utilities (“Defendant” or “NU”) breached its fiduciary duty to them. Plaintiffs’ claims are brought pursuant to the Employee Retirement Income Security Act (“ERISA”), 29 U.S.C. §§ 1001, et seq. The Court held a bench trial on these claims, and now issues the following Memorandum of Decision, including Findings of Fact and Conclusions of Law: I. EVIDENTIARY ISSUES There were several evidentiary issues presented to the Court that were deferred to this Memorandum of Decision. The Court will address those issues at the outset, before presenting its Findings of Fact and Conclusions of Law. The first issue presented to the Court was the admissibility of the affidavits of direct testimony offered by Plaintiff Lawrence LeBrun (“LeBrun”) and Plaintiff Viola Sorensen (“Sorensen”). LeBrun was gravely ill at the time he was scheduled to testify at trial and unable to appear for cross-examination and re-direct examination concerning his affidavit of direct testimony. Before the trial proceedings reached their conclusion, LeBrun died and, as a result, LeBrun’s deposition was entered into evidence. Sorensen represented to the Court by affidavit that she was unable to testify because she was tending to the health needs of her brother. The Court must decide whether her deposition transcript will be allowed into evidence. Lastly, the Court will address the admissibility of the interrogatory responses of LeBrun and Plaintiff Arthur Reil (“Reil”), who died before the trial began. Reil was never deposed. The Court will first consider the hearsay implications of each of these statements, and then further discuss Sorensen’s deposition transcript. A. Hearsay implications of Le-Brun’s, Reil’s and Sorensen’s statements NU has moved to exclude LeBrun’s and Sorensen’s affidavits of direct testimony, LeBrun’s and Reil’s interrogatory responses, and Sorensen’s deposition transcript. The basis of each objection is that these statements are inadmissible hearsay and are not otherwise admissible under any of the hearsay exceptions provided in the Federal Rules of Evidence. Under Federal Rule of Evidence 801(c), hearsay is defined as “a statement, other than one made by the declarant while testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted.” Fed.R.Evid. 801(c). A statement that fits this criterion is inadmissible, unless it falls under any of the hearsay exceptions set forth in the Federal Rules of Evidence. In this case, the Court first finds that the relevant statements are inadmissible hearsay, and second, that they are not otherwise admissible under any of the hearsay exceptions. First, the statements at issue in this case meet the definition of hearsay set forth in the Federal Rules of Evidence. Each statement was made out of court, as the declarant was not testifying at the trial or at a hearing. Specifically, the affidavits of direct testimony were prepared by Le-Brun, Reil and Sorensen, with the assistance of Plaintiffs’ counsel, before the trial began. Reil’s and LeBrun’s interrogatories were prepared in the same way, years before the trial began. Sorensen’s deposition was taken outside of a trial setting, again before the trial began. LeBrun, Reil and Sorensen offer these affidavits of direct testimony to prove the truth of the matters asserted. Plaintiffs contend in their briefing on this issue that this is not the case, and that instead they offer these affidavits to show the state of mind of the declarants. The Court cannot accept this proposition. Like any testimony, the most critical aspects of these affidavits are being offered to prove the truth of the matters asserted therein—what was said to Plaintiffs either at the time they considered retirement or at the time of their actual retirement. Moreover, the perceptions expressed by Plaintiffs in these affidavits are being offered for their truth. Statements of the type contained in these affidavits invoke the concerns meant to be addressed by the hearsay rule, and meet the definition of hearsay contained in the Federal Rules of Evidence. Second, the Court finds that the affidavits do not fall under any of the exceptions to the hearsay rule listed in the Federal Rules of Evidence, particularly Rules 803(3) and 807. The residual exception to the hearsay rule, Rule 807, does not apply. The residual hearsay rule provides: A statement not specifically covered by Rule 803 or 804 but having equivalent circumstantial guarantees of trustworthiness, is not excluded by the hearsay rule, if the court determines that (A) the statement is offered as evidence of a material fact; (B) the statement is more probative on the point for which it is offered than any other evidence which the proponent can procure through reasonable efforts; and (C) the general purposes of these rules and the interests of justice will best be served by admission of the statement into evidence. However, a statement may not be admitted under this exception unless the proponent of it makes known to the adverse party sufficiently in advance of the trial or hearing to provide the adverse party with a fair opportunity to prepare to meet it, the proponent’s intention to offer the statement and the particulars of it, including the name and address of the declarant. Fed.R.Evid. 807. The residual exception to the hearsay rule is used “very rarely, and only in exceptional circumstances.” Parsons v. Honeywell, Inc., 929 F.2d 901, 907 (2d Cir.1991). If a court is to create one of these exceptional circumstances, it must first evaluate whether the statements at issue satisfy the five requirements listed in Rule 807—trustworthiness, materiality, probative importance, the interests of justice and notice. In addition, the statements must be evaluated to assess whether the four classic hearsay dangers, which are insincerity, faulty perception, faulty memory and faulty narration, are minimized. Schering Corp. v. Pfizer Inc., 189 F.3d 218, 233 (2d Cir.1999). “Hearsay statements need not be free from all four categories of risk to be admitted under Rule 807.” Id. In this case, the Court believes that the affidavits of LeBrun, Reil, and Soren-sen do not meet this criteria. First, the Court cannot find that the statements have circumstantial guarantees of trustworthiness. The fact that these statements were prepared by, or in conjunction with Plaintiffs’ counsel is just one reason why they cannot be considered trustworthy. While the Court does not mean to imply that it believes Plaintiffs’ counsel or LeBrun, Reil and Sorensen intentionally included misstatements or untruths in their affidavits or interrogatory responses, the fact that they were prepared in the context of litigation raises concerns. The most truthful of discovery or trial declarations still cannot escape the reality that they are prepared with the “incentive to set forth the facts in a light most favorable to itself.” Kirk v. Raymark Industr., Inc., 61 F.3d 147 (3d Cir.1995). This raises strong concerns for the Court, especially because NU was prejudiced by its inability to challenge these self-interested statements at trial. In addition, the Court does not believe that any circumstances diminish its concern in this case. The fact that the affidavits in question are similar to those made by the other Plaintiffs does not make them more trustworthy. While some portions of the statements contained in the affidavits may be similar, they also contain facts specific to each Plaintiff that cannot be