RETIREMENT AND SECURITY PROGRAM FOR EMPLOYEES OF NATIONAL RURAL ELECTRIC COOPERATIVE ASSOCIATION and its Member Systems, Plaintiff,
v.
The OGLETHORPE POWER CORPORATION RETIREMENT INCOME PLAN, Defendant.
Civ. A. No. 88-0279.
United States District Court, District of Columbia.
April 25, 1989.*224 John Jay Range, Washington, D.C., for plaintiff.
William M. Hames and Thomas A. Cox, Atlanta, Ga., Willard Tom and Lee H. Pelton, Washington, D.C., for defendant.
MEMORANDUM AND ORDER
REVERCOMB, District Judge.
The plaintiff in this case seeks a declaratory judgment that it owes the defendant only "accrued" benefits generated by a retirement savings plan to which the defendant belonged from 1977-1985. The defendant, which has counterclaimed, argues that it is entitled to more money than the accrued benefits earmarked by the plaintiff for distribution to Oglethorpe's employees because it paid a substantially larger sum into the plaintiff's coffers than the amount which had accrued to its employees by the time it withdrew from the plaintiff's plan.
I. Factual Background.
Oglethorpe Power Corporation ("Oglethorpe") is a Georgia non-profit corporation owned by a consortium of 39 rural electric cooperatives located in Georgia. In January 1977, Oglethorpe contracted with an employee benefits consulting firm, the Hay Group, seeking a recommendation of an employee benefit plan for Oglethorpe's employees. The Hay Group recommended the National Rural Electric Cooperative Association, ("NRECA"), a national trade association of rural electric cooperatives, which sponsors a pension plan for its members called the Retirement and Security Program for Employees of National Rural Electric Corporations Association and its Member Systems (the "R & S Program" or "R & S"). Oglethorpe's Board of Directors authorized it to join R & S, and it joined R & S effective January 1, 1977.
Oglethorpe participated in the R & S Program for eight years, until, in January 1985, it communicated to R & S its intention to withdraw from the Program. At that time, Oglethorpe decided to create its own pension plan, known as the Oglethorpe Power Corporation Retirement Income Plan. By letter of February 27, 1985, Oglethorpe formally advised R & S that it was withdrawing from the Program effecive December 31, 1984. Oglethorpe intended to withdraw its funds from the R & S Program and place the money in the Oglethorpe Plan. It appears from the stipulated facts and from Exhibit 35 that the Administrator of R & S advised Oglethorpe by letter of the steps required to withdraw on February 6, 1985, explaining to Oglethorpe (with what Oglethorpe argues was less than perfect clarity) the method which would be used to calculate the amount transferable from the R & S Program to the Oglethorpe Plan.
Oglethorpe contributed $2,194,570.90 to fund pension benefits for its employees. Although R & S paid approximately $64,000 in benefits to Oglethorpe workers during the eight years at issue, it has no continuing liability to Oglethorpe Plan participants.
II. The Rival Methods of Calculation.
It is the calculation of the amount owing to Oglethorpe upon withdrawal which forms the basis of the dispute between the parties to this case. Section 16 of the R & S Program Specifications, which governs withdrawal from the Program, states that "an actuarial determination shall be made as to that portion of the Trust Fund which is held on behalf of Participants of the withdrawing System." At the time Oglethorpe decided to withdraw from R & S, the plaintiff calculated the amount of funds held in trust "on behalf of" Oglethorpe participants according to the "Accrued Benefits Method." This is an actuarial method which identifies the amount of retirement benefits which has accrued to each employee as an annuity based on the *225 amount of time the employee has worked. By this method, R & S calculated that the amount held "on behalf of Participants" in Oglethorpe's account was the sum of the present lump sum cash values which had accrued to all covered Oglethorpe employees. This method of determining the value of withdrawals from R & S was the standard calculation method used by R & S during this period, and it remained in effect without substantial change until July, 1987. During the time that the Accrued Benefits Method was used by R & S, 21 members withdrew from R & S, and the Accrued Benefits Method was consistently used to calculate the amount of funds transferable to those entities which withdrew.
