Alexander v. Brigham & Women's Physicians Organization, Inc.

513 F.3d 37 | 1st Cir. | 2008

513 F.3d 37 (2008)

Eben ALEXANDER III, M.D., Plaintiff, Appellant,
v.
BRIGHAM AND WOMEN'S PHYSICIANS ORGANIZATION, INC., et al., Defendants, Appellees.

No. 07-1443.

United States Court of Appeals, First Circuit.

Heard December 5, 2007. Decided January 23, 2008.

Michael Paris, with whom Colleen C. Cook and Nystrom Beckman & Paris LLP were on brief, for appellant.

David C. Casey, with whom Littler Mendelson, PC was on brief, for appellees.

Before BOUDIN, Chief Judge, SELYA, Senior Circuit Judge, and GELPI,^[*] District Judge.

SELYA, Senior Circuit Judge.

This appeal implicates the Employee Retirement Income Security Act of 1974 (ERISA), 29 U.S.C. §§ 1001-1461. It presents two issues of first impression concerning the scope of ERISA's exemption for so-called top-hat deferred compensation plans. See id. § 1051(2). After careful consideration, we conclude—as did the district court—that the plans at issue are valid top-hat plans. In the course of our analysis, we reject the appellant's claims (i) that the plans cater to more than a select group of highly compensated employees and (ii) that the applicability of the top-hat exemption hinges on an implicit requirement that affected employees possess individual bargaining power.

I. BACKGROUND

We set forth here only those (essentially uncontroverted) facts that are necessary to place this appeal into perspective.^[1] We refer readers who hunger for more detail to the district court's exegetic opinion. See Alexander v. Brigham & Women's Physicians Org., 467 F.Supp.2d 136, 137-41 (D.Mass.2006).

In 1988, plaintiff-appellant Eben Alexander III began working for the Brigham Surgical Group Foundation (BSG), a hospital-based physicians' organization that employed a host of surgeons who, like the appellant, doubled in brass as Harvard Medical School (Harvard) faculty members. In January of 2001, BSG morphed into a new organizational structure known as Brigham & Women's Physicians Organization. As this corporate shuffle has no bearing on the claims before us, we refer throughout to BSG as the employer of record.

Due to their academic affiliation, the members of BSG's full-time surgical complement were subject to special Harvard-imposed salary caps. Over time, these caps began to chafe: BSG found that they hindered its ability to recruit and retain top-flight surgeons (who could earn substantially more in private practice). To remove this impediment yet still remain compliant with Harvard's wishes, BSG created two deferred compensation plans: the Faculty Retirement Benefit Plan (FRBP) and the Unfunded Deferred Compensation Plan (UDCP). By allocating portions of a surgeon's "excess" earnings to these unfunded accounts, BSG expected that it would make employment at the hospital more attractive to surgeons.

We pause at this juncture to sketch the architecture of the plans. The critical datum under each plan is a given surgeon's net practice income (NPI), that is, the net of payments attributable to the services that he rendered less BSG's costs allocable to those services. In any year in which a surgeon does not have NPI equal to or greater than his base salary, he incurs an obligation to repay BSG for the deficit (either out of future NPI or out of pocket). In a rosier scenario—when and if a surgeon's NPI exceeds Harvard's earnings cap—the excess (up to 25% of the surgeon's salary) will be credited to his account in the FRBP.^[2] If any surplus NPI remains (that is, if the excess of NPI over the salary cap is greater than the amount consigned to the FRBP), 50% of that surplus will be credited to the surgeon's account in the UDCP and the remainder will escheat to BSG. Should a surgeon produce a negative NPI in any year, the per annum deficit will be carried forward and debited against positive balances in his FRBP and UDCP accounts.

At the plans' inception, BSG lacked a clear indication as to the number of surgeons who might earn enough to produce contributions to the FRBP.^[3] Over time, however, it became evident that only a small fraction of the surgical complement would achieve that distinction. The district court determined that the relevant years for purposes of this case were 1997, 1998, and 1999, see id. at 140, and the parties have acquiesced in that configuration. The figures for those years illustrate the "small fraction" trend. All surgeons were potentially eligible to contribute to the plans; the key was whether a given surgeon generated sufficient NPI. The surgeons en masse constituted 32.4%, 30.7%, and 27.2%, respectively, of BSG's aggregate workforce. However, the roster of surgeons who actually achieved the requisite income level(s) and thus contributed money into one or both of the plans was significantly smaller. In 1997, 8.7% of BSG's overall employee population contributed to the FRBP and 5.8% to the UDCP; in 1998, the figures were 6.2% and 3.3%, respectively; and in 1999, the figures were 4.9% and 3.1%, respectively.