corroborated by or analogized with other Plaintiffs. In addition, the Court cannot help but note that several of the Plaintiffs adjusted or clarified their affidavits while testifying at trial. Even though the adjustments made were sometimes minor in the Court’s estimation, the fact that any changes at all were made leads the Court to conclude that these Plaintiffs may also have altered their testimony if subject to examination at trial. Finally, while Gerard Turner (“Turner”), a witness for the Defendant, did testify that he “typically” told Plaintiffs things that were similar to the things included in the affidavits in question, this testimony is not specific enough to corroborate the hearsay statements sought to be admitted. While the other criteria included in the residual hearsay rule — -materiality, probative importance, interests of justice and notice — could be met in this case, the inherent lack of trustworthiness of the statements is most troubling to the Court. With this being the case, the Court cannot admit the statements under the residual hearsay exception. In addition, the statements are not admissible under the present state of mind exception under Federal Rule of Evidence 803(3). This rule provides: The following are not excluded by the hearsay rule, even though the declarant is available as a witness ... A statement of the declarant’s then existing state of mind, emotion, sensation, or physical condition (such as intent, plan, motive, design, mental feeling, pain, and bodily health), but not including a statement of memory or belief to prove the fact remembered or believed unless it relates to the execution, revocation, identification, or terms of declarant’s will. Fed.R.Evid. 803(3). The Court does not find that this rule is applicable to the statements at issue. First, this rule specifically excludes any “statements of memory or belief to prove the fact remembered or believed unless it relates to [a will].” Fed.R.Evid. 803(3). In this case, however, the statements offered by LeBrun, Reil and Sorensen almost exclusively relate to their recollections of the past, and were not made contemporaneously. If the Court were to admit them as evidence, it would be in contravention with Second Circuit cases which have held that in order for a statement to qualify for the benefits of Rule 803(3), it must “face forward, rather than backward.” See, e.g., U.S. v. Harwood, 998 F.2d 91, 98 (2d Cir.1993). In addition, the Court does not find the parallel to the court’s ruling in Leyda v. Allied Signal, Inc. that is suggested by Plaintiffs. No. 3:99cv76, slip. op. (D.Conn.) (JCH). In that case, witnesses were allowed to testify about past statements made to them by their family members, who were plaintiffs in the case. At the time the statements were made, they expressed contemporaneous perceptions that were pertinent to the issues raised in the case. The statements proposed to be offered here, however, are quite different, as they do not have the same contemporaneous aspect to them. Finally, the statements do not concern a will, but rather retirement benefits. Thus, the statements do not fall under the present state of mind exception to the hearsay rule. In sum, the Court finds that the affidavits of direct testimony and interrogatories from LeBrun, Reil and Sorensen are inadmissible hearsay, not covered by any recognized exception to the hearsay rule. B. The admissibility of Sorensen’s deposition testimony The Court also finds the deposition testimony of Sorensen inadmissible. Under the Federal Rules of Evidence, deposition testimony may be offered in lieu of live testimony if the declarant is unavailable. Fed.R.Evid. 804. Federal Rule of Evidence 804 provides that a declarant is unavailable under five circumstances: “Unavailability as a witness” includes situations in which the declarant — (1) is exempted by ruling of the court on the ground of privilege from testifying concerning the subject matter of the declarant’s statement; or (2) persists in refusing to testify concerning the subject matter of the declarant’s statement despite an order of the court to do so; or (3) testifies to a lack of memory of the subject matter of the declarant’s statement; or (4) is unable to be present or to testify at the hearing because of death or then existing physical or mental illness or infirmity; or (5) is absent from the hearing and the proponent of a statement has been unable to procure the declarant’s attendance (or in the case of a hearsay exception under subdivision (b)(2), (3), or (4), the declarant’s attendance or testimony) by process or other reasonable means. A declarant is not unavailable as a witness if exemption, refusal, claim of lack of memory, inability, or absence is due to the procurement or wrongdoing of the proponent of a statement for the purpose of preventing the witness from attending or testifying. Fed.R.Evid. 804(a). None of these circumstances are present in this case. Sorensen was not prevented from testifying based on privilege, did not defy a Court order to testify, and did not claim a lack of memory. Furthermore, while Sorensen did submit that her brother was suffering from an illness, she herself was not suffering from a physical or mental illness or infirmity. Finally, Soren-sen cannot meet the last criteria, that the proponent of the statement has been unable to compel a witness to testify by process or other means, because she herself has decided not to come to Court to testify. The Court is sympathetic to Sor-ensen’s family situation, and does not claim that she had any sinister motive in failing to appear. However, the Federal Rules of Evidence to which the Court is bound clearly and unambiguously preclude the admission of her deposition testimony because she does not meet the unavailability criteria therein. For the foregoing reasons, the Court finds that the affidavits of direct testimony submitted by Plaintiffs Reil, LeBrun and Sorensen, the interrogatory responses of Reil and LeBrun, and the deposition testimony of Sorensen are inadmissible under the Federal Rules of Evidence. II. STATUTE OF LIMITATIONS Another preliminary issue presented to the Court that was similarly deferred to this Memorandum of Decision regards the statute of limitations. The Court will rule on this issue before offering its Findings of Fact and Conclusions of Law. The Court must always address the timeliness of any plaintiffs claim for breach of fiduciary duty under § 413(2) of ERISA. Section 413(2) describes the applicable statute of limitations, as follows: No action may be commenced under this subchapter with respect to a fiduciary’s breach of any responsibility, duty, or obligation under this part, or with respect to a violation of this part, after the earlier of- (2) three years after the earliest date on which the Plaintiff had actual knowledge of the breach or violation; except that in the case of fraud or concealment, such action may be commenced not later than six years after the date of discovery of such breach or violation. 