It is plaintiff's position that Section 16 of the contract authorizes use of the Accrued Benefits Method since that method converts the accrued annuity benefit into a lump sum cash value, the aggregate of which would be the amount "held on behalf of Participants in the withdrawing System." R & S argues that this is a reasonable choice to make under Section 16, on the grounds that the approximately $1.5 million difference between what Oglethorpe has paid in and what it is entitled to receive back under the Accrued Benefits Method is authorized under ERISA, which allows plan administrators to transfer only enough assets to cover current accrued benefits. R & S further justifies payout using the Accrued Benefits Method on the grounds that Section 16 of the contract and consistent prior practice support its interpretation of the language "held on behalf of Participants" to mean accrued benefits, since the Accrued Benefits Method assumes that each employee is 100% vested in his accrued benefits. In effect, R & S is arguing that its duties as plan administrator under ERISA ought to focus on payments to employees, whose accrued benefits are protected under the Accrued Benefits Method, and that ERISA allows it to use any reasonable actuarial method which will protect employees.
Oglethorpe paid in to R & S at the "Entry Age Normal Method" level, which funds both current liabilities and pre-funds future liabilities. The pre-funding feature of the Entry Age Normal Method, acting in combination with the youthfulness of Oglethorpe's work force, created the large pre-payment amount of $2,194,570. Using the Entry Age Normal Method led to larger initial contributions than might have resulted from other methods, but Oglethorpe explains its willingness to pay into R & S using that method on the grounds that it had the effect of stabilizing annual payments to a level percentage of payroll. The practical effect of using the Entry Age Normal Method was that Oglethorpe made higher payments in order to pre-fund projected future benefits as well as those currently accruing during its period of membership. During the time Oglethorpe belonged to the R & S Program, its employees received $64,251 in benefits. Because Oglethorpe's workforce was relatively young, there was a large difference between the Entry Age Normal Method and payments reflecting only the rate of current accrual, i.e., the funding equivalent of the Accrued Benefits Method. Under the latter method, contributions for a worker are lower at the beginning of a worker's career and steadily increase as the worker approaches retirement. Under the Entry Age Normal Method, by contrast, contributions remain relatively constant over a worker's career. Oglethorpe argues that even if the language of Section 16 quoted above could support using the Accrued Benefits Method, the clause is ambiguous, and use of the Accrued Benefits Method for withdrawal is inconsistent with the use of the rival Entry Age Normal Method in other parts of the R & S Program, most notably when paying into the Program.
Oglethorpe believes that it is entitled to a proportionate share of the surplus (funds held by R & S in excess of liabilities for currently accrued benefits) held in the R & S Program. To support its position, Oglethorpe argues that the language of Section 16 is not clear as to what funds might be said to be held "on behalf of" an employee. The exact amount to which Oglethorpe believes itself entitled varies depending on which of two suggested theories of recovery the Court might adopt: under the first *226 alternative, Oglethorpe would receive a share in surplus determined by use of the "Entry Age Normal Method," which was the actuarial method used to determine its contributions to the Program; in the alternative, Oglethorpe seeks an equitable accounting and distribution based on its level of contribution. In effect, Oglethorpe's position is that it made payments on the scale it did solely as a matter of accounting convenience, since the Entry Age Normal Method of payment allows for a smooth, unvarying pattern of contribution as a workforce ages. The fact that it selected this method solely out of convenience should not, according to Oglethorpe, require it to forfeit a sum nearly three times as large as the amount of accrued benefits to which R & S believes it to be entitled.