A glance at the average income of the plan contributors conveys the magnitude of the pecuniary cleft between them and the employee population as a whole. During the three years in question, BSG employees as a whole averaged annual earnings of $83,403 (1997), $80,491 (1998), and $74,376 (1999). Meanwhile, the FRBP contributors earned on average $434,840 (1997), $476,024 (1998), and $418,059 (1999). The UDCP contributors were even more richly compensated; they earned on average $503,730 (1997), $581,320 (1998), and $483,073 (1999).

The record is pellucid that, upon recruiting the appellant, BSG introduced him to a compendium of fringe benefits. This introduction included a review of the documentation describing the. FRBP and the UDCP and an explanation of the system of credits and debits used in connection therewith.^[4] At that time—and at all times material to this case—the plans covered all full-time BSG surgeons who held Harvard faculty appointments. As written, the plans afforded no option for any surgeon to opt out. And because they were already in place when the appellant, joined BSG, he had no real opportunity to bargain over their terms.

That is not to say that the terms of the plans were set in cement. Of course, the appellant, as an individual' employee, was unable to alter those terms. Like every other surgeon, however, he was a voting member of BSG. As such, standard corporate governance and decisionmaking mechanisms stood available to him. For example, BSG's board, which included surgeons as directors, could amend the plans, subject to the concurrence of the board's compensation committee (which, under BSG's bylaws, has "final authority respecting all compensation arrangements . . . between the Corporation and its . . . Employees"). While such corporate structures are not equivalent to direct employee democracy, they are nonetheless meaningful.

Here, moreover, the record indicates that the system responded to the surgeons and their wishes. As to plan amendments, the reviews are mixed: on one occasion, BSG's executive committee rejected a proposal to amend the plans, but on a different occasion (in 1990) a proposal to revise the terms of the plans so as to make them available to Harvard faculty members of all ranks was adopted.^[5]

With this backdrop in place, we turn to the genesis of the litigation. In 2001, BSG terminated the appellant's employment. Simultaneously, it notified him that he was running a cumulative NPI deficit and that, therefore, his FRBP and UDCP accounts would' be debited by more than $400,000 to offset that deficit. The appellant protested, but to no avail.

After his ouster, the appellant sued BSG and related defendants in the United States District Court for the District of Massachusetts, asserting claims under both federal and state law. The district court deemed the state-law claims preempted, and those claims are not before us. We focus, then, on the, appellant's federal claims.

In brief, those claims allege that BSG's sponsorship and administration of the FRBP and UDCP violated ERISA's vesting and fiduciary duty requirements. The appellant seeks money damages and ancillary relief (including attorneys' fees). BSG—we use this acronym throughout this opinion as; a shorthand for all named defendants—replies that the plans, although within the realm of ERISA, were top-hat plans and thus exempt from the enumerated requirements.

After prolonged discovery and the denial of cross-motions for summary judgment, the district court conducted a bench trial. It found that the plans were maintained primarily for the purpose of providing deferred. compensation; that the surgeons as a group possessed sufficient bargaining power to alter the terms of the plans; and that the absence of individual bargaining power was irrelevant. See id. at 142, 145, 147. Having arrived at these findings, the court ruled that the plans came within the top-hat exemption and that, therefore, ERISA's vesting and fiduciary duty requirements were inapposite. See id. at 148. The court entered judgment accordingly, and this timely appeal ensued.

II. ANALYSIS

BSG crafted the two deferred compensation plans at issue here to take advantage of ERISA's top-hat provision; which applies to any "plan which is unfunded and is maintained by an employer primarily for the purpose of providing deferred compensation for a select group of management or highly compensated employees." 29 U.S.C. § 1051(2). Not surprisingly, then, the questions before us revolve around the proper scope of that exemption.

