MEMORANDUM OPINION

JOHN D. BATES, District Judge.

Plaintiff Denise Clark brings this action pursuant to the Employee Retirement Income Security Act of 1974 ("ERISA"), alleging that defendants improperly denied her a significant portion of her retirement benefits. Defendants are the law firm Feder, Semo & Bard ("Feder Semo" or "the firm") (Clark's former employer), the Feder Semo Retirement Plan and Trust ("Plan"), and two former trustees of the Plan, Joseph Semo and Howard Bard. The Court conducted a six-day bench trial beginning on April 23, 2012, on plaintiff's three remaining claims in the case.

Clark worked at Feder Semo for almost ten years before moving on to other employment in 2002. She was distributed retirement benefits after the firm unexpectedly went out of business toward the end of 2005. However, when Feder Semo went out of business, it had insufficient funds to pay out all benefits due under calculations dictated by the terms of the Plan. Lump sum distributions under the Plan's terms could be determined in two ways, and Plan participants were entitled to receive benefits in the amount of whichever determination was larger. One of the two amounts was a definite amount of money that grew as Plan participants accrued a percentage of their annual compensation in benefits and earned interest on the account balance; Plan recipients, including Clark, were regularly notified of this amount. The other was based on a calculation involving a floating interest rate, which was tied to the return on 30-year Treasuries. When the Plan went out of business, this rate was especially low, resulting in benefit distributions — and Plan liabilities — that were especially high. But the Plan had insufficient funds to pay these benefits, so each Plan recipient received a pro rata share of the amount due. Hence, although Clark ultimately received more money than the fixed amount that had been quoted to her, she received only about 53% of the full amount due under the second calculation. She claims both that the Plan's actuarial assumptions were unreasonable given the terms of the Plan and that she was inadequately informed of the risk of loss of her benefits. Furthermore, Clark contends that under either calculation her account balance was credited with a smaller percentage of her annual compensation than she was entitled to receive.

Clark's three claims accordingly involve the administration of the Plan. First, she contends that she was improperly grouped for the purpose of determining her account balance and that Semo and Bard violated their fiduciary duties in failing to correct this error when Clark made a formal appeal. Second, she contends that the firm violated ERISA's disclosure requirements by failing to disclose the risk of loss if the Plan terminated with insufficient funds and the Plan's lack of insurance to protect participants in that contingency. Third, she contends that Feder Semo and Semo and Bard, as Plan fiduciaries, failed to use a reasonable actuarial assumption for interest on the Plan's assets, causing the Plan to be underfunded.

After careful consideration of the evidence at trial, the parties' memoranda, the applicable law, and the entire record herein, and for the reasons set forth in more detail in the findings of fact and conclusions of law that follow, the Court will enter judgment for the defendants on each of Clark's remaining three claims. With respect to the improper grouping claim, the Court concludes that the decision Semo and Bard made on Clark's appeal was reasonable. With respect to ERISA's disclosure requirements, the Court finds that Clark was not harmed by any failure to disclose the risk of loss at Plan termination and the Plan's lack of insurance. And with respect to the interest rate assumption, the Court finds that the defendants did not breach their fiduciary duties to maintain the Plan on a sound actuarial basis.

I. Case History

This case has already gone through several iterations in this Court. Clark's amended complaint was filed on May 28, 2008 [Docket Entry 28]. On December 17, 2007, the Court dismissed some of Clark's claims but denied defendants' motion to dismiss on the remaining claims. Clark v. Feder Semo & Bard, P.C., 527 F.Supp.2d 112, 118-19 (D.D.C.2007) ("Clark I"). On March 22, 2012, the Court granted summary judgment for defendants on all but plaintiff's improper grouping claim. Clark v. Feder Semo & Bard, P.C., 697 F.Supp.2d 24 (D.D.C.2010) ("Clark II"). However, on September 13, 2010, the Court issued its decision on reconsideration, which vacated the partial grant of summary judgment. Clark v. Feder Semo & Bard, P.C., 736 F.Supp.2d 222, 225 (D.D.C.2010) ("Clark III"). After this decision, the Court ordered Clark to file a statement precisely detailing the nature of her remaining claims so that both the Court and the parties would have a concrete grasp of her at-times evolving allegations. See Order of Sept. 30, 2010 [Docket Entry 85]. In Clark's Statement Detailing Nature of Claims [Docket Entry # 87] ("Pl.'s Statement"), she asserted five theories of recovery. On September 7, 2011, the Court granted summary judgment for defendants on two of these claims. Clark v. Feder Semo & Bard, P.C., 808 F.Supp.2d 219 (D.D.C.2011) ("Clark IV"). Specifically, the Court concluded that Clark could not bring an "anti-cutback" claim because Clarks' benefits were not reduced by an amendment to the Plan. Id. at 226-29. The Court also concluded that Clark could not bring a claim against Semo and Bard for permitting distributions to the firm's founder and his wife in 2002 and 2005. See id. at 232-34. The case then proceeded to trial on the three remaining claims.

II. Findings of Fact

The Court conducted a bench trial over a six-day period beginning on April 23, 2012. Based on the record established at trial, the Court makes the following findings of fact. The Court will first introduce the individuals involved in this case and then explain the evolution of the firm and the Plan. Next, the Court will describe Clark's treatment under the Plan, including the decision made by the firm upon her appeal of the amount of her distribution. The Court will then explain how, in relevant part, the Plan was funded, including the actuarial assumption of interest that is the basis of one of Clark's claims. The Court will then describe the expert testimony heard at trial about the interest rate assumption. Finally, the Court will review the evidence heard about the Plan's summary plan description.

a. Individuals Involved with the Firm and the Plan

Gerald Feder founded the Feder Semo law firm, then known as "Feder & Associates,"1 in the mid-1970s. Feder Tr. 8:2-9.2 The Plan was founded in the mid-1980s. Feder Tr. 9:14-16, 29:11-16. Feder was the original sponsor and fiduciary of the Plan. Feder Tr. 11:11-12:7. The Plan was initially drafted to be primarily to the benefit of Feder and his wife, Loretta, who was also employed by the firm. Tr. 839:12-21 (Anspach). The Feders retired from the firm on December 31, 2001, but continued to receive a percentage of the firm's revenue for some time thereafter. Feder Tr. 13:6-9, 15:12-19.

Denise Clark is an attorney with an L.L.M. degree and employee benefits certification from Georgetown University Law Center. Tr. 29:10-17 (Clark). She practices in the "employee benefits" area of law, among other related areas. See Tr. 29:20-30:3, 39:18-40:1 (Clark). Clark was hired in 1993 to work at the Feder Semo law firm, then known as "Feder & Associates, P.C.," by Gerald Feder. Tr. 30:6-15 (Clark). After becoming a non-equity partner in 1997, Clark became an equity partner in the firm as a Class A Shareholder in 2000. See Tr. 31:5-35:9 (Clark); Pl.'s Ex. 7. In 2001, she became managing partner of the firm. Tr. 36:25-37:8 (Clark). Clark resigned from the firm in mid-2002 to become the general counsel of the Hotel Employees and Restaurant Employees International Union Welfare and Pension Fund. Tr. 37:20-38:11, 117:21-118:11 (Clark).

Howard Bard is an attorney who began working for the firm roughly contemporaneously with Clark in 1993. Tr. 30:18-20 (Clark). Bard practices in the employee benefits area of the law. Tr. 313:1-14 (Bard). He became a non-equity partner in the firm in 1997 and a Class A Shareholder in 2000 when Clark did. Tr. 31:5-31:7 (Clark); Tr. 319:6-10 (Bard); Pl.'s Ex. 7. He became the managing partner when Clark left the firm in 2002 and remained in that position until the firm terminated in 2005-2006. Tr. 313:17-24 (Bard). It is fair to say that, with one notable exception described below, Bard and Clark were treated equivalently in relevant respects during their time together at Feder Semo.

Joseph Semo is an attorney with substantial experience in the employee benefits field dating to the mid-1970s. Tr. 472:25-475:15 (Semo). His own firm merged with Feder & Associates in July 1998. Pl.'s Ex. 6; Tr. 31:19-32:14 (Clark); Tr. 478:15-17 (Semo). According to the terms of the merger, memorialized in a "Business Combination Agreement" ("BCA"), Semo became a Class A Shareholder in Feder Semo at that time. See Pl.'s Ex. 6.

Robert Landau is an attorney who was an employee of Feder Semo until approximately January 2000. Tr. 428:1-21 (Landau). Landau appears to have been treated roughly equivalently with Bard and Clark while at the firm, but left the firm before becoming an equity partner.

Mark Nielsen is an attorney who was also an employee of Feder Semo until the firm's dissolution in 2005. Tr. 801:3-802:5 (Nielsen). He had substantial experience with ERISA prior to joining Feder Semo, including an L.L.M. in labor and employment law with a specialty in employee benefits law and subsequent work as an investigator at the Department of Labor on ERISA compliance. Tr. 801:2-17 (Nielsen). Nielsen was involved with Semo and Bard's resolution of Clark's appeal of her distribution of benefits from the Plan. See Tr. 803:20-805:12 (Nielsen).

William Anspach, an attorney, was the Plan's outside counsel from the drafting of its restatement in the early 1990s until the Plan's termination. Tr. 838:4-839:14 (Anspach). Anspach has significant experience in the employee benefits area of law. See Tr. 831:5-836:1 (Anspach).

Dennis Reddington was the firm's enrolled actuary from the mid- to late-1990s until the Plan's termination. Tr. 602:13-16 (Reddington). He has actuarial experience dating to approximately 1989. Tr. 599:20-601:11 (Reddington).

b. The Firm

The BCA that merged Feder & Associates with Semo's firm established an "Executive Committee" for the management of the firm. See Pl.'s Ex. 6 at P0672. Since the BCA indicated that "[t]he members of the Executive Committee shall consist of the Class A Directors only," the original members of the executive committee after the merger were apparently only Feder and Semo. Bard and Clark became members upon becoming equity partners in 2000. At the time of the firm's dissolution, the executive committee appears to have consisted only of Bard and Semo. See id.; Pl.'s Ex. 7. At trial, members of the firm referred to the executive committee as the "board of directors," although others in addition to Bard and Semo appeared to participate in "board" meetings without it being clear to all participants who was and was not a voting member. See Tr. 314:8-315:11, 404:15-21 (Bard); Tr. 813:10-13 (Nielsen).

The most significant relevant event regarding the firm, at least for present purposes, was its dissolution in 2005. On July 21, 2005, Semo was informed by the general counsel of the Union Labor Life Insurance Company ("ULLICO"), the firm's largest client, that the firm would be losing the account. Tr. 488:17-489:19 (Semo). The loss of the firm's biggest client was apparently quite unexpected. Tr. 488:25-489:5 (Semo). After first contemplating trying to downsize, the firm eventually decided to shut its doors. Tr. 490:18-25 (Semo); see also Tr. 49:9-16, 119:7-120:12 (Clark); Pl.'s Ex. 44. The firm essentially ceased operations by the end of 2005. See Tr. 815:5-25 (Nielsen).

c. The Plan

The Plan was established on October 1, 1993, and the terms of the Plan were amended several times thereafter. See Pl.'s Exs. 1-3. The firm was the Plan's "plan administrator." Pl.'s Ex. 1 § 2.25; Pl.'s Ex. 3 § 2.30; Tr. 500:23-25 (Semo). The firm administered the Plan through its executive committee and its staff. Tr. 501:1-7 (Semo). The Plan also contained the following statement about Plan fiduciaries:

Exs. 1 & 3 § 11.1.

i. Benefits Under the Original Plan

The calculation of retirement benefits for a participating employee under the Plan was a multi-step process. This process did not change under the various iterations of the Plan. The Plan was a "cash balance" or "defined benefit" plan. Pl.'s Ex. 1 § 1.5. Under such a plan, each participant has a "hypothetical" (or "theoretical") account balance. Id.; see also Tr. 180:24-181:22 (Poulin). The hypothetical account balance is created by the making each year of hypothetical contributions to the account by the firm, as well as interest adjustments. See Pl.'s Ex. 1 § 1.5; Tr. 182:14-15 (Poulin). The amount of the yearly contribution made to each employee's hypothetical account varied under the iterations of the Plan, but the basic idea remained the same from the outset: contributions were determined by multiplying the amount of each participating employee's compensation in the "plan year" by a certain percentage. See Pl.'s Ex. 1 § 2.3; Pl.'s Ex. 2 at P0063; Pl.'s Ex. 3 § 2.3. Different employees received hypothetical contributions in amounts that varied not only according to salary, but also how they were "classified" under the Plan. That is, the hypothetical contribution for each employee was based on multiplying that employee's salary by a percentage that depended on which "group" (or "class") that employee fell into under the Plan's terms. Under the original iteration of the Plan, there were three such groups. The Plan provided:

Pl.'s Ex. 1 § 2.3(b). The original iteration of the Plan defined the groups as follows: "Class A shall include all Participants who are shareholders of the Employer, Class B shall include all Participants who are classified as officers and not members of Class A, and Class C shall include all Participants who are not Shareholders of the Employer or officers." Pl.'s Ex. 1 § 5.1(a). In addition to employer contributions made to the hypothetical account each year, the hypothetical account balance would also increase each year by an interest percentage, as defined in the Plan. See id.

The next step in determining the retirement benefit due to a participating employee was "accumulating" the hypothetical balance based on the age of the employee at the time of the distribution of benefits. See Pl.'s Ex. 96 ¶ 7; Pl.'s Ex. 1 § 5.1(a). Under the terms of the Plan, the hypothetical account balance would be accumulated (or "projected") at a 7% interest rate until the time when the employee would reach the age of 65 (normal retirement age). Tr. 181:8-182:1 (Poulin). In other words, one would take the employee's age at termination, determine how much time would elapse until that employee would turn 65, and compound (or "accumulate" or "project") the account balance at a certain interest rate for that amount of time. Next, the quantity would be further accumulated, again at a 7% interest rate, to the employee's expected date of death, as provided for in mortality tables. See Tr. 189:2-14 (Poulin). As an expert indicated at trial, the concept being effectuated is that this amount is the amount that would be necessary to provide the beneficiary with an annuity for the rest of their life from retirement at age 65. See Tr. 183:12-15 (Poulin).3 The Plan referred to this amount — the hypothetical account balance accumulated from current age to expected date of death — as the "Accrued Benefit." See Pl.'s Ex. 1 § 5.1(a).4 The Accrued Benefit is an important quantity, because the amount of the retirement benefit actually received by the employee is calculated from the Accrued Benefit. The benefit actually received by the participant, however, depended on how the employee opted to receive it.