29 U.S.C. § 1113. Based on this language, an ERISA plaintiff is only entitled to bring a claim for breach of fiduciary duty within three years of the earliest date he. had “actual knowledge” of the breach or violation, or, as an exception to this rule, where a plaintiff shows “fraud or concealment,” he must bring a claim for fiduciary duty within six years of the date he discovered a breach or violation. The Court has addressed this issue once before in this case. In an earlier ruling, the Court dismissed Plaintiffs’ claims as time-barred, based on its conclusion that the Plaintiffs had actual knowledge of the alleged breach or violation more than three years before they filed their claims. This conclusion was based on the Court’s assessment that the Plaintiffs had “actual knowledge” of the breach or violation at the time the early retirement programs were actually announced to NU employees, contrary to what they claimed they had been told by NU management. Although the Court dismissed Plaintiffs’ claims under the three-year statute of limitations on that basis, it granted leave to Plaintiffs to amend their Complaint in order to comply with the six-year statute of limitations. In other words, the Court allowed the Plaintiffs to amend their Complaint to allege fraud or concealment with the requisite particularity. Soon after the Plaintiffs did so, the case proceeded to trial. While the case was moving toward trial, the Second Circuit issued a decision in Caputo v. Pfizer, 267 F.3d 181 (2d Cir. 2001). One of the many issues discussed in Caputo was ERISA’s statute of limitations for fiduciary duty claims, particularly the meaning of the term “actual knowledge.” Among other things, the Appellate Court called into question an analysis provided by many courts — including this Court in this earlier decision — that the release of an early retirement program, in and of itself, would provide a plaintiff with enough information to constitute “actual knowledge” of a breach or violation of fiduciary duty. In light of Caputo, the parties requested that this Court vacate its earlier ruling on the statute of limitations issue. The Court agreed that this would be the most prudent course, and indicated that it would revisit the statute of limitations issue in this opinion, applying the Second Circuit’s decision in Caputo. The Caputo court provided guidance to lower courts in this Circuit on the issue of ERISA’s statute of limitations for fiduciary duty claims. The Appellate Court was clear from the outset that the statute of limitations it was working with in Caputo — the same statute of limitations that is at issue in this case — is “enigmatic — almost chimerical,” and is “[h]eld together by chewing gum and baling wire.” Id. at 188. Caputo added one more wad of gum — or piece of wire — to this statute of limitations. The Court addressed two issues in its ruling; first, the meaning of “actual knowledge” in the three-year statute of limitations; and second, the definition of “fraud or concealment” in the six-year statute of limitations. A. Three-Year Limitations Period On the first issue, the Second Circuit for the first time defined actual knowledge as the term related to ERISA’s statute of limitations. The case arrived at the circuit court after the district court held that plaintiffs’ fiduciary duty claims were time-barred because they had actual knowledge more than three years before they filed suit. The district court based this conclusion on its assessment that plaintiffs had actual knowledge of a breach or violation when the defendant announced the early retirement program, contrary to what they believed they had been told about the defendant’s future plans. The circuit court disagreed with the district court’s assessment of when the plaintiffs held this actual knowledge, and reversed. The Second Circuit defined actual knowledge as follows: [W]e now hold that a plaintiff has “actual knowledge of the breach or violation” ... when he has knowledge of all material facts necessary to understand that an ERISA fiduciary has breached his or her duty or otherwise violated the Act. While a plaintiff need not have knowledge of the relevant law, he must have knowledge of all facts necessary to constitute a claim. Such material facts “could include necessary opinions of experts, knowledge of a transaction’s harmful consequences, or even actual harm.” Id. at 193 (internal citations omitted). Notably, the court specifically rejected the district court’s analysis that the release of an early retirement program alone could provide enough information to a plaintiff to constitute actual knowledge: “[t]he disclosure of a transaction that is not inherently a statutory breach of fiduciary duty ... cannot communicate the existence of an underlying breach.” Id. The Court had several reasons for this conclusion. First, the Court concluded that the early retirement program offered by the defendant was not “inherently suspect, nor did it constitute a breach of fiduciary duty or an ERISA violation” at first glance, and was thus “insufficient to trigger the three-year statute of limitations.” Id. The Court said: Although the announcement should have (and did) give plaintiffs reason to suspect that [the defendant] had lied to them, “it is not enough that [plaintiffs] had notice that something was awry; [plaintiffs] must have had specific knowledge of the actual breach of duty upon which [they sued].” Id. (internal citations omitted) (emphasis in the original). Furthermore, the Court stated that to hold otherwise would impose a constructive knowledge provision to ERISA plaintiffs — a type of provision -that had been explicitly excluded from the statute by Congress. Id. at 194 (citing H.R. 391, 100th Cong., 1st Sess.1987, reprinted in, 1987 U.S.C.C.A.N. 2313-1). Instead, based on the law of fiduciary duty, the court held that “an ERISA plaintiff cannot be said to have ‘actual knowledge of the breach or violation’ until he has actual knowledge that his employer misrepresented a present fact or failed to disclose all material information known at the time of inquiry.” Id. As to the claims brought in that case, the court found that the plaintiffs did not have actual knowledge of a breach of fiduciary duty until they learned through trial testimony in a related ERISA case against the defendant that it knew about its plans to make a future offering at the time it was making contrary representations to its employees. Id. The Second Circuit, however, provided cautionary language in the midst of this conclusion. It noted that its holding was not intended to go so far as to mean that plaintiffs don’t have actual knowledge until they learn of information that would normally be revealed through discovery. Id. 195. This would, of course, defeat the purpose of a statute of limitations. This Court agrees with the Second Circuit that the statute of limitations should not be broadened to the point that employees can sue for breaches of fiduciary duty at any time. If the Court were to strictly construe the statute of limitations too harshly, however, it would create the opportunity for employers who breach their fiduciary duty to their employees through secrecy to commit not one, but two wrongs. Not only would they commit an ERISA violation, but courts would enable them to impair their employees’ ability to bring an ERISA suit based on the employers’ subterfuge. Based on the guidance provided by Caputo, the Court finds that Plaintiffs in this case acquired actual knowledge within the three-year statute of limitations. As said in Caputo, just because an early retirement program raised Plaintiffs’ suspicions at the time it was offered is not enough to constitute actual knowledge of a breach of fiduciary duty. Instead, they must have had some sort of specific knowledge that elevated their suspicion