In examining these questions, we begin by memorializing the applicable standard of review. We then orient our inquiry in terms of statutory purpose. Finally, we consider the appellant's dual assignments of error one by one and explain why we find them unconvincing.

A. Standard of Review.

It is common ground that deference is due to findings of fact made by a trial court following a bench trial. See Sierra Fria Corp. v. Donald J. Evans, P.C., 127 F.3d 175, 181 (1st Cir.1997); Fed. R.Civ.P. 52(a). Here, however, the appellant does not challenge the district court's factual findings; rather, he trains his fire on certain of the court's legal conclusions. First, he calumnizes the court's determination as to which subpopulation of employees should be considered in deciding whether a deferred compensation plan meets the top-hat provision's "select group" requirement. Second, he argues that the district court erred as a matter of law in failing to recognize that the statute implicitly requires that every individual in the select group possess bargaining power sufficient to negotiate the terms of the plans. Because these asseverations demand that we ascertain what the statute requires as a general matter,^[6] they engender de novo review. See Harvey v. Johanns, 494 F.3d 237, 240 (1st Cir.2007); In re 229 Main St. Ltd. P'ship, 262 F.3d 1, 3 (1st Cir.2001).

B. Statutory Purpose.

ERISA is a comprehensive statutory scheme enacted in response to escalating concerns about "the mismanagement of funds accumulated to finance employee benefits and the failure to pay employees benefits from accumulated funds." Massachusetts v. Morash, 490 U.S. 107, 115, 109 S.Ct. 1668, 104 L.Ed.2d 98 (1989);. see Degnan v. Publicker Indus., Inc., 83 F.3d 27, 30 (1st Cir.1996) ("ERISA is a remedial statute designed to fashion anodynes that protect the interests of plan participants and beneficiaries."). ERISA's substantive protections are designed largely "to safeguard the financial integrity of employee benefit funds, to permit employee monitoring of earmarked assets, and to ensure that employers' promises are kept." Belanger v. Wyman-Gordon Co., 71 F.3d 451, 454 (1st Cir.1995).

The top-hat provision cuts a swath through this regulatory thicket, relieving employers of many of ERISA's more onerous burdens if—and only if—certain circumstances exist. To reach that safe harbor, a plan must be "unfunded" and "maintained by an employer primarily for the purpose of providing deferred compensation for a select group of management or highly compensated employees." 29 U.S.C. § 1051(2). When those preconditions are satisfied, the plan is exempted from several of ERISA's stringencies, including rules governing plan participation, vesting, funding, and fiduciary duty. See Hampers v. W.R. Grace & Co., 202 F.3d 44, 46 n. 3 (1st Cir.2000); see also Cogan v. Phoenix Life Ins. Co., 310 F.3d 238, 242 (1st Cir.2002). Reporting and disclosure requirements remain applicable. See 29 U.S.C. §§ 1021-1031; Hampers, 202 F.3d at 46 n. 3.

The origins of the top-hat provision lie in Congress's insight that high-echelon employees, unlike their rank-and-file counterparts, are capable of protecting their own pension interests. See Gallione v. Flaherty, 70 F.3d 724, 727 (2d Cir.1995); see also Dep't of Labor Op. No. 90-14A (1990). Presuming that employees of this stature can fend for themselves, Congress relaxed some of ERISA's prophylactic obligations.

This congressional purpose is a reliable guidepost for the task of statutory interpretation. See Beckley Capital Ltd. P'ship v. DiGeronimo, 184 F.3d 52, 57 (1st Cir. 1999). Beyond that point, however, we possess few if any interpretive aids in discerning the precise scope of the top-hat provision. Thus, as a matter of both necessity and preferred practice, see, e.g., United States v. Meade, 175 F.3d 215, 219 (1st Cir.1999), we turn to the statutory text We add only that neither party has directed us to any useful legislative history

C. Select Group.

As said, the top-hat provision applies only if, among other things, a plan is maintained for the primary benefit of "a select group of management or highly compensated employees." 29 U.S.C. § 1051(2). The first issue raised by the appellant concerns the methodology used in determining whether a given group i is "select."^[7] He argues that the statute envisions that the entire cohort of employees to whom a plan was offered—here, every full-time BSG surgeon who held a Harvard faculty appointment—comprises the relevant group. Taking this approach, he says that the plans here at issue were maintained for roughly 30% of the workforce (hardly a "select" group).