The Plan afforded participants two options for receiving their benefits. The option that the Plan referred to as the "Normal Retirement Benefit" was an annuity, payable from age 65 to the participant's death. See Pl.'s Ex. 1 § 5.1; Tr. 183:6-15 (Poulin). As a factual matter, nobody seems to have taken this option. The other option was to receive the retirement benefit as a lump sum following termination from the firm. The Plan defined how a lump sum would be calculated and contained certain rules about when the lump sum could be received by "terminated participants."

With respect to timing, under the original iteration of the Plan, if the terminated participant's accrued benefit was less than or equal to $15,000, the participant could receive the lump sum as soon as administratively feasible following the end of the plan year in which employment was terminated. Pl.'s Ex. 1 § 8.6. If a terminated participant's accrued benefit exceeded $15,000, the participant could not receive the lump sum until five years after the end of the plan year in which employment was terminated. Id.5 As explained below, these rules were amended over the time period at issue here.

The amount of the lump sum that a participant would receive was the larger of two quantities. Tr. 184:11-185:1 (Poulin); Tr. 604:15-22 (Reddington). The first amount was the "present value" of the Accrued Benefit calculated using the Plan assumptions — that is, using the 7% interest rate. This amount would be equal to the hypothetical account balance at termination — the account balance before accumulation. See Tr. 604:15-22 (Reddington); Pl.'s Ex. 71.6 The second amount was the "present value" of the Accrued Benefit calculated using the so-called "GATT rates." See Tr. 189:15-190:1-2 (Poulin); Tr. 604:15-605:25 (Reddington); see also Pl.'s Ex. 2 at P0062. The GATT interest rate is a "moving average of the 30-year Treasury rate." Tr. 743:4-15 (Altman); see also Tr. 185:2-15 (Poulin); Pl.'s Exs. 24, 25 at D766.

One of the two possible amounts for the lump sum was calculated using a variable interest rate; hence, the lump sum actually due to Plan participants could vary from the hypothetical account balance to different degrees over time. The lower the interest rate used in a present value calculation, the larger the present value will be, since money received in the future will be less discounted. Thus, the lower the GATT rate, the greater the lump sum distribution would be in excess of the hypothetical account balance. This disparity is known as the "whipsaw effect." Tr. 204:22-205:13 (Poulin); Tr. 605:12-606:5 (Reddington). Because the whipsaw effect is based on taking the present value of benefits that would be received at age 65, the effect will be more significant for younger participants. See Tr. 282:8-10 (Poulin). Quite importantly, as it turned out, the GATT rate was substantially lower than 7% during the time period at issue in this case and many of the Plan participants were significantly younger than 65. Hence, the lump sum due to many participants was substantially larger than their hypothetical account balances. See Tr. 285:8-12 (Poulin).

ii. Revisions to Plan

The first revision to the Plan that is relevant to this case was the "Third Amendment," executed October 1, 1998, a few months after Semo joined the firm. Pl.'s Ex. 2 at P0064.7 The Third Amendment redefined the grouping of employees and created an additional group as follows:

Pl.'s Ex. 2 at P0066. Gerald Feder clearly qualified under this amendment for treatment under Class A. Tr. 806:22-25 (Nielsen). But as this Court has previously noted, the Third Amendment contained a patent ambiguity: Class A Shareholders who were born on or after January 1, 1950, could be placed in either the second group (which received an employer contribution, or "service credit," of 20% of compensation) or the third group (which received an employer contribution of 10% of compensation). See Clark II, 697 F.Supp.2d at 32. At the time of the Third Amendment's enactment, Semo was in this category; Clark and Bard were not because they were not yet Class A Shareholders. See Tr. 540:5-25 (Semo). The BCA explicitly provided, however, that Semo would be "classified under the Firm's pension plan as a participant entitled thereunder to a rate of contribution of twenty percent (20.0%)," Pl.'s Ex. 6 § 3(i), assuring him treatment as a Class B participant. See Tr. 539:10-541:3 (Semo). When Bard and Clark became Class A Shareholders in March 2000, they entered into this ambiguous category.8

A minor amendment to the Plan was made in October 2000. That amendment added an exception to the rules about when participants could receive lump sum distributions. Under the exception, a participant who had at least five years of service with the firm prior to termination would be eligible for a lump sum distribution upon turning age 63 and incurring a one-year break-in-service. Pl.'s Ex. 2 at P0067. The only other change was to add Clark and Bard's names to the Plan. Id.

In August 2003, the terms of the Plan were completely restated, effective October 1, 2002, with more substantial changes from the earlier version, apparently to bring the Plan into compliance with newly enacted laws that are not relevant here. See Pl.'s Ex. 3 § 1.2, Pl.'s Ex. 37. The August 2003 restatement of the Plan made one substantial relevant change. The restated Plan indicated:

[T]he groupings shall be defined as follows:

Pl.'s Ex. 3 § 5.1(d). Hence, the restated Plan eliminated the ambiguity between the second and third groups by putting individuals who could have previously been in either group clearly into the second group, which received an employer contribution ("service credit") of 20% of compensation.9 The restated Plan also stated that "[f]or the period commencing October 1, 1993 and ending September 30, 1998, the groupings were defined as follows: Group A included all Participants who were shareholders of the Employer, Group B included all Participants who were classified as officers and not members of Group A[,] and Group C included all Participants who were not Shareholders of the Employer or officers." Pl.'s Ex. 3 § 5.1(e). The restated Plan made no explicit mention of any other period.

Anspach testified at trial that he decided to make this revision to the groupings while preparing the restatement to the Plan for compliance with the statutory changes. Anspach testified that he intended to resolve the ambiguity in the Plan language, but had no intention to change the grouping of any specific individuals by making the change, although he also stated that he "may have had a conversation with somebody at the firm" about the groupings but could not recall. See Tr. 923:13-16, 924:22-925:7, 935:17-21 (Anspach). Anspach also testified that he did not intend the change in the groupings to have retroactive effect prior to the restatement's effective date of October 1, 2002. Tr. 927:15-928:10 (Anspach). Anspach sent a "cover memorandum" to the firm about the 2003 restatement, which focused on the new statutory requirements and did not mention the changes to the grouping section. See Pl.'s Ex. 37.

In September 2003, the firm made a significant change to the Plan by freezing altogether the accrual of benefits. The firm amended the Plan to include the statement that "[t]here shall be no further accruals of benefits after September 30, 2003." Pl.'s Ex. 38 at P0218. The other provisions of the Plan did not change.

On July 29, 2005, shortly after the firm lost its biggest client, the firm amended the Plan to remove the timing restrictions on when former employees with accrued benefits of greater than $15,000 could take their lump sum distributions. See Tr. 570:5-572:25 (Semo). The Plan was amended to add that, effective August 1, 2005, the restrictions regarding participants with more than $15,000 in accrued benefits applied only to "Participants who are or have been a shareholder of the Employer on or after December 31, 2004." Pl.'s Ex. 45 at D0124. In other words, former employees such as Clark were immediately eligible to take their lump sum distributions. Finally, the firm terminated the Plan completely on September 26, 2005. Pl.'s Ex. 50 at D0127. From a benefits perspective, the result of the Plan's termination was that everyone, rather than just former employees like Clark, became immediately eligible for lump sum distributions of their Plan benefit.

d. Treatment of Bard and Clark Under the Plan

i. Clark's Grouping and Benefits

While an employee of the firm, Clark regularly received benefit statements regarding her participation in the Plan which provided her hypothetical account balance, contribution by the firm, and calculated annuity amount. Tr. 57:7-15 (Clark). Upon leaving the firm, Clark had an accrued benefit in excess of $15,000, so under the Plan's terms at that time she was not immediately eligible to receive a distribution upon separation. After the firm lost its largest client and amended the Plan to remove the timing restrictions on lump sum distributions, Clark received a letter from Semo, dated September 2, 2005, informing her of the change. See Pl.'s Ex. 46 at D0001; Tr. 49:17-50:20 (Clark). The letter indicated that Clark was entitled to a monthly annuity benefit in the amount of $4,860.65 or a lump sum payment of $227,647.75. Id. at D0004. The letter also included election forms for requesting the benefit, which Clark did not return. Clark emailed Semo to ask how her benefit had been calculated and requested Reddington's calculation of the lump sum amount. Tr. 51:9-20 (Clark); Pl.'s Ex. 47 at D0017.

On September 12, 2005, Semo provided Clark with Reddington's calculation worksheet, which has been admitted into evidence. Pl.'s Ex. 47 at D0019-20. The worksheet indicates that Clark received a benefit accrual of 8% of her annual compensation for each of the four plan years ending on September 30, 1995, to September 30, 1998. The worksheet indicates that Clark received a benefit accrual of 10% of her annual compensation for each of the four years for the plan years ending on September 30, 1999, to September 30, 2002. In other words, Clark was placed in "Group D" for the years prior to her becoming an equity partner (Class A shareholder) and was placed in "Group C" for her remaining years as a shareholder of the firm. Clark's hypothetical account balance as of September 30, 2005, was $159,815.

Clark then received another letter from the firm dated September 30, 2005. See Pl.'s Ex. 53. This letter indicated that the Plan was terminating and distributing all benefits, making all Plan participants eligible for immediate lump sum distributions. The letter referenced the most recent amendment to the Plan (removing the timing restriction on former employees) and stated that "[a]t that time, it was believed that sufficient funds existed in the Plan to make full distributions to all participants with small cutbacks to the firm's shareholders." Id. at D0021. However, the letter explained that, due to "the dramatic decrease in GATT rates over the past few years, on a termination basis the Plan has unfunded liabilities" and that therefore each participant's lump sum was "proportionately reduced so that the aggregate amount paid equals the trust assets." Id. at D022; Tr. 68:23-69:15 (Clark). Accordingly, the letter indicated that Clark was only eligible for a lump sum payment of $166,541.71.

Clark then directly emailed Anspach. Tr. 70:14-73:20 (Clark). Clark inquired about the reduction in her lump sum amount, as well as her accrual of benefits and other issues. Pl.'s Ex. 55 at P0625. Anspach explained that, as a result of a drop in GATT rates, the lump sum due to Clark under the Plan's terms had actually increased to $312,380.83 from $227,647.56 in the time between the first and second letters.10 Id. at P0624. However, benefits were then reduced pro rata by approximately 47% so that benefits in the aggregate would equal the Plan's assets. Id. at P0624. Anspach noted that the amount of $166,541.71 was still greater than Clark's hypothetical account balance, which was $158,899.96. Id.

On October 17, 2005, Clark returned the benefits election form. See Pl.'s Ex. 58. At that time, she also sent the firm a formal "Appeal of Benefit Calculation and Plan Termination" with several questions about her benefit, including questions about the applicable percentage by which she accrued benefits and whether there were any changes to the "actuarial earnings assumptions." See Pl.'s Ex. 59; Tr. 74:19-76:4 (Clark). Clark received a letter from Anspach dated December 14, 2005, with responses to each question. See Pl.'s Ex. 85; Tr. 76:16-77:24 (Clark). With respect to Clark's accrual percentage, the letter stated: "For plan years ending 09/30/95-09/30/98, 8% was the applicable percentage. For plan years ending 09/30/99-09/30/02, 10% was the applicable percentage. We note that for the above years Howard Bard received the same percentage as Ms. Clark." Pl.'s Ex. 85 at D0106. With respect to actuarial assumptions, the letter referenced an increase in the interest rate assumption from 7% to 8% and stated that "Ms. Clark was integrally involved with making this decision." Id. at D0108. The letter concluded with a decision denying Clark's appeal and "reconfirm[ing] that the correct amount of Ms. Clark's lump sum benefit is $166,541.71." Id. at D0109.

ii. Bard's Grouping and Benefits

Reddington's calculation of Bard's benefits under the Plan has also been introduced into evidence and was apparently produced to the plaintiff during discovery in this case. See Pl.'s Ex. 66; Tr. 13:18-22. The worksheet indicates that Bard, like Clark, received a benefit accrual of 8% of his annual compensation for each of the four plan years ending on September 30, 1995, to September 30, 1998. Id. at D0735. In other words, Bard and Clark were both placed under "Group D" for the years prior to becoming equity partners. The worksheet indicates that Bard, like Clark, received a benefit accrual of 10% of his annual compensation for the next three years — that is, for plan years ending on September 30, 1999, to September 30, 2001. Id. Again, Bard and Clark were both placed in "Group C" for these years. But the worksheet indicates that Bard, unlike Clark, received a 20% benefit accrual for the plan year ending on September 30, 2002. Hence, Bard was placed in "Group B" for the fourth year and Clark was placed in "Group C," despite the fact that both Bard and Clark were Class A Shareholders during this year. In accordance with the 2003 restatement of the Plan, which explicitly placed Class A shareholders of appropriate age in Group B, Bard also received a 20% benefit accrual for the following plan year, which ended on September 30, 2003, after Clark had left the firm. Id.; see also Pl.'s Ex. 39 (actuarial worksheet indicating "all but Semo & Bard are class D (8%)" with annotations of "B" next to Semo and Bard's names).

iii. Decision on Clark's Grouping

The Court has been presented with a substantial amount of evidence regarding Clark's grouping under the Plan. This information appears to have all been considered by the firm when Clark made her formal appeal, with the exception of the fact that Bard was grouped at a higher accrual rate than Clark for one year during which both were at the Firm. This fact seems not to have been understood by defendants at the time of Clark's appeal.