to actual knowledge. In this case, it was the Plaintiffs’ exposure to some other information which made the Plaintiffs believe that NU had committed a breach of fiduciary duty. Plaintiffs universally stated that they were made aware of their potential claim after either reading about similar claims in the newspaper or seeing an advertisement by Plaintiffs’ counsel within the three-year statute of limitations. The fact that a few of the Plaintiffs consulted lawyers is of no consequence. Each time a Plaintiff consulted an attorney NU contacted the Plaintiff shortly after and assured the Plaintiff that there were no legal improprieties, thus diminishing any belief the Plaintiffs might have had a legal claim existed. The Court finds that prior to the advertisements and newspaper articles, Plaintiffs did not have the actual knowledge to start § 413’s statute of limitations. The court vacates its ruling to the contrary, and finds that Plaintiffs may pursue their claims. B. Six-Year Limitations Period In the alternative, the Court finds that Plaintiffs may pursue their claims under the six-year statute of limitations because they have properly alleged fraud or concealment. ERISA’s statute of limitations provides that “in the case of fraud or concealment, [a breach of fiduciary duty action] may be commenced not later than six years after the date of discovery of such breach.” Whether plaintiffs have adequately shown fraud is “relevant to when ‘the action’ may be timely brought, ‘the action’ itself is still one for breach of fiduciary duty.” Diduck v. Kaszycki & Sons Contractors, Inc., 974 F.2d 270, 282 (1992). In Caputo, the Second Circuit defined “fraud or concealment” as follows: [T]he six year statute of limitations should be applied to cases in which a fiduciary either: (1) breached its duty by making a knowing misrepresentation or omission of a material fact to induce an employee/beneficiary to act to his detriment; or (2) engaged in acts to hinder the discovery of a breach of fiduciary duty. Caputo, 267 F.3d at 190. As to the first provision, the Court found that the elements of common law fraud are applicable. Id. at 188-93. These elements include: (1) a material false representation or omission of an existing fact, (2) made with knowledge of its falsity, (3) with an intent to defraud, and (4) reasonable reliance, (5) that damages plaintiff. Id. Specifically, the Court found that information is material “if there is substantial likelihood” that a misrepresentation of future pension benefits “would mislead a reasonable employee in making an adequately informed decision about if and when to retire.” Id. at 192 (citing Ballone, 109 F.3d at 123). As discussed later in this opinion, in this case NU intentionally withheld important information from its employees by virtue of its confidentiality policy. NU concealed information not only from its employees as a whole, but also from the human resources employees that were trained to provide benefits-related information. There is no question that this information was material. The failure to inform can constitute a material false representation in the same way as an affirmative misstatement, especially in the context of a fiduciary relationship. Diduck, 974 F.2d at 276. In addition, “[t]o hold a defendant liable for fraud it must be shown that he acted with ‘knowledge,’ that is, he either knew of the falsity of the representations or believed them to be false.” Id. This knowledge element cannot be demonstrated through negligent behavior, but may be met if a “defendant makes a representation with reckless disregard as to its truth.” Id. In this case, the Court finds that the knowledge element has been met. By limiting the information it provided to its human resources employees, NU guaranteed that the representations made to employees by the company would be incomplete. While incompleteness does not automatically constitute falsity, the confidentiality policy promulgated by NU certainly does show a “reckless disregard” for truth that satisfies the knowledge element of common law fraud. In addition, the intent of NU’s policy can be considered as an encouragement to employees to retire without waiting for an early retirement program to be released. While intent is difficult to prove, in this context the Court may infer intent because of NU’s special relationship with Plaintiffs. The Second Circuit has said that the ordinary requirements for proof are relaxed when the fraud is alleged to have been committed by a fiduciary, as long as there is a “substantial certainty” that a nondisclosure would be relied upon by a plan beneficiary. Id. at 278; see also Mathews v. Chevron, 2002 WL 826804, *11, 2002 U.S. Dist. LEXIS 7738, *31 (“[U]nder ERISA, employers are deemed to intend the reasonably foreseeable consequences of their misinformation when that misinformation induces earlier than otherwise retirements.”). In other words, a reasonably foreseeable consequence of the confidentiality policy—and resulting misinformation—was that employees who might otherwise delay their retirement would instead retire. This action would be to these employee’s detriment, as they would fore-go inclusion in a forthcoming early retirement program. The reliance element of fraud is essentially causation in fact. Id. Thus, NU’s conduct need not have been the “exclusive inducing cause” of plaintiffs’ actions, but only an “essential or inducing cause.” Id. As discussed later in this opinion, the statements and/or omissions made by NU to each Plaintiff was an “essential or inducing canse” of their retirement, notwithstanding the other life circumstances that contributed to their decision to retire. Finally, the damage element has been met in this case, as each plaintiff who retired before the offering of an early retirement program suffered a monetary loss. III. FINDINGS OF FACT At trial, the Court heard evidence on two general issues: first, the evolution of NU’s early retirement programs; and second, the particular circumstances of each Plaintiffs’ retirement from NU. The Court will now review its Findings of Fact on these issues: A. Early retirement programs at NU In 1987, NU was developing a strategy to prepare for the challenge of competition in a deregulated environment. NU concluded that to be financially competitive it should implement cost-saving initiatives, including contracting its workforce. In 1987, NU conducted a survey of downsizing techniques employed by other large companies. As a result of this survey, NU was presented with numerous cost-saving tools, including the implementation of early retirement programs. These type of programs were recommended to NU management as a way to meet their objective to reduce its employee base. At this point, it might be helpful to briefly describe the concept of “early retirement programs.” Various terms were used throughout trial to refer to these type of early retirement programs, such as “golden handshakes” or “special retirement programs” or “incentives.” In essence, early retirement programs differed from “normal” retirement at NU in the following respects: employees were eligible for early retirement at age 55, while normal retirement eligibility began at age 65; and those taking early retirement would have a reduction to their accrued benefit based on early commencement of that benefit, while those taking normal retirement would get their entire accrued benefit. The Court