III. Standard of Review.
Subsequent to the trial in this case the Supreme Court heard argument in Firestone Tire and Rubber Co. v. Bruch, ___ U.S. ___, 109 S. Ct. 948, 103 L. Ed. 2d 80 (1989). Judgment in this case was withheld pending the decision in Firestone, because the issue before the Supreme Court was the standard of review to apply in reviewing denial of ERISA benefits. Firestone was decided on February 21st, 1989. The parties have had the opportunity to address the holding in Firestone as they believe it affects the standard of review in this case, and upon consideration of their arguments the Court has concluded that the actions of R & S are to be reviewed under the arbitrary and capricious standard.
The Supreme Court held in Firestone that the arbitrary and capricious standard should not be imported wholesale into ERISA, and found that a de novo standard should be applied to cases where a plan administrator has no discretion to interpret uncertain plan terms. 109 S. Ct. at 956. However, the Court also held that the arbitrary and capricious standard is the correct standard to apply when the contract grants interpretive discretion to the administrator. The Court stated that
[c]onsistent with established principles of trust law, we hold that a denial of benefits challenged under § 1132(a)(1)(B) is to be reviewed under a de novo standard unless the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan.
Id. at 956.
The implication of Firestone is that if the plan gives the fiduciary discretionary authority to construe the terms of the plan, his decisions should be reviewed under the arbitrary and capricious standard. Section 17.01 of the R & S Program Specification grants the Committee complete discretion to interpret the terms of the plan, stating in relevant part that
[t]he Committee shall have authority to determine all questions arising in connection with the Program, including its interpretation, and may adopt rules for the procedures of the Committee....
The decision or action of the Committee in respect of all matters within the scope of its authority shall be conclusive and binding on all persons.
Discretion granted to this extent is the type of contractually mandated delegation which the Supreme Court means to be reviewed under the arbitrary and capricious standard. The language of § 17.01 is at least as broad as the delegation contained in Central States, Southeast & Southwest Areas Pension Fund v. Central Transport, Inc., 472 U.S. 559, 568, 105 S. Ct. 2833, 2839, 86 L. Ed. 2d 447 (1985), cited favorably in Firestone.
IV. Use of the Accrued Benefits Method.
It remains to set forth the contours of that standard and to determine whether R & S's use of the Accrued Benefits Method was arbitrary or capricious.
In ERISA cases, the arbitrary and capricious standard has been interpreted to include a requirement of "reasonableness" in operation which restrains the fiduciary's discretion. In Edwards v. Wilkes-Barre Publishing Co. Pension Trust, 757 F.2d 52, 56-7 (3d Cir.1985), the court interpreted the fiduciary responsibilities of ERISA trustees to require a choice between reasonable alternatives; in order to perform *227 their duties as fiduciaries, plan trustees must select a "reasonable interpretation of the terms of the Plan," 757 F.2d at 56, but "[i]t is for the trustees, not the courts, to choose between two reasonable alternatives." Id. at 57 (quoting Tomlin v. Board of Trustees, 586 F.2d 148, 151 (9th Cir. 1978)). See also Stewart v. National Shopmen Pension Fund, 795 F.2d 1079, 1083 (D.C.Cir.1986). This reading of the arbitrary and capricious standard is consistent with the focus in Firestone on applying trust law principles, where "[a] trustee may be given power to construe disputed or doubtful terms, and in such circumstances the trustee's interpretation will not be disturbed if reasonable." 109 S. Ct. at 954.
Although Oglethorpe argues that the selection of the Accrued Benefits Method was improper because it unfairly leaves Oglethorpe with less than it would have if the calculation were made under the Entry Age Normal method, "ERISA does not require plan fiduciaries to maximize the benefits of departing employees[;] it only requires them to make a reasonable choice among possible alternatives." Foltz v. U.S. News & World Report, 663 F. Supp. 1494, 1519, (D.D.C.1987), aff'd, 865 F.2d 364 (D.C.Cir.1989). To determine whether a reasonable choice among alternatives had been made, Foltz used the four-part test established in Donovan v. Carlough, 576 F. Supp. 245, 249 (D.D.C.1983), aff'd 753 F.2d 166 (D.C.Cir.1985). As restated in Foltz, the Carlough test runs as follows:
In determining whether a fiduciary's interpretation of the terms of a plan document is arbitrary or capricious, four factors should be considered:
(1) whether the interpretation is contrary to the language of the plan; (2) whether it is consistent with the purposes of the plan; (3) whether it is consistent with the purposes of the particular provision itself; and (4) whether it is consistent with prior interpretations and whether beneficiaries were on notice of the interpretation.