On November 4, 2005, subsequent to Clark's appeal letter, Anspach emailed Semo to alert him to the ambiguity in the Third Amendment. Pl.'s Ex. 67; Tr. 896:18-897:12 (Anspach); Tr. 1073:18-22 (Semo). More specifically, the email stated: "The good news is that the amendment ties together our position that Denise received ... 10% for plan years ending 9/30/99, 9/30/00, 9/30/01 and 9/30/02. "The problem with the Third Amendment is that there appears to be a drafting error and the definitions of Class B and Class C are unclear." Pl.'s Ex. 67. The email also indicated that the Plan was restated in September 2003 "for the plan year beginning October 1, 2002" and that "[s]ince Denise was gone by that date, this restatement does not affect her." Id. Anspach's email also stated that he had received a fax from Reddington in August 2003, which in turn contained a memorandum from Bard to Reddington dated June 2000. Pl.'s Ex. 67; Tr. 897:13-18 (Anspach). The Bard memorandum states that, under the Third Amendment, "contributions [are] as follows:

Pl.'s Ex. 36 at PAG1145. The memorandum also indicates that it is being sent "in order to enable us to have final numbers by the end of this week, so that we may make our final contribution to the Plan by the June 15, 2000, due date." Id.11

Anspach sent further emails to Semo and Bard in the next few days on the topic of Clark's grouping. One of the emails, dated November 9, 2005, asked: "Since [Bard] and [Clark] became Class A shareholders on March 31, 2000, why does the June 6, 2000 fax memo state that [Clark] and [Bard] are members of Class C and entitled to 10% compensation?" Pl.'s Ex. 69; see Tr. 944:11-947:9 (Anspach). The email also indicated that "[e]xcept for the June 6, 2000 fax from [Bard], [Reddington] does not think that he has any written documentation regarding the groupings[;] [h]e believes that each year Jaci [Moline] verbally provided this information in telephone conversations." Pl.'s Ex. 69. Jacqueline Moline was the firm's office manager. Tr. 91:24-92:12 (Semo). At trial, Anspach confirmed this understanding of how Reddington received information about groupings from the Firm. Tr. 944:18-945:2 (Anspach). Reddington also confirmed this fact at trial. Tr. 603:1-5, 641:9-642:11 (Reddington).

Anspach, on behalf of the firm, asked Reddington to prepare calculations putting Bard and Clark in Group B — 20% accruals — for plan years 2000, 2001, and 2002. Tr. 901:12-17 (Anspach); Pl.'s Ex. 69. Reddington made these calculations and provided them to Anspach, who in turn provided them to Bard and Semo. Tr. 405:8-18 (Bard); Tr. 665:14-670:10 (Reddington); Tr. 901:7-902:15 (Anspach); Pl.'s Exs. 70-71.12 According to Reddington's calculations, placing Clark in the 20% category for these three years would increase her hypothetical account balance to $225,325.72 (as compared to $158,899.96 under the original grouping) and her unreduced lump sum amount to $440,428.98. Pl.'s Ex. 71. As a pro rata share of the Plan's assets, Clark would be entitled to $216,005.88 (as compared to $166,541.71 under the original grouping). Pl.'s Ex. 70. Hence, Clark would be entitled to an additional $49,464.17 in lump sum distribution. Id. Bard's unreduced lump sum would increase to $559,343.66 and his pro rata share would increase to $274,371.54. Pl.'s Exs. 70-71.

Semo and Bard eventually decided to deny Clark's appeal. See Tr. 405:23-406:6 (Bard); Tr. 564:13-18 (Semo); Tr. 912:22-913:18 (Anspach); Exs. 72-74. Bard and Semo considered Clark's appeal with the assistance of Anspach and Nielsen. Bard testified that he reviewed the BCA to see if there was any provision about himself and Clark and that he "did not see anything in there or anything in the [Third] amendment" about their placement. Tr. 410:18-411:20 (Bard). Bard also testified that it was his understanding that the second grouping was created for Semo when he joined the firm and that he and Clark were intended to be placed in the third grouping. Tr. 323:18-325:18 (Bard). Bard had no explanation for why Diana Peters — whose identity was not discussed further at trial — was also included in the second grouping. Tr. 324:13-325:11 (Bard). Semo testified several times that "[w]hatever doubts [he] had" about the appropriate classification were "clarified" by Bard, who stated against his financial interest that he and Clark were intended to stay in the third grouping. Tr. 548:22-25, 557:6-12, 1079:1-15, 1096:14-1097:12 (Semo). Semo indicated that this factor "weighed heavily" in the decision. Tr. 1079:15 (Semo). Additionally, Semo noted that the BCA provided that his own accrual percentage would be 20% and that no similar agreement existed for Clark and Bard. Tr. 548:3-22, 1076:20-1077:22 (Semo). Semo also testified that he "probably" would have received less money if the decision on the appeal had been to raise Clark's distribution. Tr. 549:10-13 (Semo).

Nielsen testified that Bard and Semo asked him to join the deliberation on Clark's appeal — despite the fact that he had already begun new employment elsewhere by that time — because he had benefit claims experience and "they wanted someone who had perhaps a more set of fresh eyes [sic] to go over the issues with them." Tr. 804:21-805:12 (Nielsen). Nielsen testified that Semo and Bard told him that the Plan was designed to put Clark and Bard in the 10% category and that he found it significant that this statement was against Bard's own financial interest. Tr. 810:20-812:4 (Nielsen). Nielsen testified that he did not object to denying the appeal because "[i]t did not sound like a particularly objectionable interpretation to me." Tr. 811:14-19 (Nielsen). Anspach testified that he concurred with this result, primarily because Clark had originally been grouped in the 10% category and, after reviewing the available information, the group did not "f[ind] anything different to change the answer" and so "made a decision to leave everything the same and leave it how it was." Tr. 900:14-906:13 (Anspach).

Anspach sent Clark the formal notice denying her appeal, which went through several drafts. The erroneous statement that "Howard Bard received the same percentage as Ms. Clark" during the years in question was apparently not in the original draft of the letter, but was added at some point during the editing process. Compare Pl.'s Ex. 85 at D0106, with Defs.' Ex. 1 at PR050. Anspach testified at trial that "[s]omeone at the firm" told him in November 2005 that Bard and Clark received the same accrual percentage, though he could not recall who at the firm told him. Tr. 959:12-960:2, 1004:4-9 (Anspach). Semo testified that he believed the statement to be true at the time of Clark's appeal and that he relied on Anspach and Reddington for the information. Tr. 550:4-550:16, 558:15-19, 1095:8-1096:10 (Semo). Bard testified similarly that he believed the statement to be true at that time and did not learn of his own more favorable grouping until his deposition in this case. Tr. 320:9-322:2, 417:11-418:1 (Bard).

iv. Distributions to the Feders

Following his retirement on December 31, 2001, Gerald Feder was eligible for a distribution, apparently because he had reached age 65.13 Clark oversaw this distribution with substantial advice from Anspach and Reddington. See Tr. 103:23-106:12, 107:7-108:12, 109:23-112:14 (Clark). Feder was not able to receive the full lump sum amount resulting from the GATT rate calculation because of certain restrictions on distributions to "highly compensated employees." See Tr. 61:3-20 (Clark); Tr. 641:5-8 (Reddington); Pl.'s Ex. 24. According to Reddington, IRS regulations "restrict[] benefits payments to highly compensated employees as a measure of protection so highly compensated employees do not terminate and leave a plan underfunded for nonhighly compensated employees. The Plan cannot be less than 110% funded on a current liability basis." Pl.'s Ex. 24. According to calculations provided by Reddington, Feder's full lump sum distribution amount as of April 2002 was $1,083,263, but he only was able to receive $779,082. See Pl.'s Exs. 22-26; Tr. 61:21-62:18 (Clark). Loretta Feder's distribution a few months later was similarly reduced. See Pl.'s Exs. 23, 28; Tr. 642:17-643:10 (Reddington).

Anspach testified at trial that Gerald Feder could not receive the remaining portion of his benefit until the Plan was funded to a certain percentage — Anspach "believe[d]" the percentage was 110% — on a current liability basis. See Tr. 1010:3-15 (Anspach). (The phrase "current liability" is explained more fully below.) According to Clark, she had a conversation with Semo in 2004 in which he indicated that the Plan was not funded in a manner that would allow Feder "to get the balance of his distribution out of the plan." Tr. 60:2-61:2 (Clark). Semo was not specifically asked about this conversation during his trial testimony, but did testify that he was not "concerned about Mr. Feder" in making decisions about the funding of the Plan. Tr. 1115:17-1116:9 (Semo).

Gerald Feder was also involved in discussions with Semo and Bard around the time that the firm lost its largest client in July 2005. According to Semo, in the period between when the firm lost the client and when the firm terminated, Semo and Bard spoke to Feder about the possibility of their forfeiting "about a third" of their distributions under the Plan in order to make full distributions to other employees such as Clark. Tr. 573:23-575:21 (Semo); see also Tr. 413:9-414:1 (Bard). Semo testified that "[i]t was only when it was realized that number had to be much bigger than a 30 percent haircut that Mr. Feder said, no, he wanted his benefit" and that this effectively foreclosed the option of making a full distribution to Clark. Tr. 575:7-24 (Semo).

e. Plan Funding

i. Plan Terms and Assumptions Regarding Funding

The Plan also contained provisions relating to funding. The Plan stated: "It is the intention of the Employer to continue the Plan and make regular contributions to the Trustee each year in such amounts as are necessary to maintain the plan on a sound actuarial basis and to meet minimum funding standards as prescribed by any applicable law." Pl.'s Exs. 1 & 3 § 4.1.

The Plan itself did not define what was meant by "maintain[ing] the plan on a sound actuarial basis." Reddington, the Plan's actuary, testified that the Plan's funding was based on "actuarial assumptions" that are "not hard-coded into the plan document." Tr. 607:21-608:1, 608:25-609:2 (Reddington); see also Tr. 223:4-13 (Poulin). As with other similar plans, the actuary performs calculations, based on actuarial assumptions, to inform the plan sponsor how much needs to be contributed to fund the Plan. Tr. 261:17-262:1 (Poulin). In other words, the actuarial assumptions dictated the minimum and maximum allowable contributions to the Plan. Two such assumptions are relevant here. First, the Plan was funded under the assumption that participants would take their benefits as an annuity at normal retirement age, rather than as a lump sum at an earlier time. Tr. 606:20-608:7 (Reddington). Reddington testified that "[t]hat was a very common feature of small plans, not to assume that participants take their distributions earlier." Tr. 608:7-9 (Reddington). Second, the Plan was funded under an assumption about the rate of return on the investment of its assets — the so-called "interest rate assumption." Tr. 608:10-21 (Reddington); see also Tr. 262:5-8 (Poulin); Pl.'s Ex. 106 at 26 (defining "Valuation Liability Interest Rate" as "the assumption as to the expected interest rate (investment return)"). This assumption is the basis of one of Clark's three claims in this case. As Clark's expert has explained, "if a high interest rate assumption is used, the contributions to the plan will be lower than if a lower interest assumption was used. By selecting an unreasonably high interest assumption, minimum funding requirements can be decreased to lower levels." Pl.'s Ex. 100 ¶ 4. And as the Court has previously noted, the interest rate assumption "mattered because to the extent the market interest rate decreases relative to the Plan's projected interest rate, the present cash value of a beneficiary's annuity increases. And if the present value of a beneficiary's annuity increases, but the Plan's funding remains constant, there is a potential for the Plan to have unfunded liabilities." Clark III, 736 F.Supp.2d at 232 (citing Hirt v. Equitable Ret. Plan for Emps., Managers, & Agents, 533 F.3d 102, 109 (2d Cir.2008)).

ii. Interest Rate Assumption

The interest rate assumption was 7% prior to the fall of 2001, but was then raised to 8% in 2002, after Clark became managing partner of the firm. At that time, there was some communication between Clark and Reddington about the interest rate. The testimony at trial regarding this communication was inconsistent. Reddington testified that Clark indicated to him that the firm's minimum required contribution to the Plan was "larger than what they wanted to contribute" using the 7% interest rate assumption, so he performed calculations and informed Clark what the minimum contribution would be using an 8% interest rate. Tr. 609:8-610:1 (Reddington). Clark testified that an inquiry about the interest rate came from Reddington to her and that she understood that Feder and Semo had previously agreed to raise the rate and had communicated that to Reddington prior to her becoming managing partner. Tr. 47:11-48:8, 162:2-9 (Clark). Semo testified that Clark made a recommendation to the board of directors based on discussions she had had with Reddington, that the board of directors approved the change, and that he never had any other conversations about the interest rate, including with Feder. Tr. 523:22-524:15 (Semo). Bard testified that he remembered Clark presenting the issue at a board meeting, but that he couldn't recall who was managing partner when the issue was first raised. Tr. 352:18-353:1 (Bard). Both Reddington and Clark testified that at some point Clark informed Reddington to go forward with raising the interest rate assumption from 7% to 8%. Tr. 162:7-9 (Clark); Tr. 609:20-610:1 (Reddington).

iii. Funding of the Plan Over Time

Reddington provided Actuarial Valuation Reports to the firm with various statistics about the Plan for each plan year during the time period in question. The "accrued benefits" statistic was the sum total of the Plan participants' hypothetical account balances — essentially, the value of benefits if participants took their benefit at normal retirement age. See Tr. 249:15-251:2 (Poulin). The Plan's "current liability" and "funded current liability percentage" statistics, however, were indicators of the Plan's liabilities on a lump sum distribution basis and how well-funded the Plan was on that basis, because "current liability" is calculated based on a rate similar to the GATT rate. See Tr. 248:16-249:4 (Poulin); 726:21-727:13 (Altman). Nonetheless, the current liability statistic did not exactly reflect the Plan's liabilities upon termination. See Tr. 679:7-680:2 (Reddington); 726:21-25, 758:13-15 (Altman). The interest rate assumption affected the "minimum required contribution" amount — the amount of money Feder Semo was required to contribute to the Plan each year — as well as what the reports referred to as "Accrued Liability." See Pl.'s Ex. 17 at D0407; Tr. 233:4-234:8 (Poulin). The firm appears to generally have made the "minimum required contribution" each year, although there was some dispute at trial over whether Reddington correctly calculated the contribution in the final year of the Plan (i.e., at termination). See Tr. 306:9-15 (Poulin); Tr. 876:15-22 (Anspach).

In January 2003, Reddington provided an Actuarial Valuation Report to the firm for the plan year ending September 30, 2002. Pl.'s Ex. 17; Tr. 503:16-24 (Semo). The report indicated that the Plan's "[f]unded current liability percentage for current year" was 55.76% and that the Plan had accrued benefits with a present value of $1,272,414 and assets with a market value of $1,048.576. Pl.'s Ex. 17 at D0391, D0405. This was a significant drop in funding from the prior year, primarily due to the distributions to the Feders. The report indicated that the minimum required contribution from the firm to the Plan that year was $150,276. Id. at D0391.