will also take this opportunity to outline NU’s basic corporate structure vis-a-vis employee benefits. The corporate benefits section, a division of the human resources group, was responsible for the design and administration of benefits for NU, including early retirement programs. During most of the relevant time period, Keith C. Coakley (“Coakley”) was the senior manager in the corporate benefits section. As such, Coakley was the senior manager at NU most responsible for retirement- and benefits-related work. Coakley’s direct supervisor in 1991 was Cheryl Grise (“Grise”), who was then assistant vice president of human resources. At that time, Barry Ilberman (“Ilberman”) was the vice president of human resources, and was supervised by NU president and chief operating officer Bernard M. Fox (“Fox”). Fox reported to William B. Ellis (“Ellis”), chief executive officer of NU and chairman of its board of directors. In May 1991, Grise assumed the post of vice president of human resources, replacing Ilber-man. For some time before that, she had assumed Ilberman’s supervisory duties with regard to employee benefits. With that background in mind, the Court will now discuss NU’s exploration of early retirement programs. NU was initially reluctant to offer early retirement programs because it believed that they were too costly, especially when compared with other options such as layoffs in its workforce. In June 1990, however, NU offered a limited early retirement program to 25 employees. In order to qualify for the June 1990 early retirement program, an employee had to have at least ten years of service to NU, and had to be at least 55 years old. The June 1990 early retirement program included an additional five years of service, and an additional five years for pension calculations. Another feature of the 1990 early retirement program was a $550 per month “social security bridge” for those employees who had not yet reached the age of 62. In late 1990 it was still a NU priority to cut costs. In early-Deeember 1990, Coak-ley asked NU’s actuaries, Towers-Perrin, to prepare a report on the potential costs of a system-wide early retirement program. The Towers-Perrin report, dated December 13, 1990, reviewed NU’s June 1990 early retirement offering. In addition, the report responded to NU’s request that Towers-Perrin review the cost of extending a similar program system-wide, including officers, employees exempted from the previous program, and clerical/technical employees of the entire company. The proposed eligibility for this system-wide program was those employees aged 55 and older with at least ten years of service. The proposed program included a monthly $550 “social security bridge” to age 62, and an additional monthly $275 social security bridge after age 62. In addition, the program included an addition of five years of service. The Towers-Perrin report assumed a 1991 offering date, included a suggested design for the program, discussed issues surrounding NU’s capability to provide the program from the retirement plan, and estimated acceptance rates and pension costs. Listed as “next issues” in the report were: “resolve tax reform required changes, finalize program design, implement program (prepare communications materials, prepare individualized employee calculations, train counselors, hold employee briefings/Q & A sessions), amend plan, and determine financial impact.” Later in December 1990, Coakley made a presentation to the NU management committee, which focused on various cost-saving downsizing scenarios, including system-wide early retirement programs. The management committee consisted of NU’s top executives, including Fox and Ellis. While Coakley does not recall presenting the actual Towers-Perrin report to the management committee during this presentation, Coakley referenced the report in his preparation for the meeting. During his presentation, Coakley presented six different downsizing options, four of which were various versions of early retirement programs. The remaining two options were both severance programs not tied to retirement. Coakley’s presentation of the four early retirement programs for employees included assumed acceptance rates, cost estimates with potential savings, descriptions of targeted employees, and early retirement net reduction calculations for various classes of employees. Of the four versions of early retirement programs offered by Coakley, two involved giving employees six months pay at retirement in a lump sum at retirement, while the remaining two offered three months pay at retirement in a lump sum. Finally, the four versions reflected variations of the categories of eligible employees. In January 1991, Coakley followed up his December 1990 presentation to the management committee with a meeting with Fox. During that meeting, Coakley discussed the six downsizing options he had presented to the management committee, including both the severance alternatives and early retirement programs. With regard to the early retirement programs, Coakley’s agenda with Fox included a discussion of issues such as social security supplements for those employees past age 62, special incentives for employees with long years of service, legal issues related to extended periods, and the regrouping of various classes of employees. The materials prepared for the meeting included plan design materials, sample pension calculations, and detailed cost charts. Soon after, Grise instructed Coakley to ask Towers-Perrin to do some development work for a possible system-wide early retirement program. Coakley instructed one of his employees, Paula Roberge (“Roberge”), to communicate with Towers-Perrin on this work. Roberge was a human resources representative and benefits analyst in the corporate benefits section. Roberge was supervised by Coakley, who testified that any communications Roberge had with Towers-Perrin were most likely passed by him as well. From March 1991 through April 1991, Towers-Perrin was involved in developing an early retirement program. Towers-Perrin’s documents reflect that in March 1991 NU was “considering” an early retirement incentive program which was expected to result in a one-time cost of about $25 million. During this same month, NU issued a “Confidential Position Paper” regarding the potential retirement of certain Fossil/Hydro units, which contemplated, among other things, an early retirement program for these employees, as well. On April 20, 1991, Roberge discussed with Towers-Perrin a possible early retirement program that would be offered to NU employees on May 15, 1991 with an acceptance date of June 30, 1991. As of April 24, 1991, Towers-Perrin had prepared a detailed summary of the proposed early retirement program. The proposed program required an employee to be age 55 and have at least ten years of service with NU. It also included three months of severance pay, and the addition of five years of service and five years of age for pension calculations. In addition, it included a monthly $550 “social security bridge” to age 62. On April 29, 1991, Roberge sent a memo to Betsy Magnarelli (“Mag-narelli”), an employee of Towers-Perrin, which indicated that NU was “finalizing” the details of this early retirement program, including the “titles and groups of employees to be offered this program.” On that same day, Coakley sent a confidential memo to Grise about preparing communications with employees