663 F.Supp. at 1514.
The Court believes that the present case is similar enough to Carlough and Foltz to make these factors pertinent. Although the Court has no briefs before it defending the decision-making process by which R & S decided to select the Accrued Benefits Method, "that is no basis for overturning a decision that is entirely consistent with the Plan document and with ERISA's substantive requirements." Foltz, 865 F.2d 364, 374-75. The Court will consider the factors outlined in Foltz and Carlough in determining whether R & S acted arbitrarily or capriciously in selecting the Accrued Benefits Method.
(1) Whether Using the Accrued Benefits Method is Consistent with the Language of the Plan.
The method for determining what money is due to Oglethorpe upon withdrawal is not specified in the contract. As noted above, ERISA requires R & S to make a reasonable choice between available interpretations of the contract language, meaning that the Court should not require any particular interpretation, so long as the interpretation R & S chooses is reasonable. "Choices between reasonable alternatives ... are for the trustees, not the courts." Stewart v. National Shopmen Pension Fund, 795 F.2d at 1083.
Oglethorpe has argued that the termination clause, § 18.01 of the contract, contains language which implies that R & S will use the Entry Age Normal Method to calculate refunds upon withdrawal, since use of the Accrued Benefits Method is specified in § 18.01, while more ambiguous language is used for withdrawals under § 16. Oglethorpe claims that this shows that the drafter knew how to express the Accrued Benefits Method with clarity when that method was intended, thus implying that the absence of comparable specificity in § 16.02 shows that the Accrued Benefits Method was not intended to be used there. Section 18.01 states that in the event of a termination, "the interest of affected Participants in the retirement benefits accrued under the Program to the date of termination or partial termination, to the extent then funded, shall be nonforfeitable." Compared with the language of Section 16.02, which refers to "that portion of the *228 Trust Fund which is held on behalf of Participants of the withdrawing system," § 18 embodies a clear description of the Accrued Benefits Method. But § 18.01 deals with the termination or partial termination of a program, rather than a withdrawal as occurred in this case. The § 16 expression "that portion of the Trust Fund which is held on behalf of Participants" is considerably more ambiguous, and would, as an ambiguity, be construed against the drafter according to traditional contract principles. But R & S's actions are to be reviewed under the arbitrary and capricious standard, rather than as a contract subject to de novo review. Since R & S's discretion is specifically provided for in § 17 of the plan, its decision to interpret the ambiguous language of § 16 differently from the clear language of § 18 will be disturbed only if it is unreasonable. For purposes of this portion of the Carlough test, it is sufficient for the Court to find that the Accrued Benefits Method is consistent with the language of the plan. The Court finds that it is consistent, since the language "on behalf of participants" refers to Oglethorpe's employees, and under the Accrued Benefits Method they receive all accrued benefits held on their behalf. It is not inconsistent with the language of § 16.02 for "held on behalf of" to mean "accrued to," so long as the Accrued Benefits Method is a reasonable reading of § 16.02 per se from an actuarial viewpoint. (See Part (3) below). For the reasons stated below, the Court concludes that the Accrued Benefits Method is a reasonable interpretation of § 16.02 per se, and is therefore an acceptable resolution of the ambiguity of that language when contrasted with the specificity of § 18.
(2) Whether the Accrued Benefits Method is Consistent with the Purposes of the Plan.