On March 3, 2003, Reddington emailed the firm about the Plan, and, in particular, the Plan's funding. Pl.'s Ex. 34; Tr. 616:25-617:13 (Reddington). The email stated: "Gerald and Loretta Feder have now retired. The Plan was set-up to maximize the Feder's [sic] benefits. The Plan design needs to be reviewed to best advantage the current owners." Pl.'s Ex. 34 at 1. The email indicated that, as of September 30, 2002, the Plan had liabilities in the amount of $1,216,000 and assets in the amount of $1,002,000 and that therefore the Plan had an "Underfunded Status" of "[l]iabilities exceed[ing] assets by $214,000." Id. Under the heading "How does the Plan become sufficiently funded?", the email listed three options: "Larger Contributions in 2002/03 and future years ($250,000 instead of the $157,000 for 2001/02)," "Contributions similar to the 2001/02 contribution and good asset performance," and "Lower contributions ($75,000 to $100,000) and freezing of accruals." Id. The email explained that a participant's lump sum distribution is based upon the greater of the cash balance account and the calculation involving the GATT rate and stated that GATT rates "are currently at a historical low point." Id. at 3. The email stated that "[t]herefore, on a plan termination basis, the Plan is even more significantly underfunded." Id.

After receiving Reddington's email, the Firm's executive committee met and considered the funding issue. Documents in the record suggest that the firm first decided not to take any action to address the funding issue and that Anspach expressed his view that this path did not make sense, although at trial Anspach, for one, did not recall this fact. See Pl.'s Ex. 35; Tr. 845:13-856:13 (Anspach). In any event, the committee then resolved to address the funding issues by freezing benefit accruals and "putting in additional monies over the next several years." Tr. 1067:3-1068:8, 1071:5-10 (Semo); see also Tr. 360:12-20 (Bard); Tr. 616:16-620:3 (Reddington); Tr. 851:8-854:12 (Anspach). Semo testified that "in his mind" the firm would make additional contributions to the Plan over a period of three to five years to remedy the "ongoing" underfunding of approximately $200,000 and the additional underfunding on a "termination basis." Tr. 509:12-511:14, 533:21-535:7 (Semo). Semo indicated that he believed the firm would have "extra funds" available as the firm's revenue-sharing agreement with Feder wound down. Tr. 497:8-16, 509:23-510:2 (Semo). He also testified that the firm was not intending to contribute more funds to address the possibility of immediate lump sum distributions because "we weren't thinking of terminating the plan." Tr. 535:8-9 (Semo). It does not appear that anyone ever suggested or considered changing the interest rate assumption. See Tr. 352:8-15, 353:9-18 (Bard); Tr. 524:20-25, 527:19-528:16 (Semo).

As indicated, the Plan was amended to freeze accruals as of September 2003. Tr. 855:25-856:4 (Anspach). The Actuarial Valuation Report for plan year ending September 30, 2003, which was certified in July 2004, indicated that the Plan's "[f]unded current liability percentage for current year" was 49.25%. Pl.'s Ex. 18 at D0313. In November 2004, Reddington emailed the firm. Reddington indicated that "[t]he minimum required contribution for the plan year end[ing] September 30, 2004 is $22,404" and that this amount "is much lower than in the past because of the freeze in benefits that took place at the end of the previous plan year." Pl.'s Ex. 42. at 1. Reddington also stated that "[t]he plan is still very underfunded on a plan termination basis" and that "[w]e would strongly recommend that the final contribution for the plan year exceed the minimum funding requirement." Id.

For the plan year ending September 30, 2004, the firm contributed $122,404 to the Plan in installments from February 2005 to May 2005. Pl.'s Ex. 13 at D0239; Tr. 632:5-633:21 (Reddington). Semo and Anspach testified that they believed that this was approximately $100,000 more than the minimum required amount. Tr. 858:23-859:2 (Anspach); Tr. 1069:4-1070:13 (Semo). The Actuarial Valuation Report for plan year ending September 30, 2004, which was dated July 11, 2005, indicated that the minimum required contribution was $62,263 — apparently contradicting Reddington's email. Pl.'s Ex. 19 at D0222; see Tr. 651:13-24 (Reddington).

For plan year ending September 30, 2005, the firm made its final contribution to the Plan in the amount of the minimum contribution, $4,774, in one installment made in December 2005. Pl.'s Ex. 14 at D0161; Tr. 653:18-654:20 (Reddington). Reddington testified that this minimum contribution was approximately $58,000 less than the minimum contribution would have otherwise been, because the contribution for the September 30, 2004, plan year was included as a "[p]rior year credit balance." Tr. 654:10-20 (Reddington); see Pl.'s Ex. 14 at D0163. According to the Actuarial Valuation Report for the plan year ending September 30, 2005, which was dated June 30, 2006, the Plan's "[f]unded current liability percentage" was 79.1%. Pl.'s Ex. 20 at D0149; Tr. 620:22-621:11 (Reddington). At that time, the report indicated that the present value of accrued benefits was $1,675,517 and the value of the Plan's assets was $1,392,275. Pl.'s Ex. 20 at D0135; Tr. 624:9-18 (Reddington).14 Somewhat inexplicably, Reddington appears to have included Bard and Clark's hypothetical account balances in the Actuarial Valuation Report as if Clark's appeal had been resolved in both employees' favor — that is, as if they had been in the 20% accrual grouping for the three years in question. Compare Pl.'s Ex. 20 at D0150, with Pl.'s Ex. 71; see Pl.'s Ex. 96 at 10-11; Tr. 673:16-23 (Reddington). As a result, the present value of the accrued benefits that were actually paid out must have actually been somewhat lower than $1,675,517. As indicated, each plan participant received a pro rata share of 53.31% of their benefit as provided for by the lump sum calculation. Pl.'s Ex. 56; Tr. 679:20-23 (Reddington).

f. Expert Testimony on Reasonableness of Interest Rate Assumption

The Court heard testimony from two expert witnesses regarding the reasonableness of the 8% interest rate assumption.15 Claude Poulin was admitted as an expert witness on behalf of Clark to provide testimony on the general subject of the design, interpretation, administration, and review and compliance of defined benefit pension plans, including offering opinions as to the reasonableness of certain actuarial assumptions. Tr. 179:11-16 (Poulin). Poulin testified that he believed the 8% interest rate assumption was unreasonable. Tr. 232:8-11 (Poulin). He stated that, under ERISA, "actuarial assumptions must individually and in the aggregate ... tak[e] into account the experience under the plan as well as reasonable expectations as to the future experience under the plan." Tr. 231:20-24 (Poulin). Poulin testified that the interest rate assumption did not take into account the features of the Plan that participants could take their distributions as lump sums, rather than receiving a benefit at retirement age, and that lump sum distributions were based on GATT rates if that rate produced a larger lump sum. Tr. 220:18-221:7, 229:21-230:18, 285:24-286:1 (Poulin). He testified that the liabilities of the Plan were computed assuming the participants would retire at age 65, when in practice, for a plan of this type, it would be reasonable to expect that participants will take a lump sum distribution when it becomes available. Tr. 226:13-227:13, 232:4-7, 289:22-290:5, 300:6-10 (Poulin). Poulin testified that the "experience of the plan" was to make lump sum distributions; he noted that the Feders both took their distributions as lump sums, although he also noted that he was not aware of anyone else taking lump sum distributions prior to the Plan's termination. Tr. 221:8-18, 229:16-24, 289:10-17, 300:20-301:5 (Poulin). Poulin also indicated that benefit distributions generally do not "come due at the same time" — i.e., all at once — in pension plans. Tr. 287:15-17 (Poulin).

Poulin also testified that an 8% assumption might have been reasonable if the Plan had "offsetting" actuarial assumptions, but the Plan did not have any such assumptions. Tr. 230:24-231:8 (Poulin). He stated that his opinion about the reasonableness of the 8% assumption was based on the fact that the Plan's terms indicated that the Plan "had to be fully funded." Tr. 230:19-23, 286:14-287:3 (Poulin). Poulin testified that he believed 6% would have been a reasonable interest rate assumption. Tr. 232:12-233:3 (Poulin). He testified that had the Plan utilized a 6% assumption, the minimum required contribution would have been approximately $400,000 greater in total for plan years 2002-2004; had the Plan utilized a 7% assumption, the contribution would have been approximately $275,000 greater; and had the Plan utilized a 5.38% assumption (the average GATT rate for this period), the contribution would have been approximately $500,000 greater. Tr. 236:1-237:20 (Poulin). Poulin stated that merely freezing the Plan would not remedy the underfunding issue because the "whipsaw" effect was a feature of the Plan. Tr. 228:16-229:9 (Poulin). Poulin testified that the whipsaw effect was a common feature of similar plans during this time period. Tr. 282:11-284:6 (Poulin).

Ian Altman was admitted as an expert witness on behalf of the defendants to provide testimony on the design, interpretation, administration, and review and compliance of defined benefit pension plans. Tr. 711:13-17 (Altman). Altman testified that an 8% interest rate assumption was reasonable in this case. Tr. 721:1-7 (Altman). He testified that he believed that the rate was reasonable because the Plan's returns had been in excess of 8% a year for the years in question and the Plan had a diversified portfolio of stocks and bonds. Tr. 721:8-722:20, 733:13-734:3, 779:2-780:11, 786:18-21 (Altman). Altman also indicated that market surveys from the time period indicated an average funding interest rate assumption of about 8%. Tr. 721:13-723:20 (Altman). He noted that one such market survey included plans with a mean salary increase assumption of 5.1%, while the Plan here did not use a salary increase assumption because it was a cash balance plan that did not rely on salary assumptions. Tr. 740:17-742:1, 783:14-784:14 (Altman). Altman testified that neither the distribution to the Feders nor the Plan's whipsaw feature affected his view of the reasonableness of the interest rate assumption. Tr. 723:21-725:10 (Altman). He testified that neither of these facts affected the long-term expectation for investment returns, which is what the interest rate assumption is intended to reflect. Id.; see also Tr. 734:17-21 (Altman). He also noted that the Feders took their retirement at age 65, making the GATT rate less significant with respect to them even though the distributions were taken as lump sums. Tr. 723:21-725:10 (Altman); see also Tr. 749:18-750:2, 753:10-18 (Altman). Altman also indicated that he understood that actuarial assumptions must be reasonable in the aggregate but that this did not change his impression of the reasonableness of the interest rate assumption. Tr. 734:4-22, 745:2-12, 777:16-25 (Altman). He noted in particular that the Plan used a conservative mortality table, which in his view was an off-setting assumption. Tr. 781:9-20 (Altman).

Altman testified that he would expect that the funding of the Plan would have improved over time, since the accrued benefits were frozen and the firm was continuing to contribute to the Plan. Tr. 730:19-731:4 (Altman). He stated that the Plan sponsor took steps to mitigate the underfunding of the Plan "but it was the termination of the plan tied to the fortunes of the firm that led to the plan terminating at a time when there was underfunding." Tr. 731:18-23 (Altman). He also testified several times that the degree of underfunding of a Plan on a termination basis does not affect his assessment of the reasonableness of the actuarial assumptions if he expects the Plan to be ongoing and that an actuary should only consider underfunding on a termination basis "[w]hen it became either apparent or likely that termination was coming." Tr. 736:1-18, 769:7-14, 778:19-779:1, 784:15-18 (Altman). Altman indicated that if he were the Plan actuary, he would "have some cognizance" of the fact that terminated participants were eligible to receive lump sum distributions in five years, but noted that he would be uncertain about what the GATT rates would be at that time. Tr. 746:13-747:1 (Altman). Finally, Altman testified that, as a rule, he doesn't expect his clients to understand actuarial assumptions. Tr. 786:3-17 (Altman).

g. Summary Plan Description and PBGC Insurance

Clark received a copy of the Feder Semo Summary Plan Description ("SPD"), originally dated September 1994, from Gerald Feder. Tr. 64:6-9 (Clark); Pl.'s Ex. 4 at 9. The SPD, which was drafted by Anspach, was also made available in the office kitchen. Tr. 64:8-9 (Clark); Tr. 961:13-19 (Anspach). Among other things, the SPD described the valuation of Plan benefits as an annuity and described the rules for receiving lump sum distributions. See Pl.'s Ex. 4 at 2-3. The SPD indicated that "[a]s a general rule, if a distribution is made in a single sum payment, the amount of your retirement benefit will equal your hypothetical account balance" and noted that "[d]ue to the application of actuarial equivalents, in certain circumstances, the single sum payment of your retirement benefit may exceed your hypothetical account balance." Id. at 2. Under the heading "Can the Plan be Amended or Terminated?", the SPD stated: "Yes. The Plan can be amended and terminated. However, no amendment may deprive you of the vested amount of your accrued benefit. If the Plan terminates, you will become 100% vested in your accrued benefit and your accrued benefit will be distributed to you in accordance with the terms of the Plan." Id. at 6-7. Under the heading "Is This Plan Covered By the Pension Benefit Guaranty Corporation ("PBGC")," the SPD stated: "Benefits under this Plan are insured by the PBGC if this Plan has more than 25 participants. If PBGC covers this Plan and, on termination, if the Plan does not have sufficient assets to pay the benefit, the PBGC will provide part or all of that benefit. The amount of benefit guaranteed is subject to certain limitations." Id. at 7.

The SPD was reissued in September 2003, after Clark's departure from the firm. See Pl.'s Ex. 5 at P0234; Tr. 578:25-579:6 (Semo). Clark testified that she did not see or receive the SPD after leaving the firm until she requested a copy in September 2005. Tr. 145:9-10 (Clark).

The Plan did not apply to PBGC for insurance coverage. Tr. 363:3-8 (Bard); Tr. 966:8-12 (Anspach). Anspach testified that under the ERISA statute, a professional service organization's plan is covered by PBGC insurance if it has more than 25 "active participants." Tr. 931:2-9 (Anspach); see 29 U.S.C. § 1321(c)(3). Anspach testified that he put the language that "[b]enefits under this Plan are insured by the PBGC if this Plan has more than 25 participants" into the SPD because "I believe at the time I was doing this I knew they were under the [25 participant] threshold, but ... it's not the type of a client that I would necessarily get information if they go over the threshold[,] [s]o I put in the rule because if they have 15 and they go over 25 ... then they would become covered at that point in time." Tr. 931:1-15, 964:11-966:1 (Anspach). Anspach testified that his inclusion of this conditional statement in the SPD was unusual and that he "almost always ... would make the definitive statement." Tr. 966:2-8 (Anspach). Anspach also testified that the "correct sentence" would have included the word "`active' between the words '25' [and] `participants.'" Tr. 964:11-19 (Anspach).