about the early retirement program, including sample text for these communications. On April 30, 1991, Roberge sent another memo to Magnarelli, which enclosed a “draft that [Coakley was] proposing for an early retirement program.” According to Towers-Perrin records, Roberge reiterated that NU wished to release this program to employees by May 15,1991. In early May 1991, Roberge continued to communicate with Towers-Perrin about details of the program. During this month, employee communication materials were being finalized by NU and Towers-Perrin, and a “first draft” of the document employees were to be given was created. As of this point, however, NU had yet to receive final approval of the early retirement program. A May 3, 1991 memo from Roberge to Magnarelli mentioned that May 1, 1991 retirees would not be eligible for the early retirement program “since we have yet to receive a decision.” Ro-berge relayed a similar message again on May 8, 1991, when she told Debbie Hart (“Hart”), another Towers-Perrin employee, that “we have not received the official word yet.” It appears that sometime around May 9, 1991, NU decided to put the early retirement program on hold. Roberge testified that in a May 9, 1991 phone call to Mag-narelli, she told Towers-Perrin that the program scheduled for June 30, 1991 was in a “holding pattern” because the “[Connecticut Department of Public Utility Control] [hereinafter “CDPUC”] would not believe the expense.” Roberge did not have any personal recollection of this conversation, but instead based it on a “Chronology of Events” prepared by Towers-Perrin in preparation for this litigation. This chronology — apparently based on Towers-Per-rin documents — reads as follows: “5/9/91. Note to file from [Magnarelli] regarding phone call with [Roberge], Window no longer on for 6/30. In holding pattern. [CDPUC] would not believe expense.” Roberge testified that it was “more probable than not that the comment about the expense was made to her by [Coakley] ... and that [she] passed those words on to Towers-Perrin.” Coakley did not disagree with Roberge’s assessment. No matter the cause, the record shows that NU’s plan for the May 15, 1991 release of an early retirement program was in a “holding pattern” as of May 9, 1991. In terms of what this “holding pattern” meant, Coakley agreed that the holding pattern was similar to a plane circling an airport waiting to land. During the holding pattern, Towers-Perrin continued to work on program details. For example, on May 14, 1991, Towers-Perrin wrote to NU: “since most of the work on this portion of the package is behind us, we should be able to move quite quickly when you decide to proceed.” On that same date, Towers-Perrin forwarded to NU a second draft of the plain language summary that employees would use with respect to the system-wide early retirement program. In addition, a coverage test was prepared by Towers-Perrin for the program on May 22, 1991. Coakley testified that the purpose of this work was to be prepared for the time when NU decided to proceed with the system-wide early retirement program. He explained that this advance work was helpful, since human resources typically did not have much time to implement such programs once final approval was made. On July 17, 1991, Roberge called Mag-narelli to inform Towers-Perrin that “the open window [a euphemism for an early retirement program] seems to be on again, possibly effective 10/1/91.” There is a dispute between the parties about which early retirement program Roberge was referring to when she made this statement. NU contends that during the time that the system-wide early retirement program was in a “holding pattern,” Towers-Perrin continued to work on a smaller early retirement program for NU’s Fossil/Hydro units. NU argues that in her July 17, 1991 telephone call, Roberge was referring to the smaller Fossil/Hydro early retirement program, not the system-wide early retirement program put on hold in May 1991. In her affidavit of direct testimony, Ro-berge supports this interpretation. At trial, however, Roberge testified that she had no personal recollection of her conversation with Magnarelli. She explained that she formulated her affidavit of direct testimony based on the Towers-Perrin “Chronology of Events” document which was prepared by Towers-Perrin in preparation for this litigation. Similarly, while Coak-ley’s affidavit of direct testimony offered the same interpretation of Roberge’s statement, he testified at trial that he had no personal knowledge of Roberge’s conversation with Magnarelli. In contrast to Ro-berge’s and Coakley’s suppositions about the July 17, 1991 conversation, Plaintiffs offered their own hypothesis — that Ro-berge was referring to the system-wide early retirement program. The personal recollections of those involved in the July 17, 1991 conversation, or lack thereof, leave the Court with the job of reaching the most logical conclusion about the meaning of Roberge’s words to Magnarelli. After reviewing both parties’ analysis, the Court concludes that the most likely scenario is that Roberge was referring to the system-wide early retirement program, rather than the smaller Fossil/Hydro program. The Court reaches this conclusion for several reasons. First, it is only logical that the “on again” language would refer to an early retirement program that was once “off.” In this case, only the system-wide early retirement program was in a “holding pattern.” Coakley testified that work on the smaller Fossil/Hydro early retirement program had continued throughout the period the system-wide early retirement program was in a “holding pattern.” Second, the circumstances of Roberge’s employment with NU make it unlikely that she would communicate with Towers-Perrin about the smaller Fossil/Hydro program. Both Roberge and Coakley testified that they did not believe that Roberge had ever worked on the Fossil/Hydro program. Several other factors are of note. The Towers-Perrin memo indicates that the early retirement program referenced by Roberge would be “on again” as of October 1, 1991. This is the same date that the system-wide early retirement program was actually announced to employees. The main contrary evidence is the testimony of Roberge and Coakley. This testimony was based not on their personal recollection, but almost exclusively on the “Chronology of Events,” a document the Court deems inherently unreliable. This exhibit was not prepared contemporaneously, but rather prepared by Towers-Perrin for NU around the same time that this litigation was filed, five years after the events it purported to summarize. It was also prepared in anticipation of litigation, as reflected in the fact that it was faxed to NU’s outside counsel, its counsel in this litigation, just one day after it was sent to NU. Absent these infirmities, the document is also unreliable because many of its statements contain several levels of hearsay, and the Court has not been shown its underlying documents. Even if Roberge’s and Coakley’s reliance on this document, as opposed to their personal recollection, were reasonable, their conclusion is inconsistent with the rest of the document. In a later entry, dated August 20-23, 1991, it is indicated that the smaller Fossil/Hydro early retirement program was “back on.” If the Court were to credit Roberge’s and Coakely’s speculation about the July 17 conversation, there would be