Oglethorpe has argued that the Accrued Benefits Method, while it may be acceptable when a plan is 100% funded, is not a fair method to calculate withdrawals when a fund contains surplus, as R & S did. In Foltz, the Court of Appeals upheld the trial court's finding that the fund had not been arbitrary or capricious because the disputed valuation decisions were consistent with the "exclusive pursuit of pecuniary advantages for plan beneficiaries," despite the fact that the actual result of the defendant's valuation decisions tended "to favor a rolling class of future beneficiaries over those present and past." 865 F.2d at 374. The arbitrary and capricious standard allows plan fiduciaries the discretion to balance competing interests of present and future claimants while maximizing assets to the aggregate advantage of all beneficiaries. See Fink v. National Savings and Trust Co., 772 F.2d 951, 955-56 (D.C. Cir.1985).
R & S argues further that the fiduciary duty it owed was not to Oglethorpe, but rather to Oglethorpe's employees, the ERISA "participants" or beneficiaries of the plan. § 404(1). The Accrued Benefits Method protects Oglethorpe's employees, as required by ERISA, and does so arguably at lower cost and with a substantial disincentive to early withdrawals which might affect the future viability of the R & S Program as a whole. Under the Accrued Benefits Method, Oglethorpe's employees will be refunded 100% of their accrued benefits. The money left in R & S's surplus accounts will be available to benefit a class of future beneficiaries.[1]
*229 (3) Whether the Accrued Benefits Method is a Reasonable Interpretation of § 16.02.
The Foltz test requires the Court to determine if the method selected adequately implements § 16.02. In this regard, both experts testified that an acceptable actuarial method for implementing § 16.02 would have certain characteristics, including:
1. Satisfying the requirements of ERISA;
2. Consistent application to all withdrawing systems;
3. Cost efficiency in adminstration;
4. Encouraging systems to remain in the program;
5. Producing a sum certain in case of withdrawal.
In addition, the independent actuarial study commissioned by the parties, and performed by Alexander & Alexander, concluded that the Entry Age Normal Method was not a desirable way to implement § 16.02, and should "not be used under any circumstances." Ex. 40 at 22. Plaintiff's expert testified that the complexity of the Entry Age Normal Method would make it more expensive to administer, and his testimony was not contradicted by Oglethorpe's expert. Given this evidence, the Court concludes that the Accrued Benefits Method was at least as good a choice from the standpoint of implementing § 16.02 as the Entry Age Normal Method would have been. R & S's expert witness, Mr. Cohen, admitted at trial that § 16.02 could cover "any computation under a reasonable recognized actuarial method" (Transcript at 69) and that neither method of calculation was precluded by the language of the plan. (Transcript at 72). If, as the plaintiff's expert admits, the language of § 16 does not preclude the use of the Entry Age Normal Method, while not specifically excluding either that method or the Accrued Benefits Method, the employment of the Accrued Benefits Method appears to be a reasonable one.
Oglethorpe presented no evidence at trial that the R & S interpretation of § 16.02 was contrary to its language, but has made the equitable claim that the existence of considerable surplus in the plan at the time it withdrew made it unreasonable for R & S to calculate its assets solely on the basis of what was needed to cover accrued benefits to Oglethorpe's employees. However, Oglethrope's expert, Mr. Stratton, acknowledged that the Accrued Benefits Method was a permissible interpretation of the section, both when the trust was 100% funded and when there were surplus assets. In this regard, the D.C. Circuit has held that an administrator is not obliged to make decisions on a case-by-case basis reflecting funding levels at the time of the decision. In Stewart v. National Shopmen Pension Fund, 795 F.2d 1079 (D.C.Cir.1986) the court held that if a policy is reasonably designed to promote the financial security of a plan, it could be followed consistently even if deviation from a policy would not undermine actuarial soundness in a given case:
[T]he fund is not obliged to show that its policy's application to these facts is justified by the need to protect the actuarial soundness of the Fund.... A trustee's fiduciary duty, after all, extends to all a fund's participants. And failure to abide by [the policy] would amount to granting the appellants a preference not accorded to others. In truth, to accept appelants' argument would be to turn our standard of review on its head and require the Fund to act capriciously.
Id. at 1083-84 (emphasis in original).