A Plan filing with the IRS in 2002 indicated that "[a] total of 31 persons were participants in or beneficiaries of the plan at the end of the plan year." Pl.'s Ex. 9 at P0428. After Clark filed her appeal, Anspach sought to verify with the PBGC that only plans with more than 25 participants currently accruing benefits were covered by PBGC insurance. Tr. 975:2-976:25 (Anspach). The PBGC indicated that its "informal position" — apparently, the consensus of attorneys in the general counsel's office — was that "active participants" refers "to currently working employees" but that "they do not have any opinion letters on this issue." Id.; Pl.'s Ex. 63 at 1.

Clark testified that, prior to September 2005, she did not know whether the Plan was insured by the PBGC. Tr. 59:16-19 (Clark). Clark stated that she "relied" on the language in the SPD regarding PBGC insurance in September 2005 "with regards to whether or not there was any backup to what the firm wasn't able to pay.... What I was trying to figure out was whether there would be any backup should the firm not be able to pay anything." Tr. 144:5-11 (Clark). Clark also testified that she "[p]robably [did] not" rely on the SPD language regarding PBGC insurance before September 2005 "because prior to September 2, I had no reason to think that the firm was not going to pay the benefits that were due to each and every participant in the plan as stated under the last set of benefit statements we had received." Tr. 144:24-145:4 (Clark). Clark also stated that she was aware PBGC insurance premiums were not in the firm's budget when she was a shareholder in the firm. Tr. 138:21-141:20 (Clark). Bard testified that he did not know prior to September 2005 whether the Plan was covered by PBGC insurance, that he was aware the firm was not paying insurance premiums to PBGC, and that he did not know whether PBGC's insurance coverage is dependent on "being current on your premium payments." Tr. 363:9-364:11 (Bard).

III. Conclusions of Law

The Court reviews the evidence under the "default rule for civil cases," the "preponderance of the evidence" standard. CIGNA Corp. v. Amara, ___ U.S. ___, 131 S.Ct. 1866, 1881, 179 L.Ed.2d 843 (2011). The Court considers each of Clark's three claims — the improper grouping claim, the summary plan description claim, and the interest rate assumption claim — in turn.

a. Improper Grouping Claim

Clark contends that she was improperly classified under the Plan in the 10% benefit accrual category (Group C) for plan years 2000, 2001, and 2002, and that she is entitled to the additional benefits she would have received with 20% allocations for those years, plus interest. Pl.'s Statement at 4. Clark brings this claim under 29 U.S.C. § 1132(a)(3) for breach of fiduciary duty against Semo and Bard, the "fiduciaries who decided not to correct her benefit before distributing the Plan's assets," to the extent that "monetary recovery for that violation is unavailable because the Plan's assets have been distributed." Pl.'s Statement at 3.16

i. Legal Framework

The Court previously denied summary judgment on the administrative record to Bard and Semo on the breach of fiduciary duty claim. The Court found that "Clark has demonstrated that defendants Semo and Bard were aware of her grouping in a less advantageous category and failed to provide a reasonable explanation for why she was so classified and why her benefits were not adjusted prior to the disbursement of Plan assets upon its termination." Clark IV, 808 F.Supp.2d at 226. The Court has also previously stated that "[a]tlhough the Court agrees with Feder Semo that the 1998 amendment contained an ambiguity that permitted the firm to classify Clark in either Group B or Group C, the mere presence of an ambiguity is not a `final, fully considered, and reasoned explanation for the court to evaluate.'" Clark II, 697 F.Supp.2d at 32 (quoting Hall v. Nat'l R.R. Passenger Corp., 559 F.Supp.2d 38, 48 (D.D.C.2008)). The Court noted that Anspach "raised the question animating Clark's improper grouping theory during her benefit appeal, but there is no evidence that Feder Semo resolved it either during the appeal or earlier." Id. Additionally, the Court indicated that it could not, on the record before it, "conclude that the firm's decision not to retroactively apply the 2003 restatement of the 1998 amendment to Clark was reasonable" and that the evidence that Bard was placed in the 20% grouping for the 2002 plan year was "evidence in the administrative record to suggest that the firm should have corrected [Clark's] account credit pursuant to the 2003 restatement for at least the 2002 plan year." Id. at 32-33. With respect to retroactively applying the 2003 restatement, the Court noted Anspach's statement that "[s]ince Denise was gone by [the restatement's effective date], this restatement does not affect her," Pl.'s Ex. 67, but concluded that "a statement of the Plan's outside attorney does not evidence the reasons why Feder Semo adopted this position." Id. at 32.

Under 29 U.S.C. § 1132(a)(3) (or "ERISA § 502(a)(3)"), "[a] civil action may be brought ... by a participant, beneficiary, or fiduciary (A) to enjoin any act or practice which violates any provision of this subchapter or the terms of the plan, or (B) to obtain other appropriate equitable relief (i) to redress such violations or (ii) to enforce any provisions of this subchapter or the terms of the plan." In Varity Corp. v. Howe, 516 U.S. 489, 491-92, 116 S.Ct. 1065, 134 L.Ed.2d 130 (1996), the Supreme Court concluded that § 1132(a)(3) authorizes individual ERISA plan beneficiaries to bring a lawsuit for harm due to a breach of fiduciary duty. And in CIGNA, 131 S.Ct. at 1878-80, the Supreme Court ruled that beneficiaries may bring suits for monetary relief against plan fiduciaries for breach of fiduciary duty under § 1132(a)(3). See Clark IV, 808 F.Supp.2d at 224-26 (relying on Varity Corp. and CIGNA); see also CIGNA, 131 S.Ct. at 1879 ("The case before us concerns a suit by a beneficiary against a plan fiduciary (whom ERISA typically treats as a trustee) about the terms of a plan (which ERISA typically treats as a trust)."). ERISA provides that "a person is a fiduciary with respect to a plan to the extent... he exercises any discretionary authority or discretionary control respecting management of such plan or exercises any authority or control respecting management or disposition of its assets." 29 U.S.C. § 1002(21)(A)(iii); see Varity Corp., 516 U.S. at 527, 116 S.Ct. 1065. Bard and Semo have not denied that they fall into this category.

In Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989), the Supreme Court indicated that a court reviews a claim under 29 U.S.C. § 1132(a)(1)(B) under a deferential standard of review when "the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan." See Clark II, 697 F.Supp.2d at 31. It is undisputed that such discretionary authority existed here under the terms of the Plan. See Clark II, 697 F.Supp.2d at 31 n. 3. Furthermore, the Supreme Court stated in Varity Corp. that "characterizing a denial of benefits as a breach of fiduciary duty does not necessarily change the standard a court would apply when reviewing the administrator's decision to deny benefits" because "Firestone, which authorized deferential court review when the plan itself gives the administrator discretionary authority, based its decision upon the same common-law trust doctrines that govern standards of fiduciary conduct." 516 U.S. at 514-15, 116 S.Ct. 1065.

Hence, the Court concludes that the Firestone standard applies here to Clark's breach of fiduciary duty claim against Semo and Bard. The D.C. Circuit has defined this standard as one of "reasonableness." Wagener v. SBC Pension Benefit Plan-Non Bargained Program, 407 F.3d 395, 402 (D.C.Cir.2005) (citing Block v. Pitney Bowes, Inc., 952 F.2d 1450, 1452, 1454 (D.C.Cir.1992)); accord Moore v. CapitalCare, Inc., 461 F.3d 1, 11 (D.C.Cir.2006). The standard is deferential and "sometimes indicated by the term `arbitrary and capricious review,' other times signaled by the phrase `abuse of discretion.'" Block, 952 F.2d at 1454 (quoting Firestone, 489 U.S. at 109-11, 113-15, 109 S.Ct. 948). The parties agree. See Pl.'s Proposed Findings of Fact and Conclusion of Law [Docket Entry 128] ("Pl.'s F & C") at 48 (citing Block); Defs.' Resp. to Pl.'s Proposed Findings of Fact and Conclusions of Law [Docket Entry 131] ("Defs.' Resp.") at 96. The parties also agree that "[a] plan administrator must administer the provisions of a policy `consistently.'" Canada Life Assurance Co. v. Estate of Lebowitz, 185 F.3d 231, 238 (4th Cir.1999) (quoting de Nobel v. Vitro Corp., 885 F.2d 1180, 1188 (4th Cir.1989)).

So far, so good. But disagreements have emerged between Clark and Semo and Bard regarding the distinction between bringing a claim against the Plan administrator and bringing a claim against other Plan fiduciaries. The more straightforward of these disputes involves Clark's assertion of a conflict of interest. Clark asserts simply that, "to the extent an abuse of discretion standard applies, any conflicts of interests must be weighed." Pl.'s F & C at 48-49. Clark relies on Metropolitan Life Insurance Co. v. Glenn, 554 U.S. 105, 112, 128 S.Ct. 2343, 171 L.Ed.2d 299 (2008), in which the Supreme Court held that a conflict of interest is present when "a plan administrator both evaluates claims for benefits and pays benefits claims." The Court rejected the notion that "a change in the standard of review, say, from deferential to de novo review" is warranted under such circumstances. Id. at 115-16, 128 S.Ct. 2343. Rather, the Court simply concluded that "a conflict should `be weighed as a factor in determining whether these is an abuse of discretion'" and that judges should "determine lawfulness by taking account of several different, often case-specific, factors, reaching a result by weighing all together." Metro. Life, 554 U.S. at 115, 117, 128 S.Ct. 2343 (quoting Firestone, 489 U.S. at 115, 109 S.Ct. 948).17 Bard and Semo contend that "[t]he conflict of interest analysis is inapplicable" to them because Bard and Semo did not "individually serve as both the plan administrator and the source of plan funds." Defs.' Resp. at 98.

The Court agrees with Clark on this issue. The Court sees no reason under Metropolitan Life to distinguish completely between plan administrators and other plan fiduciaries. It is readily apparent that a plan fiduciary could have a conflict of interest in making decisions about benefit claims even if he or she is not the source of funds for the plan. The evidence in this case presents such a situation: as Semo testified at trial, he would likely himself receive a larger pro rata share of his lump sum distribution if less of the Plan's limited funds were distributed to Clark (as would all other Plan participants). See Tr. 549:10-13 (Semo). Bard's situation was more complicated, since he too would benefit from denying Clark's appeal by receiving a larger pro rata share, but would also have a more viable claim for his own more favorable grouping if Clark's appeal were granted. Considering not just whether a fiduciary was conflicted but how and to what degree seems exactly the type of "weighing" analysis that Metropolitan Life contemplated. It may well be true that the nature of the conflict is different for plan fiduciaries than for plan administrators, but that is simply a consideration to be weighed by the Court in its assessment under the abuse of discretion standard.

Another area of disagreement between the parties is the Court's consideration of evidence outside the administrative record. On this topic, the Court previously stated:

Clark II, 697 F.Supp.2d at 31. Clark now contends that "the abuse of discretion standard extends, moreover, only to `the specific basis upon which the Plan administrator relied in its administrative denial of benefits.'" Pl.'s F & C at 48 (quoting Spradley v. Owens-Ill. Hourly Emps. Welfare Benefit Plan, 686 F.3d 1135, 1140-42 (10th Cir.2012)). Bard and Semo maintain, by contrast, that "the Court should consider evidence used for the denial of benefits that is not included in the administrative record" because "[u]nlike the Spradley case, this case concerns a claim of breach of fiduciary duty that must be reviewed following a trial on the merits." Defs.' Resp. at 97. The main issue here, of course, is the erroneous statement in the letter denying Clark's appeal that "for the above years Howard Bard received the same percentage as Ms. Clark." Pl.'s Ex. 85 at D0106. Clark believes that this statement makes the reasons given in the letter "untrue," making the de novo standard apply instead. Pl.'s F & C at 50.

There is some force to Clark's argument, but the Court believes that it must review the actions of Semo and Bard largely under the deferential standard by which fiduciary action is typically judged, with appropriate consideration given to which evidence appeared in the administrative record in evaluating the weight of evidence. The Court certainly still believes that, in order for the Court to determine if the decision on Clark's appeal was reasonable, "`it is important for the plan to provide a final, fully considered, and reasoned explanation for the court to evaluate.'" Hall v. Nat'l R.R. Passenger Corp., 559 F.Supp.2d 38, 48 (D.D.C.2008) (quoting Commc'ns Workers of Am. v. AT & T, 40 F.3d 426, 433 (D.C.Cir.1994)). Furthermore, the Court still must, as the D.C. Circuit has instructed, "review ERISA-plan benefit decisions on the evidence presented to the plan administrators, not on a record later made in another forum." Block, 952 F.2d at 1455. The Court does not believe that a completely different standard applies for fiduciaries, as the defendants would have it.

Nonetheless, in this case, the Plan did articulate an explanation for its decision: it believed that the appropriate accrual percentage for Clark for the relevant years was 10%, in particular because Bard was grouped that way for those years. The Court is therefore not in the position faced in Communications Workers of America, where "the District Court had nothing before it, except the arguments of counsel, to which it could defer." 40 F.3d at 433. Furthermore, Semo and Bard are not, as in Spradley, trying to rely on some other basis for the reasonableness of their decision. See Spradley, 686 F.3d at 1141 ("The specific reasons and specific provisions supporting Defendant's broad coverage argument have changed, and we will not permit Defendant to sandbag Plaintiff with its after-the-fact interpretation of an entirely different section of the Plan."). Rather, they are defending the reasonableness of the terms of the decision they made at the time and articulated to Clark.