two separate entries — one on July 17, 1991 and the other on August 20-23, 1991 — which indicate that the Fossil/Hydro program was “back on.” It does not make sense for there to be two entries, almost one month apart, indicating that Towers-Perrin had been told that the same early retirement program was “back on.” In any event, to the extent that the Court must resolve the dispute about Roberge’s July 17, 1991 phone call with Magnarelli, it resolves it as described above — that on July 17, 1991, Roberge told Towers-Perrin that the system-wide early retirement program was back “on.” No matter what Roberge said to Towers-Perrin on July 17, just a few days later Coakley suggested to Grise that NU offer “any [system-wide] one we might be considering for this year” concurrent with a Fossil/Hydro early retirement program in order to limit administrative costs. About one week later, Grise sent a memo to Fox relaying Coakley’s suggestion. Consistent with his recommendation, on August 5, 1991 Coakley prepared timelines for both a Fossil/Hydro early retirement program and a system-wide early retirement program. On that same day, Coak-ley met with Fox to discuss these possibilities. At this point, the Court believes it would be helpful to discuss the role of the CDPUC in relationship to NU. The CDPUC had controls over the rates NU could charge its customers. NU believed that CDPUC decisions affected many of its operational decisions, especially those that would result in expenditures. This included the offering of an early retirement program. Throughout this time period, NU was negotiating a rate increase with CDPUC. In particular, in June 1991, NU took part in a series of hearings with the CDPUC with regard to its request for a rate hike. In early August, NU learned that the CDPUC had approved its rate hike. It is likely that during July and August 1991, NU was working on both the Fossil/Hydro and the system-wide program, as reflected in Coakley’s suggestion to Grise and Fox. NU had a so-called “fire drill” contingency policy, whereby NU would do much of the preparation work for early retirement programs in advance so that NU could be prepared when the board issued its final approval. Coakley suggested at trial that the work being done by Towers-Perrin was “cannibalized” by one program for another. In other words, the general language and provisions which NU had been examining for one program could be used interchangeably with another program. The record does show that on August 20, 1991, Diane Antoni (“Antoni”) of Towers-Perrin spoke with Roberge, who indicated that the “window is back on with a 12/1/91 effective date. Provisions are the same but reasons are different than in May. The eligible group has changed from about 700 to 50 employees.” This was likely in reference to the Fossil/Hydro program, which was offered to a much smaller employee population than the system-wide early retirement program. On August 26, 1991, Coakley presented the Fossil/Hydro early retirement program to the board of directors. On August 30, 1991, Towers-Perrin indicated that the “ÑUSCO open window project is on track once again.” The program referred to in this memorandum involved fifty employees, and contemplated a late-August announcement date. Since the number of employees was so small, this comment was also most likely in reference to the smaller Fossil/Hydro program, not the system-wide early retirement program. Soon after, on September 11, 1991, Towers-Perrin sent Roberge a memorandum with sample statements for “the early retirement program.” It assumed a December 1, 1991 retirement date. On September 16, 1991, employees eligible for the Fossil/Hydro early retirement program received packets in the mail. On September 17, 1991, Magnarelli drafted a memorandum to other Towers-Perrin employees which stated that NU may offer a “new open window” for approximately 500 employees, with a retirement date of December 30, 1991. The memo indicated that Roberge had sent to Magnarelli a description of the program which “has some changes from last time.” The memo indicated that the final decision on the program would be made on September 30, 1991, and that packets would go to employees on October 1,1991. From mid- to late-September 1991, Ro-berge and Towers-Perrin worked on specifications of the program, including lists of eligible employéés and drafts of materials. On September 18, 1991, Roberge sent Magnarelli a list of those individuals who would be eligible for the “next program.” On September 23, 1991, Coakley met with his staff to discuss employee financial counseling in relationship to these early retirement programs. In late-September 1991, various drafts of the program were exchanged between Towers-Perrin and Roberge. Finally, on September 80, 1991, Coakley presented the system-wide early retirement program to the board of directors, and it was approved. The program was announced to employees on that same day, with a response date of November 18, 1991, and a retirement date of December 31,1991. In order to be eligible for the 1991 early retirement program, employees had to have at least ten years of service with NU. For those employees who decided to retire by December 31, 1991, the early retirement program provided to eligible employees a benefit calculation that included a “bonus” of five additional years of service that would lead to benefits being calculated as if the employee were five years older than he actually was. If an employee was younger than age 62 when he retired, he would also receive a “social security bridge” of $550 a month. Employees also had the option to let NU set a retirement date for them prior to December 31, 1992, rather than accept the December 31, 1991 date. If .an employee chose this option, the employee would receive an additional lump sum payment of three months base pay. Of the 631 employees eligible for the 1991 early retirement program, 438 accepted. During the next few years, NU continued to consider efforts to improve its financial condition. NU had set a goal of reducing its workforce by approximately 800 employees, and thus NU had to reduce its workforce by approximately an additional 400 employees. Sometime in 1992-1993, NU began what was called the “Competitive Preparedness Review” (“CPR”). CPR was focused on NU’s Fossil/Hydro organization. The final presentation to the CPR steering committee was made on April 14, 1993. One of the primary recommendations to the CPR steering committee was the implementation of an early retirement program. On May 12, 1993, the final report of the CPR was issued. This report included a summary of an early retirement option, which included cost analysis. The goal was to implement the early-retirement program “near term,” and impact 1994 budgets. There were other ideas that were “immediate,” and would impact 1993 budgets. In a memo written by Towers-Perrin on June 15, 1993, Towers-Perrin discussed numerous details of this early retirement program, and indicated that non-discrimination tests were being performed. On June 28, 1993, Roberge sent a memo to Towers-Perrin which indicated that only minor changes would be made to the materials Towers-Perrin and NU had prepared for the 1991 offering. On June 30, 1993, Roberge wrote a memo to Towers-Perrin which indicated that it was “a go” for an early retirement program. By early July 1993, target announcement and retirement dates for various types of NU