Therefore, the Court agrees with Mr. Cohen that the level of funding at the time Oglethorpe withdrew should not have affected R & S's decision to apply the Accrued Benefits Method to Oglethorpe as it had to the 21 systems which had withdrawn *230 previously. Since Oglethorpe has conceded that the Accrued Benefits Method is a reasonable method when some surplus assets exist, the Administrator was not required, under Stewart v. National Shopmen Pension Fund, to alter his procedure when assets appreciated, even if the result appears harsh.
Bearing in mind that both actuarial experts included ease of administration, consistency and cost-effectiveness among the goals of a reasonable method of calculation, it would seem appropriate to stick to a given method once it has been chosen, even if the result might work against an individual employer, so long as the employees are protected as required by ERISA. The Court concludes that refusal to distribute surplus assets at a time when the funding ratio was 174% was not arbitrary or capricious.
(4) Consistency of Application and Notice to Oglethorpe.
Up to the time Oglethorpe withdrew, the Accrued Benefits Method had been applied consistently to the 21 systems which had withdrawn from the Program.
The substantial early withdrawal penalty created by use of the Accrued Benefits Method discourages withdrawal from the program, to say the least. But this raises the question why, then, did Oglethorpe withdraw? In response, Oglethorpe has argued that it did not know, at the time, that the amounts "held on behalf of" its employees would be calculated using the Accrued Benefits Method. Although Oglethorpe did receive a letter from R & S which explained the withdrawal process, Oglethorpe contends that the letter was ambiguous and vague. Problems with ambiguity of notice do incline in Oglethorpe's favor under the Foltz test, but it is only one of the Foltz factors and is not sufficient to overcome the weight of evidence that the Accrued Benefits Method is reasonable and was consistently applied.
V. Conclusion.
For the foregoing reasons, the Court concludes that selection of the actuarial method for calculating amounts transferable to withdrawing systems under § 16.02 is committed to the discretion of the plan administrator, and should, therefore, be reviewed under the arbitrary and capricious standard. The selection of the Accrued Benefits Method by R & S is a reasonable implementation of § 16.02 because it is actuarially sound, and meets the requirements for proper administration of an ERISA fund as set forth in the case law. Use of the Accrued Benefits Method is neither arbitrary nor capricious. Accordingly, a declaratory judgment is entered to the effect that R & S's use of the Accrued Benefits Method is approved.
ORDER
In accordance with the Memorandum of today's date, declaratory judgment is entered for the plaintiff.
SO ORDERED.
NOTES
[1] In effect, this treatment of Oglethorpe acts as a substantial early withdrawal penalty on the Oglethorpe's plan participants. From the perspective of the R & S Program, as from the perspective of the retirement funds of which it is an aggregate, it is legitimate to penalize withdrawals to the degree necessary to deter withdrawls to a reasonable extent. Both sides' experts agreed on the legitimacy of that goal. Just how big a withdrawal penalty would have to be to deter withdrawals efficiently is difficult to calculate, but the level of forfeiture exacted by R & S in this case seems to the Court to be at the borderline of reasonableness, and may have exceeded what would be necessary to insure the stability of R & S's future. Indeed, R & S itself must think so, since it has since changed its method of calculating withdrawals in the direction suggested by Oglethorpe, i.e., towards splitting up the surplus amounts among withdrawing employers. It is unlikely that R & S would have done so if the Accrued Benefits Method were the only method capable of stemming a deleterious rate of withdrawal. However, the selection of the Accrued Benefits Method did serve the purpose of the plan at the time Oglethorpe withdrew, since it preserved the benefits of the plan's beneficiaries at the accrued level, thereby satsifying § 404, while being more cost-effective for the plan to administer, as opposed to the Entry Age Normal Method which would, according to the independent study, be inefficient. It also satisfies the criterion identified at trial of producing a sum certain upon withdrawal, while Mr. Stratton showed some difficulty in explaining how the Entry Age Normal Method would work in practice.