In reviewing Bard and Semo's behavior in coming to the decision that was articulated to Clark, then, the Court will be, as stated in Block, restricted to "the evidence presented to the plan administrators." 952 F.2d at 1455. But evidence admitted at trial about how the decision-makers in fact considered Clark's appeal — at the time they were considering it — falls into this category. To conclude otherwise would force fiduciaries defending themselves from these claims to rely only on documentary evidence that literally could be put into an "administrative record," rather than an evidentiary record of how they actually went about making the decision. Relatedly, although the same "standard of care" applies for plan fiduciaries as for plan administrators, it does not seem entirely fair to apply the same formal requirements for consideration of plan action to consideration of fiduciary action; what must be assessed is whether the fiduciary fulfilled his or her duty appropriately, not necessarily the administration of the plan writ large, for which the fiduciary may have only been partially responsible. In other words, the Court must consider the role that the fiduciary played, even though it applies the same standard of care to every actor. This situation is far removed from Block, in which the plaintiff argued that a separate entity's decision not known by the plan administrator was relevant to the review of the decision. See 952 F.2d at 1455-56. In considering the decision that was made in the full light of the trial evidence, the Court will certainly not consider reasons other than those on which the defendants actually relied when they were considering Clark's appeal at the end of 2005.

In sum, as with conflicts of interest, the Court believes it must consider all factors, giving each factor the weight that it deserves, including whether it appeared in writing at the time the decision was being made. This process will best fulfill what the Supreme Court has referred to as the "combination-of-factors method" for judicial review of fiduciary conduct. See Metro. Life, 554 U.S. at 119, 128 S.Ct. 2343. And the Court believes this standard is appropriate for what is ultimately an equitable determination.

Finally, the parties also disagree about whether Clark must show willful or bad faith conduct to establish the defendants' breach of fiduciary duty. For this proposition, Semo and Bard rely on, among other cases, Boivin v. U.S. Airways, No. 03-2373, 2005 WL 713622 (D.D.C. March 17, 2005);18 Morgan v. Independent Drivers Ass'n Pension Plan, 975 F.2d 1467 (10th Cir.1992); and Burke v. Latrobe Steel Co., 775 F.2d 88 (3d Cir.1985). Clark, in response, cites, among other cases, In re Mailman Steam Carpet Cleaning Corp., 196 F.3d 1 (1st Cir.1999); and Leckey v. Stefano, 501 F.3d 212 (3rd Cir.2007).

The Court's review of the cases cited by both parties reveals that the legal standard is quite clear and not subject to much, if any, disagreement: "[W]hen a trustee exercises his authority, a mere mistake will not render him liable for a loss. Only fault — in the form of bad faith or negligence — will. When, on the other hand, a trustee takes action that exceeds his authority, he is strictly liable for any loss (and accountable for any profit)." Leckey, 501 F.3d at 224 (citing Restatement (2d) of Trusts § 201 cmts. a-c (1959)); Morgan, 975 F.2d at 1470 (quoting identical language); In re Mailman Steam Carpet, 196 F.3d at 7 ("[Surcharge] is most fittingly defined as `[t]he imposition of personal liability on a fiduciary for wil[l]ful or negligent misconduct in the administration of his fiduciary duties.'" (second alteration in original) (quoting Black's Law Dictionary 1441 (6th ed.1990))). Leckey explicitly noted that it was clarifying the discussion in Burke. Leckey, 501 F.3d at 223. And the brief statement in Boivin on the topic does not unequivocally state that bad faith is required (as opposed to negligence) or otherwise warrant deviation from this clear rule.

As far as the Court can tell, there is no allegation here that Semo and Bard exceeded their authority as fiduciaries. Hence, the Court will apply the following standard: a fiduciary breaches his duty if he behaves either negligently or in bad faith (willful misconduct). The Court also notes that a straightforward rendition of the negligence standard appears in both the Plan itself and the ERISA statute. See Exs. 1 & 3 § 11.1 ("Each [f]iduciary shall act with the care, skill, prudence and diligence under the circumstances that a prudent man, acting in a like capacity and familiar with such matters, would use in conducting an enterprise of like character and with like aims."); Katsaros v. Cody, 744 F.2d 270, 279 (2d Cir.1984) ("ERISA requires a pension fund fiduciary to act `solely in the interest' of a plan's participants and beneficiaries, and to discharge his duties `with the care, skill, prudence, and diligence ... that a prudent man acting in a like capacity and familiar with such matters would use in the conduct of an enterprise of a like character ....'" (quoting 29 U.S.C. § 1104(a)(1)(B))).

ii. Application to Present Facts

Clark's contention that she was improperly grouped under the Plan is not frivolous. In the Court's view, Plan fiduciaries reviewing Clark's 2005 appeal, possessed of all the facts, may well have reasonably concluded that Clark should have been grouped in the 20% category for plan years 2000, 2001, 2002, or at least for the 2002 year. Yet the Plan was ambiguous, and Semo and Bard undertook a reasonable process and made a reasonable decision about Clark's grouping. The process was not perfect. But the Court concludes that Semo and Bard fulfilled their fiduciary duties under the relevant standard of care.

The Plan, as revised by the Third Amendment, was ambiguous. Clark has repeatedly tried to avoid this fact by pointing out that the "Group B" (or "Class B") language was not, when read alone, ambiguous. See, e.g., Pl.'s F & C at 49 ("[T]here is `no ambiguity' in the provision in the Third Amendment that `All Participants who are Class A Shareholders of the Employer and who were born on or after January 1, 1950' are in the 20% classification (or `Group B')."). Clark's counsel also tried to elicit testimony from witnesses to the effect that the Group B language, when read alone, is not ambiguous. See, e.g., Tr. 939:14-942:12 (Anspach). But it does not make sense to take one sentence of any document — statute, contract, or, in this case, ERISA plan — out of context and declare that it is "unambiguous." By Clark's logic, the language defining Group C was also "unambiguous," which would mean that Clark should have unambiguously been placed in two groups at once. The Plan simply cannot be read that way. Hence, the case law about fiduciary decisions that contradict the unambiguous language of the plan in question simply is not helpful here.

On the basis of the evidence at trial, the Court finds credible Semo and Bard's testimony that their decision on Clark's appeal was based on their understanding that Bard and Clark had been intentionally placed in the 10% grouping for plan years 2000, 2001, and 2002 and that there was no reason to change that grouping. Bard testified clearly and credibly that he thought that Group B was created for Semo. See Tr. 323:18-324:9 (Bard). From the chronological sequence of events, it appears that Anspach sent the firm Bard's June 2000 memorandum at the very same time that he alerted the firm to the "drafting error," on November 4, 2005. See Pl.'s Exs. 36-37; Tr. 896:18-897:12 (Anspach); Tr. 1073:18-22 (Semo). The Court believes it would have been reasonable for Bard and Semo to draw the conclusion from this memorandum that Bard and Clark were intended to be placed in the 10% category for the relevant years, especially as it confirmed Bard's understanding of the arrangement. It is true that the memorandum could also be read, as Clark insists it must be, to relate only to the firm's "final contribution to the Plan" for that particular plan year. But that is not the issue. The issue is whether Bard and Semo could have reasonably drawn the conclusion from the memorandum that 10% was the appropriate grouping. The Court believes that the memorandum could have reasonably given Bard and Semo that impression.

The defendants have made much of the fact that Bard's input into Clark's appeal was contrary to his self-interest, since any action to improve Clark's grouping for the years in question probably would have improved his own status. The Court finds that Bard's incentive in this regard somewhat supports the veracity of his testimony that he truly believed the 10% grouping was appropriate for both himself and Clark. To be fair to Clark, there may also have been significant informal pressure on Bard to resolve the situation in a way more favorable to his current than to his former co-workers, even if this meant he himself would be put slightly at a detriment. In any case, this argument was of particular significance to Semo, who testified repeatedly that he relied on Bard's statement against what Semo perceived was Bard's self-interest in coming to the conclusion that Bard and Clark had been appropriately grouped. The Court finds this testimony credible, especially as it was corroborated by Nielsen, who also indicated that he found Bard's statement against interest significant. In other words, the Court finds credible Semo's assertion that he relied on Bard's understanding of the Plan documents because Bard had an incentive to read the documents in the other way. This is true even though Semo had somewhat of an incentive to resolve the appeal against Clark, since resolving it in her favor would have resulted in a smaller pro rata share for himself. Although this fact would have given Semo an incentive to resolve the appeal against Clark, the Court finds credible the testimony that both he and Nielsen were legitimately persuaded by Bard's statement against interest.19 And the Court finds that Semo behaved reasonably by deferring to Bard's understanding under those circumstances.

In coming to these conclusions about Semo and Bard's behavior, the Court takes into consideration the process by which they reviewed Clark's appeal at the end of 2005. It is not difficult to believe that Semo and Bard were not predisposed to resolve the matter in Clark's favor. At this point, Clark had left the firm more than two years prior and was seeking additional money that would come at the expense of current employees' distributions. Furthermore, at least once in the record, Anspach makes reference to "t[ying] together our position," Pl.'s Ex. 67, suggesting that the defendants were not approaching the question with entire equanimity. On the other hand, the reasonableness of Semo and Bard's conduct is supported by the fact that they involved Nielsen in the process as a "set of fresh eyes to go over the issues with them." Tr. 804:21-805:9 (Nielsen). The fact that Semo and Bard brought in a third party without preconceptions about the outcome of Clark's appeal suggests that they considered the issue in good faith. Furthermore, the fact that Nielsen's testimony on why the group came to the decision that they did — the group relied on Bard's understanding as to the original intent — strongly corroborates Bard and Semo's testimony that this was the real reason why they resolved Clark's appeal in the way that they did.

To be sure, the Court does find significance in the fact that Bard was actually grouped in the 20% category for the 2002 plan year, while Clark was grouped in the 10% category. Inconsistent treatment of similarly situated participants by Plan fiduciaries is troubling. And Bard's 2002 grouping also arguably suggests that Clark and Bard were not put into the 10% category with clear intentionality. Were the Court deciding which group to place Clark into in the first instance, this fact might accordingly carry significant weight. But the Court is mindful of its role in reviewing a decision by a fiduciary. The question is not whether, all things being equal, Clark might have been fairly moved into the 20% grouping, but rather whether Bard and Semo's actions were reasonable. And the Court finds credible Bard and Semo's testimony that they had no idea that Bard had been placed in the 20% category for the 2002 plan year at the time that they were resolving Clark's appeal. Indeed, it is not even clear to the Court that Bard's 2002 grouping can properly be considered to be part of the administrative record in this case. Clark has indicated that she received this information during discovery. See Tr. 13:18-20 (Clark). It is not clear when the information surfaced, and certainly not self-evident that the information was before Bard and Semo when they came to their decision on Clark's appeal. Anspach appears to have been the source of Bard and Semo's information about the groupings, and there is no particular reason to believe that Anspach provided the information about Bard's grouping in the 2002 plan year to Semo and Bard in the context of Clark's appeal. Semo testified that he did not recall looking at such calculations and relied on Anspach's statements that the two were grouped identically. See Tr. 549:14-550:9 (Semo). In any case, the Court finds credible Bard and Semo's testimony that they believed Bard and Clark to have been treated the same during their overlapping years at the firm. Furthermore, given that Bard and Clark were generally treated the same while at the firm — which is further borne out by the fact that they were treated the same for all the other overlapping years — the Court is not persuaded by Clark's contentions that Bard and Semo had some further duty to perform "due diligence" and verify Anspach's statements. Anspach presented information indicating that Bard and Clark had been grouped in the 10% category for the years in question and that information comported with the general way that Bard and Clark had been treated. It was reasonable for Bard and Semo to believe that Anspach's representations were accurate.

The Court also does not agree with Clark that Bard and Semo's determination with respect to Clark was inconsistent with their treatment of themselves. Bard and Semo do not seem to have had any role in Bard being placed in the higher grouping for the year that Bard and Clark were both at the firm. As noted, the Court finds credible Bard, Semo, Anspach, and Nielsen's testimony that they intended to treat Bard and Clark the same for those years. Furthermore, Semo was simply not similarly situated to Bard and Clark. The BCA expressly provided that he receive 20% accruals, and he was clearly viewed as a more senior member of the firm.

The Court is not especially persuaded by the testimony from Anspach that the judgment on Clark's appeal was made due to "a decision to leave everything the same and leave it how it was" because the group did not "f[ind] anything different to change the answer." Tr. 900:14-906:13 (Anspach). These statements border on the arbitrary; they come close to articulating the point of view that a decision should stand in light of no further information even if the original decision was not supported by any particular reasoning. If that were all that Semo and Bard were relying on, the ambiguities in the Plan would probably have to be resolved in Clark's favor. But the Court is sufficiently persuaded by the testimony that Bard and Semo legitimately believed that the initial placement decision had been purposeful, and that therefore Bard and Semo's resolution of Clark's appeal was reasonable. Furthermore, the testimony at trial indicated that Reddington was regularly receiving information from Moline about Plan groupings, and this fact could have fairly suggested to Semo and Bard that the initial grouping was purposeful and should only be revisited for a good reason.

The Court does not put much weight in the fact that Reddington appears to have placed Bard and Clark in the 20% grouping in his preparation of the final Actuarial Valuation Report. Since the Report was dated June 2006, it cannot itself have factored into the decision on Clark's appeal. There was no testimony at trial that indicated that Reddington had any authority to place individuals in particular groupings, nor was there any testimony, or other reason to believe, that Reddington had been instructed to place Bard and Clark that way by someone with authority to make that decision. That Reddington did so seems to have been first "discovered" by Clark's expert, Poulin; there is no indication that Bard and Semo were even aware of it. See Pl.'s Ex. 96 at 10-11. It appears to the Court that Reddington, having been told to run the numbers in this fashion in the consideration of Clark's appeal, simply included them in the Actuarial Valuation Report for no particular reason.

Finally, there is the issue of the 2003 restatement of the Plan and how it relates to Clark's grouping under the earlier Third Amendment. The 2003 restatement is relevant in two respects. First, in resolving the ambiguity between Group B and Group C by placing participants such as Clark and Bard into Group B, the restatement could arguably suggest that the firm intended to place these participants into the more favorable grouping all along. However, this reading of the restatement was belied by Anspach's testimony, which indicated that he drafted the 2003 restatement to resolve the Third Amendment's ambiguity without any thought to how it would affect particular Plan participants or how participants were previously grouped under the Plan. Anspach's cover letter accompanying the restatement supports this assessment, since it does not even mention the change. See Pl.'s Ex. 37. Substantial testimony at trial also indicated that Anspach and Reddington were somewhat removed from the day-to-day operation of the firm and dealt with the Plan's operation without much involvement with the firm's personnel, suggesting that Anspach was unlikely to have made the change with certain individuals in mind. The Court finds this testimony credible, even though Anspach did also state that he "may have had a conversation with somebody at the firm" about the groupings but could not recall. Tr. 935:12-17 (Anspach). In any case, there certainly is no evidence in the record to indicate that Bard and Semo had some knowledge about the intent behind the restatement at the time they were reviewing Clark's appeal.