employees were listed, the eligible employees had been defined, and a budget had been set up for all program-related costs. By July 13, 1993, drafts of the offering materials were distributed to NU by Towers-Perrin. On July 23, 1993, a “final” draft of the program was developed, with regard to which Roberge indicated “everything is fine.” All materials were scheduled to be ready by July 29, 1993, when envelopes would be collated and stuffed, with July 30, 1993 as the targeted mailing date. On July 30, 1993, the board of directors adopted the 1993 early retirement program. It was announced to employees on August 3, 1993. Materials had been mailed a day earlier. In June 1994, Coakley drafted a memorandum indicating that NU was considering other force reduction programs that would be targeted at a younger population of employees, since NU needed “to retire as many people as possible.” In June 1994, Coakley recommended that NU follow the “current design” for a program. Soon thereafter, in July, August, October and November, 1994, various early retirement programs were announced by NU. On October 24, 1994 the board of directors approved an early retirement program, and on the following day, October 25, 1994, the early retirement program was announced to NU employees. The documentary ■ evidence NU produced in discovery regarding the early retirement programs, but especially the 1993 and 1994 programs, is limited. NU retained the final versions of some plan and program documents, but generally not correspondence or drafts leading to those final versions. B. NU’s confidentiality policy and communication procedures The testimony at trial indicated that during this entire time period NU had'a policy that any discussion of early retirement programs was to be kept confidential until the programs were formally announced. Grise testified that “information relating to the development of early retirement projects was maintained in the strictest confidence until a program was finally approved and implemented.” By “approval,” Grise meant approval by the board of directors and/or chief executive officer Ellis. Grise testified that “[ujntil- a decision had been made to offer an early retirement program, we did not disclose to employees or to anyone else whether we were considering a potential design of an early retirement program or whether to offer such a program.” Coakley testified that Grise instructed him to perform any work on downsizing scenarios “confidentially” and to “disclose it on a need-to-know basis.” Coakley testified that “to the extent that [his] subordinates were involved, [he] asked them not to share their developmental work with anyone until there was a final approval of an actual program by [NU], as it was the policy of [NU] that until such approval, information concerning the internal consideration or the development of any downsizing project, including early retirement, was maintained confidential.” As Coakley wrote in an internal NU memorandum: “Decisions to offer early retirement programs and/or involuntary termination programs are made at the highest levels of management and maintained in the strictest of confidence until approval and implementation. Until approved and announced, no one within the Company is authorized to represent that such programs will or will not be offered.” Coak-ley described the situation this way: until an early retirement program was approved by the board “we really didn’t have a program.” Consistent with this policy, members of the NU human resources group, including corporate benefits counselors such as Turner, were not made aware of the development of offerings, even though Turner and those like him were considered the preferred resource available to NU employees to obtain. information about their re: tirement. Even if corporate benefits counselors were somehow made aware of the development of any early retirement program, they were instructed, to follow NU’s confidentiality policy with regard to these programs, and thus not share the details with NU employees. These type of human resources counselors were an integral part of NU’s benefits structure. NU’s position was that employees were generally advised that their supervisors did not know of potential changes to employee benefits, and that they should instead contact the human resources department with any retirement- or benefit-related inquiries. At trial, the Plaintiffs did not disagree that the human resources department, and those such as Turner, were the most appropriate contact for these type of inquiries. Plaintiffs consistently testified, however, that NU encouraged employees to speak first with their supervisor, who then might refer them to human resources if he could not assist the employee. In particular, Plaintiff Robert L. Klein (“Klein”), a mid-level management employee, testified that he was charged with explaining some human resources policies to his employees. Klein testified that if he couldn’t answer a particular human resources-related question posed to him, he would then refer the employee to the human resources department. C. Plaintiffs’ retirements The Court will now discuss the details of each Plaintiffs’ retirement. 1. Paul Broga Paul Broga (“Broga”) retired on February 1, 1993 at age 55, after 33 years at NU. On August 24, 1992, Broga first indicated to Turner that he intended to retire effective February 1, 1993. In November 1992, Broga met with Turner to discuss his retirement. Broga brought up the topic of early retirement programs, and Turner responded that “there was nothing coming down the pike.” That same month, Broga also approached his supervisor, who was a senior company manager, with similar questions. His supervisor said that he was not aware of any program being offered in the future. Broga asked similar questions of other NU employees, supervisors and senior company managers in January 1993, and received similar responses — that they knew nothing about any such programs. Several factors led to Broga’s decision to retire. At the time of his retirement, Bro-ga, although ten years younger than his wife, had some heart problems. He was aware that the release of an early retirement program in the future was possible, but testified that if in January 1993 he had known that the company was thinking about offering an early retirement program, he would have delayed his retirement. He testified that he felt that he had a good job, and was content to stay as much as a year longer. In September 1993, after his retirement, Broga found out about the 1993 early retirement program. While he was upset, he did not realize that a legal wrong had been committed, and did not retain legal counsel and file a claim until several years later. 2. Robert Gamache Robert Gamache (“Gamache”) retired on July 1, 1993 at age 58, after 27 years at NU. In Fall 1992, Gamache asked Turner about retiring. In May 1993, Gamache asked Turner about the possibility that an early retirement program would be offered. Turner said that he didn’t know of anything. Gamache testified that at this time of his retirement, he was “fed up” with his coworkers and had effectively “thrown in the towel” with his work at NU — all factors which led to his retirement. Gamache understood that anything was possible in the future, and that he was taking a risk by leaving. Gamache testified, however, that if NU had told him it was considering offering an early retirement pro