Second, there is also the argument that the 2003 restatement actually had retroactive effect, meaning that the restatement itself put Clark into the 20% category for the 2000, 2001, and 2002 plan years. Anspach's November 2005 email to Semo indicated unequivocally that "[s]ince Denise was gone by [October 1, 2002], this restatement does not affect her." Pl.'s Ex. 67. Semo and Bard never seem to have second-guessed that assessment. And Anspach testified at trial that he intended, in drafting the 2003 restatement, for it to affect only plan years after its effective date. But the restatement itself is not so clear. It amended the groupings such that Bard and Clark, if the new groupings were to apply to them (as they did to Bard in the 2003 plan year), would receive 20% accruals. And the restatement indicated what the old groupings were "[f]or the period commencing October 1, 1993 and ending September 30, 1998," Ex. 3 § 5.1(e), but did not indicate what the groupings were for the period beginning October 1, 1998. That is, the restatement indicated what the new groupings were and noted what the prior groupings were for the period from October 1, 1993, to September 30, 1998, but did not note anything about the groupings from October 1, 1998, to the date of the restatement. One reading of this language would be that the restatement's new groupings actually superseded the Third Amendment for the period beginning October 1, 1998. See Tr. 1129:14-1133:11 (Semo). On the other hand, the restated Plan does not actually say that it has any retroactive effect, it had an effective date of October 1, 2002, and Anspach clearly testified that he drafted it with the intent only to affect accruals from the effective date forward. Defendants have also noted that Clark did not raise this interpretation of the restatement until submitting findings of fact and conclusions of law following trial.20 Indeed, Clark's counsel's opening statement at trial arguably suggested that the 2003 restatement did not apply retroactively. See Tr. 12:19-23 ("So there was some initial confusion, but then in the year 2003, that mistake was apparently identified, and Mr. Bard was put into the 20 percent classification for two years. But they didn't correct his classification all the way back, and they didn't correct Ms. Clark's at all.")

The Court is mindful that, if the Plan as restated in 2003 unambiguously indicated that Clark was entitled to a 20% accrual percentage for the three years in question, she would have a strong case that she was entitled to that grouping upon the Plan's termination See Wagener, 407 F.3d at 404 ("As plaintiffs justly contend, it is patently unreasonable for the Committee and other Plan officials who are authorized to administer the Plan to interpret the Plan in a manner that discriminates against plaintiffs in direct contravention of the Plan's plain language."). But the Plan's language is not unambiguous on this point. The Plan provided for accrual of benefits over time. The 2003 restatement contained an explicit effective date of October 1, 2002. As far as the Court can tell, it would have been completely inconsistent with everything else in the Plan for an amendment to reach back in time and change account balances previously accrued for time periods prior to the effective date of the amendment. (Indeed, doing so in a way that would hurt Plan participants would usually be illegal. See Clark IV, 808 F.Supp.2d at 227.) It is therefore reasonable to read the Plan's language as providing for new groupings for benefits accrued beginning October 1, 2002, but not before. The Court finds Anspach's testimony that this was his intent in drafting the language credible, especially given that it was the interpretation that he expressed from the beginning upon raising the issue with Semo. Furthermore, the Court is also aware that Wagener was a suit against the plan itself, whereas here Semo and Bard would be personally liable to Clark. Even if Semo and Bard had honestly misinterpreted the Plan — and the Court is not convinced that they did — it would not be clear that they would be liable for a breach of fiduciary duty for an honest mistake based on Anspach's representations. See Morgan, 975 F.2d at 1471 ("As the case is presented to us, the defendant trustees made a mistake in interpreting the plan.... [T]he trustees are not subject to strict liability for their mistake.").

In sum, the Court finds credible Bard and Semo's testimony, corroborated by Nielsen, that they relied on Bard's understanding that he and Clark were intended to be placed in the 10% category for plan years 2000, 2001, and 2002. The Court concludes that this justification was reasonable, given the ambiguous Plan language, and was not a breach of fiduciary duty. The Court acknowledges that the statement that Bard was treated the same as Clark for the relevant years was false, but the Court finds credible the evidence that Bard and Semo did not know that the statement was false and behaved reasonably in believing that it was true. The Court is mindful of both the fact that the 2003 restatement resolved the ambiguity in what would have been Clark's favor (had she remained at the firm) and that the 2003 restatement arguably could be interpreted to apply retroactively to Clark. Nonetheless, the Court concludes that the restatement does not provide any insight into how the previous iteration of the Plan should be interpreted, and concludes that the 2003 restatement could reasonably be interpreted by Semo and Bard not to apply retroactively, especially given Anspach's statements to that effect. Accordingly, the Court concludes that Semo and Bard did not behave negligently or in bad faith and therefore that they did not breach their fiduciary duty.

b. Summary Plan Description Claim

Clark contends that Feder Semo violated ERISA's disclosure requirements by failing to disclose in the summary plan description the risk of loss of benefits at Plan termination and the Plan's lack of PBGC insurance. Pl.'s Statement at 7-8.21 Clark maintains that the SPD's statement that "[i]f the Plan terminates, you will become 100% vested in your accrued benefit and your accrued benefit will be distributed to you in accordance with the terms of the Plan," Pl.'s Ex. 4 at 6-7, conflicted with the actual Plan provisions that distributions would be made only to the extent funded. Pl.'s Statement at 7-8. Clark claims that the SPD therefore violated two Department of Labor Regulations. First, Clark contends that the SPD violated the requirement to include "a statement clearly identifying circumstances which may result in ... loss ... of any benefits that a participant or beneficiary might otherwise reasonably expect the plan to provide on the basis of the description of benefits." 29 C.F.R. 2520.102-3(l). Second, Clark contends that the SPD violated the requirement to include, "[i]f the benefits of the plan are not insured [by PBGC], a statement of this fact, and reason for the lack of insurance." 29 C.F.R. 2520.102-3(m)(1). Clark asserts that if she had been informed of "the risk of loss on plan termination," she "could have complained about the underfunding" and "would have elected the $227,000 lump sum offered in the September 2, 2005 letter." Pl.'s F & C at 28; see also Tr. at 59:23-60:10 (Clark). Feder Semo maintains that the SPD made the appropriate disclosures. Defs.' Proposed Findings of Fact and Conclusions of Law [Docket Entry 130] ("Defs.' F & C") at 28-30. The firm also maintains that even if the SPD were deficient, Clark would not be able to prevail on her SPD claim because she has not shown that Feder Semo was unjustly enriched or that she was harmed by the SPD language. Id. at 30-32.

The Court previously denied Feder Semo's request for summary judgment on Clark's SPD claim. The Court stated:

Clark IV, 808 F.Supp.2d at 230-31.

Pursuant to 29 U.S.C. § 1022(a), a summary plan description must be furnished to plan participants and beneficiaries and must "be sufficiently accurate and comprehensive to reasonably apprise such participants and beneficiaries of their rights and obligations under the plan." The Supreme Court may well have granted certiorari in CIGNA in order to clarify the standard by which harm must be shown for a claim based on this provision. See Tr. of Oral Argument at 33:16-20, 2010 WL 4859511, CIGNA, 131 S.Ct. 1886 (No. 09-804) (Breyer, J.) ("[A]s long as you show some kind of reliance and harm, and then we're back to what I thought we granted this for, which is why not say if harm is likely, then the burden shifts?"). The Supreme Court indicated that, although a fiduciary can be surcharged "only upon a showing of actual harm," that actual harm "may sometimes consist of detrimental reliance, but it might also come from the loss of a right protected by ERISA or its trust-law antecedents." CIGNA, 131 S.Ct. at 1881; see also id. ("In the present case, it is not difficult to imagine how the failure to provide proper summary information, in violation of the statute, injured employees even if they did not themselves act in reliance on summary documents — which they might not themselves have seen — for they may have thought fellow employees, or informal workplace discussion, would have let them know if, say, plan changes would likely prove harmful.").

Somewhat unfortunately, the standard is not especially clear following CIGNA. The distinctions between "actual harm," "detrimental reliance," and "injury" are not obvious, and the cases interpreting CIGNA have revealed the difficulty. In Skinner v. Northrop Grumman Ret. Plan, 673 F.3d 1162, 1167 (9th Cir.2012), the Ninth Circuit concluded that "considering that [plaintiffs] did not rely on the inaccurate SPD, they establish no harm for which they should be compensated." It is difficult to reconcile this conclusion with CIGNA's statement that employees may have been "injured ... even if they did not themselves act in reliance on summary documents." 131 S.Ct. at 1881. On the other hand, this Court agrees with Skinner's rejection of the argument that "the `harm' of being deprived of [a] statutory right to an accurate SPD is a compensable harm," because such an interpretation "would render [the plan administrator] strictly liable for every mistake in summary documents." 673 F.3d at 1167; see CIGNA, 131 S.Ct. at 1881 ("We believe that, to obtain relief ..., a plan participant must show that the violation injured him or her."). In contrast, in Killian v. Concert Health Plan, 680 F.3d 749, 757 n. 7 (7th Cir.2012), the Seventh Circuit concluded that it "need not discuss the ramifications of [CIGNA] here, however, because [plaintiff] fails to demonstrate that any breach on the part of [defendants] actually caused his harm." That court concluded that "a reasonable fact finder could only conclude that [plaintiffs] would have" taken the same course of conduct regardless of whether they were provided an accurate SPD. Id. at 758. The Seventh Circuit has since granted en banc review of Killian, although the partial dissent did not differ with the majority on this issue. See id. at 768 (Ripple, J., concurring in part and dissenting in part) ("There is, therefore, no evidence that [plaintiff] incurred these medical bills because she did not know....").

As noted above, relief under 29 U.S.C. § 1132(a)(3) is for "appropriate equitable relief," and the Supreme Court was clear in CIGNA that the remedy arises under "the law of equity." 131 S.Ct. at 1881. Under the facts of this case, and with the benefit of the full record, the Court determines that Clark has not shown that she was harmed by the SPD.

To be sure, the SPD was not a model of clarity. The statement that "[b]enefits under this Plan are insured by the PBGC if this Plan has more than 25 participants," Pl.'s Ex. 4 at 7, was itself arguably inaccurate. The Plan did at times have more than 25 participants — albeit not "active" participants, a word not used in the SPD description — and yet the Plan was not covered by PBGC insurance. And even if the statement were accurate (i.e., if "participants" was understood to mean current employee participants), it would be hard to conclude that it was "a statement of th[e] fact" that "the benefits of the plan are not insured" by the PBGC, as required by 29 C.F.R. § 2520.102-3(m)(l). Furthermore, the Court cannot agree with Feder Semo that the SPD actually indicated that participants might receive less than their full benefit distribution. The firm tries to find this meaning in the sentence stating that "[i]f PBGC covers this Plan and, on termination, if the Plan does not have sufficient assets to pay the benefit, the PBGC will provide part or all of that benefit." Pl.'s Ex. 4 at 7. This statement does imply that there might be circumstances in which the Plan had insufficient assets and Plan members would receive only "part" of the benefit. But it is hardly "a statement clearly identifying circumstances which may result in ... loss ... of any benefits," as required by 29 C.F.R. § 2520.102-3(l). On the other hand, Clark's reliance on Burstein v. Retirement Account Plan for Employees of Allegheny Health Education & Research Foundation, 334 F.3d 365 (3d Cir.2003), is misplaced. That case involved an SPD that "conflict[ed] with the plan language," id. at 378, which is not the case here, where the SPD simply failed to provide, at least in a straightforward manner, the relevant information.

In any case, the Court cannot conclude that a clearer SPD would have conveyed to Clark any information that she did not already know or that could plausibly have affected her behavior.22 Clark testified that she was provided with and was aware of the SPD during her tenure at the firm, but candidly admitted that she did not rely on it for any information. That alone would not be enough to preclude recovery, but it certainly does not help her case. More importantly, though, is the fact that Clark was the managing partner of Feder Semo, intimately aware not only of the firm's affairs, but also the Plan's, and an experienced ERISA lawyer to boot. As a factual matter, the Court finds it simply implausible that Clark was unaware that retirement plans that terminate unexpectedly can only make distributions to the extent those distributions are funded. She cannot deny that she knew the Plan was less than "fully funded," since she herself oversaw a significant distribution to the Feders that reduced the Plan's assets below its liabilities. And with respect to PBGC insurance, it is simply not credible for Clark to contend that a better SPD would have alerted her to the fact that the Plan did not have the insurance when she herself would have been responsible for providing for the insurance (or not) as managing partner. Clark has admitted that she was aware the firm was not paying PBGC insurance premiums, and she accordingly notes that "PBGC insurance coverage is not dependent on timely payment of insurance premiums," Pl.'s F & C at 55 (citing 29 U.S.C. § 1307(d)). But she did not actually testify — nor has she ever really asserted — that she was unaware of the firm's lack of insurance. Given her experience and her role at the firm, the Court concludes it is not credible that she was unaware of this fact.

There may well be individuals who are not knowledgeable about the workings of retirement plans generally or specifically the retirement plan in which they participate. Even if they themselves did not rely on the SPD, those individuals can plausibly say they have been harmed by an inadequate SPD, because an inadequate SPD can be said to deny information to even those employees who do not themselves rely on the document. But a plaintiff who is fully knowledgeable of whatever information was missing from an SPD cannot fairly be said to have been harmed by the SPD's inadequacies. To hold the plan administrator liable under such circumstances would be, as the Ninth Circuit indicated, to hold plan administrators "strictly liable" — if that is the right phrase — "for every mistake in summary documents." Skinner, 673 F.3d at 1167. That cannot be right. Clark was such a fully knowledgeable employee. Indeed, the Court suspects that the reason that Clark never read the SPD is because she was confident — and quite appropriately so — that she knew everything it had to say about the Plan, and more.

Finally, the Court notes that what Clark contends she would have done had the SPD been more accurate is also not a particularly strong argument for injury. The time period in which it would even be plausible for Clark to assert that she could have been harmed by the SPD — that is, the time period in which Clark was not actually herself administering the Plan as managing partner of the firm — was after she left the firm. At that point, Clark was not in a strong position to negotiate for greater benefits or consider leaving the firm, which are the paradigmatic examples of how an employee might gainfully react to a fully disclosed inadequacy in a retirement plan. Cf. Tr. of Oral Argument at 36:17-37:24, 2010 WL 4859511, Cigna, 131 S.Ct. 1886 (No. 09-804) (Alito, J.). Clark asserts that she was injured because she would have either "complained about the underfunding" or would have responded to the September 2, 2005, letter more quickly. But there is little reason to believe that either of these actions would have actually resulted in Clark ultimately receiving more money. With respect to complaining about the underfunding, she testified at trial that she actually did stress to Semo the importance of funding the Plan because "[t]here are others of us [other than Feder] who are depending on the plan." Tr. 63:9-16 (Clark). It is hard to believe that Clark would have been somewhat more vocal if the SPD had more clearly articulated the risk of loss or that this would have had any effect on how the Plan was funded. Semo and Bard were aware the Plan was somewhat underfunded and had taken action to put more money into the Plan; by all accounts, they were blindsided by the loss of their biggest client and subsequent unexpected closure of the firm. That additional complaining from a former employee would have had any impact strains credulity. Furthermore, it is quite clear from the record that no one received the distributions that they were quoted in the September 2 letter and that the change in the firm's status from tottering to closing — not the speed with which participants returned their election forms — was the reason. Hence, the Court also believes that Clark's claim that she returned her election form more slowly due to the inadequate SPD has no import. In sum, even if a more adequate SPD were to have done a better job of disclosing the relevant information — which, as indicated, the Court concludes would not have actually told Clark anything she did not already know — it is hard to believe that Clark was in a position to do anything with that information that would have made a difference, and hence hard to conclude that she has even articulated a viable theory of how she was harmed by the inadequate disclosure. Hence, the Court concludes that Clark cannot recover on her SPD claim.

c. Interest Rate Assumption Claim

Clark contends that Feder Semo and individual defendants Semo and Bard breached their fiduciary duty to "maintain the Plan on a sound actuarial basis" by failing to correct the underfunding of the Plan caused by an "unreasonable" 8% interest rate assumption. Pl.'s Statement at 13. Clark relies on 29 U.S.C. § 1083(h)(1)(A), which requires that "the determination of any present value or other computation under this section shall be made on the basis of actuarial assumptions and methods — [] each of which is reasonable (taking into account the experience of the plan and reasonable expectations)." She also maintains that the Plan's own funding policy requires that the Plan be maintained on a "sound actuarial basis." Pl.'s Statement at 15. Defendants counter that actuaries, rather than plan sponsors, have the duty of ensuring that actuarial assumptions utilized by a plan are reasonable. Defs.' F & C at 37-38. They note that Reddington consistently certified that the actuarial assumptions used to fund the Plan, including the interest rate assumption, were reasonable. Id. at 38. Furthermore, defendants contend that the 8% interest rate assumption was reasonable. Id. at 38-39. They argue that the 8% assumption was reasonable as a reflection of the Plan's historical investment returns. Id. Defendants argue that ERISA plan sponsors are required to fund plans on an "ongoing basis," rather than a "current liability" (i.e., liabilities at Plan termination) basis. Id. at 39. They assert that Poulin's expert testimony that the interest rate assumption should have taken into account that all Plan participants would receive a lump sum distribution was based on hindsight and did not reflect the expectations of the Plan. Id. at 39-40. Defendants dispute that the Plan language about "fully funding" the Plan was ever intended to indicate funding based on "current liability." Id. at 40-41. Finally, defendants argue that they behaved diligently as fiduciaries in trying to remedy the underfunding when Reddington presented the issue. Id. at 41-42.

The Court is not persuaded by defendants' argument about the role of actuaries in setting Plan assumptions. On this topic, the Court has already stated: "Defendants first contend that selecting the interest rate is the role of the actuary, not the plan sponsor or other plan fiduciaries. But plaintiff has presented evidence that defendants were at least involved in the rate selection process, and even the Plan's outside counsel noted that he thought" the plan's sponsor rather than the actuary has the final authority on all assumptions. Clark IV, 808 F.Supp.2d at 232. The Court is not swayed by defendants' current arguments to the contrary. It is true that, as the Supreme Court has noted, under ERISA "the assumptions and methods used in calculating withdrawal liability," as well as in setting actuarial assumptions, "are selected in the first instance not by the trustees, but by the plan actuary." Concrete Pipe & Prods. v. Constr. Laborers Pension Trust, 508 U.S. 602, 631-33, 113 S.Ct. 2264, 124 L.Ed.2d 539 (1993) (citing 29 U.S.C. §§ 1082(c)(3), 1401(a)(3)(B)). But Concrete Pipe also noted that "[i]t may be that the trustees could, in theory, replace the actuary's assumptions with their own." Id. at 633 n. 19, 113 S.Ct. 2264. The other cases cited by defendants are similarly not on point. In Citrus Valley Estates, Inc. v. C.I.R., 49 F.3d 1410, 1414 (9th Cir.1995), the court did state that "Congress consciously left the specifics of [individually defined benefit] plan funding in the able hands of professional actuaries," but was contrasting that principle with "legislating mandatory funding assumptions and methods" (emphasis added). That is, the court was making the point that actuarial assumptions are left up to the plan, not distinguishing between the actuary's role and the role of plan fiduciaries. The Fifth Circuit was making a similar point in Vinson & Elkins v. C.I.R., 7 F.3d 1235, 1238 (5th Cir.1993), stating, for example, that "[t]he statute refers to the actuary's best estimate, not that of a court or of outside experts." In this Court's view, the question for the finder-of-fact is whether those involved with running the Plan behaved consistently with their fiduciary duties in the role that they played in setting the actuarial assumptions. Where, as here, the Plan fiduciaries were involved in setting the interest rate assumption, it is no defense to say "the actuary made me do it."

Nonetheless, the Court concludes that neither Feder Semo nor Semo and Bard breached their fiduciary duties with respect to the interest rate assumption. With respect to the interest rate assumption standing alone, the Court finds that the assumption was within the range of reasonableness. Clark has not disputed that the 8% assumption was consistent with the Plan's prior earnings on its investments. This assumption, in retrospect, reflected a rosy picture of what the future would bring and the Plan did not continue earning at that rate. But this is only clear with hindsight; the firm cannot reasonably be faulted for failing to anticipate that the future would deviate from consistent past experience. Defendants' expert, Altman, likewise presented credible evidence that 8% was a common assumption at the time. Clark has countered by arguing that other plans were making off-setting assumptions elsewhere to compensate for the 8% interest rate assumption, but again has not really disputed that an 8% assumption was reasonable specifically with respect to the earnings on the Plan's assets. Defendants' argument with respect to the reasonableness of the 8% interest rate assumption is further bolstered by the fact that Clark herself was the Plan official most involved in changing the rate from 7% to 8%.

Clark's more fundamental complaint about the interest rate assumption relates to the role the assumption played in the overall underfunding of the Plan. As a matter of law, the Court has some misgivings about this theory of recovery. To be sure, the Court recognizes that, in addition to the requirement under § 1083(h)(1)(A) that "each" actuarial assumption "is reasonable (taking into account the experience of the plan and reasonable expectations)," ERISA imposes an additional requirement that the actuarial assumptions, "in combination, offer the actuary's best estimate of anticipated experience under the plan," 29 U.S.C. § 1083(h)(1)(B). The express language of § 1083(h)(1)(B) puts the onus of ensuring actuarial assumptions that are reasonable in combination more clearly on the actuary than does § 1083(h)(1)(A), as § (h)(1)(B) specifically refers to "the actuary's best estimate." It strains common sense to believe that Plan fiduciaries normally would have some independent sense of the complicated holistic assessment of whether actuarial assumptions are, in the aggregate, reasonable, beyond what the actuary indicates. Furthermore, Clark's claim here is not really about the interest rate assumption per se, but rather about the overall funding of the Plan. And there is at least some authority for the proposition that "there is no freestanding fiduciary duty to fund a pension plan" beyond the minimum required contributions, Cress v. Wilson, No. 06 Civ. 2717(JGK), 2008 WL 5397580, at *9 (S.D.N.Y. Dec. 29, 2008), which suggests that Plan fiduciaries do not have an independent duty to ensure that plans have sufficient overall funds. But putting all that aside and assuming that the defendants could be faulted for their role in funding the Plan employing actuarial assumptions that led to an overall underfunding, despite the individual reasonableness of the interest rate assumption, the Court cannot conclude that the defendants violated their fiduciary duties.

The Court finds credible the large amount of testimony indicating that no one at Feder Semo had any inkling before July 2005 that the firm might lose its largest account or that the firm's closing was anything approaching imminent. It is hard to say that Clark has even disputed that this was a completely unexpected outcome. And it is also clear from the record that the inability of the Plan to meet its obligations was a direct result of the Plan's termination and would not have happened otherwise. Only the Plan's termination made lump sum distributions available to all participants simultaneously; before August 2005, all Plan participants younger than 65 would have had to wait at least five years after leaving the firm to take a lump sum distribution. Hence, there is no reason to believe that the Plan would have been unable to meet its liabilities if the firm had not lost the major account. Additionally, it was not only the fact that immediate lump sum distributions were required but also the especially low GATT rates at which they were calculated that resulted in Plan participants receiving only 53% of their calculated benefit. Clark's argument, therefore, must be not only that the defendants should have kept extra funds in reserve for the very unexpected contingency that the Plan would terminate, but also that the defendants should have planned for the contingency that GATT rates would be especially low when the Plan unexpectedly terminated. There is no way to conclude that a failure to anticipate these contingencies was a breach of fiduciary duty.

The Court finds the testimony of defense expert Altman and Clark's expert Poulin to be helpful and illuminating about how ERISA plans function in general and how this Plan functioned specifically. However, the Court does not find that the experts shed much light on the key point in contention. Altman testified that he thought 8% was a reasonable interest rate assumption as an expected rate of investment returns during the period in question, and that he did not think plans had an obligation to be prepared for distributions upon termination if termination of the Plan was not anticipated. See Tr. 721:1-722:20, 733:13-734:3, 736:1-18, 769:7-14, 778:19-780:11, 784:15-18, 786:18-21 (Altman). Poulin did not really dispute that 8% was reasonable as an assumption about the expected rate of return on the Plan's investments, but stated that he believed plans should be funded with an eye toward termination liability and that, in turn, actuarial assumptions should reflect liabilities on plan termination. See Tr. 220:18-221:18, 226:13-227:13, 229:16-231:8, 231:20-24, 232:4-11, 285:24-286:1, 289:10-290:5, 300:6-301:5 (Poulin). The issue having been teed up, the experts did not have much more to say. Neither expert testified, for example, about how common unexpected plan termination is, which would shed some light on whether a responsible fiduciary would be wise to anticipate that contingency (or not). Cf. Tr. 287:15-17 (Poulin). The Court was left with the impression that reasonable experts could differ about whether plans should be funded with respect to termination liability. And at the end of the day, the Court believes that the determination of what the fiduciaries should have reasonably anticipated must be decided by the finder-of-fact. Here, without any clear, persuasive guidance from the experts, the Court concludes that the defendants could not reasonably be expected to fund the Plan to account for termination liability when that contingency was completely unexpected.

The Court is not persuaded by Clark's repeated attempts to find testimony indicating that the Plan's requirement to be "fully funded" somehow translated to "fully funded at Plan termination." No witness with any firsthand knowledge actually articulated this idea, despite ample prodding from Clark's counsel. See, e.g., Feder Tr. 48:8-22; Tr. 1021:14-1022:13 (Anspach). The only witness to actually state that the intention of the Plan was to be 100% funded on a termination basis was Poulin, who merely testified about his understanding of Feder's deposition testimony — hardly a convincing foundation for Poulin's conclusion. See Tr. 210:25-211:14 (Poulin). Clark was able to elicit substantial testimony from several witnesses that the fiduciaries had a responsibility to "fully fund" the Plan, but this raises rather than answers the question of what that phrase means.

Furthermore, the Court finds credible the testimony indicating that defendants were aware of the Plan's underfunding with respect to its ongoing or expected liabilities and took reasonable steps to remedy that deficiency. Defendants were notified by Reddington that the Plan was underfunded by approximately $200,000 and contributed what they understood to be $100,000 above the minimum contribution to remedy the shortfall. See Pl.'s Ex. 34; Tr. 858:23-859:2 (Anspach); Tr. 509:12-511:14, 533:21-535:7, 1069:4-1070:13 (Semo). It appears from the record that the defendants may have actually, unbeknownst to them, contributed less than $100,000 above the minimum contribution, see Pl.'s Ex. 19 at D0222, but the Court finds nothing in the record to suggest that they had any reason to know this, given Reddington's representations. Since defendants were putting additional monies over the minimum contribution into the Plan, the Court does not find credible Clark's implied assertion that defendants were intentionally underfunding the Plan to prevent Gerald Feder from receiving his distribution. The fact that defendants were taking reasonable steps to remedy the underfunding on an ongoing basis indicates that they were taking their fiduciary duties seriously and behaving reasonably. Again, defendants had no reason to expect that the Plan would suddenly be required to pay all its distributions immediately and simultaneously. To be sure, defendants did only make the minimum contribution in the final year when the Plan was terminating, and this action resulted in taking a "credit" of the previous contribution above the minimum. See Tr. 653:18-654:20 (Reddington); Pl.'s Ex. 14 at D0161, D0163. But that one action is insufficient to change the overall determination that the defendants were responsibly carrying out their fiduciary duties. The fact that Semo and Bard apparently approached Feder about forfeiting some of their own benefits to pay for Clark's distribution certainly suggests they were taking their duties seriously as the firm terminated. Hence, the Court finds that defendants did not violate their fiduciary duties with respect to the Plan's interest rate assumption.

IV. Conclusion

Having considered the extensive evidence in the record and the parties' extensive arguments, and applying the appropriately deferential standards when assessing defendants' conduct, the Court determines that Clark cannot prevail on any of her three claims. Clark presents a reasonable argument that she should have been grouped in a higher accrual category for three years, but Bard and Semo came to a similarly reasonable determination that Clark was appropriately grouped in the lower category. Clark cannot reasonably contend that she was harmed by the Plan's summary plan description. And the Plan's interest rate assumption was reasonable. Accordingly, judgment will be entered for the defendants. A separate order has been issued on this date.