The district court rejected this argument. It held that the relevant inquiry involved only those surgeons who in fact crossed the NPI threshold and contributed money to the plans. In so holding, the court concluded that "[a]s a practical matter" the plans were "maintained" for the surgeons who actually contributed and not for the more sizable group of surgeons who nominally were covered by the language of the plan documents. Alexander, 467 F.Supp.2d at 144. Operating from this coign of vantage, the court found that the FRBP never was maintained for more than 8.7% of the relevant universe and that the UDCP never was maintained for more than 5.8% of the relevant universe. Id. So viewed, both groups were "select." Id.

The lower court's rationale fits readily within an unforced reading of the statutory text. After all, a substantial precondition—the accumulation of NPI in excess of the salary cap—must be satisfied before any surgeon can squirrel away money in either of the plans. That gets the grease from the goose: owing to the height of the threshold that contributors need to surmount, only a relatively small fraction of BSG surgeons contributed to either plan in the relevant years. Common sense suggests that it is this small fraction that should be taken into account in determining the composition of the select group for, whom the plans were maintained.

We are fortified in this impression about the dimensions of the select group by the flanking statutory phrases: "maintained" and "highly compensated." In view of Congress's linkage of these terms, it would seem very strange to say that the plans were being "maintained by an employer primarily for the purpose of providing deferred compensation" to employees who were not highly compensated, who did not cross the income-generating threshold needed to become plan contributors, and who might never be able to do so.

The more natural and sensible reading of the statute is that a plan is "maintained" for a group of employees only if those employees realistically have the capacity to benefit from it. Thus, for the purpose of determining whether a plan services a select group, we find relevant here only those employees who actually earned enough NPI to fund plan contributions. They are the employees for whom the money in the plans is being managed and held; they are thus the employees whom Congress had in mind when it devised the contours of the top-hat carve-out.

This conclusion is not at odds with the case law. The other courts that have spoken to this issue confronted categorically different situations—either situations in which employers or employees retained the discretion to contribute to the deferred compensation plans when and if they chose or situations in which every member of a designated group of employees received plan contributions albeit in differing amounts depending on some variable (say, the incidence of commissions). See, e.g., Demery v. Extebank Deferred. Comp. Plan (B), 216 F.3d 283, 285 (2d Cir.2000); Guiragoss v. Khoury, 444 F.Supp.2d 649, 654 n. 4 (E.D.Va.2006); Carrabba v. Randalls Food Mkts., Inc., 38 F.Supp.2d 468, 471 (N.D.Tex.1999). Thus, unlike this case, either every employee within the designated group received some benefit from the plan or someone possessed the ability to fund the plan as a matter of discretion (so that, crucially, the ability to fund the plans was always in existence).

The BSG plans are quite different surgeons have no automatic entitlement to deferred compensation. Rather, any particular surgeon can contribute to the plans if—and only if—he or she generates sufficient NPI to surpass the salary cap. The surgeons who succeed in that endeavor can and do contribute; those who fail have no option to deposit money into the plans. Under these circumstances, it would be curious—and, ultimately, incorrect—to say that those plans were "maintained by an employer primarily for the purpose of providing deferred compensation" to the latter cadre of surgeons, that is, surgeons incapable of contributing to the plans.

The appellant has a fallback position. He argues, in the alternative, that ERISA's statutory definition of "participant" should define the relevant employee universe. Under that definition, "participants" include, inter alia, any employee "who is or may become eligible to receive a benefit of any type from an employee benefit plan." 29 U.S.C. § 1002(7). The appellant contends that, applying this definition, all the BSG surgeons should be included as part of the "select group" because each one of them signed the same plan documents and therefore might, if his or her earnings soared, "become eligible" to receive plan benefits.

This contention lacks force because the insertion of the term "participant" is made of whole cloth. The top-hat provision does not refer to the term "participant" at all, nor does any legislative history suggest that Congress intended to import that defined term into the top-hat provision. Indeed, the affirmative use of the term "participants" in a distinct provision of section 1051, see 29 U.S.C. § 1051(3)(A), strongly implies that the definition does not apply sub silentio throughout the statute. See Duncan v. Walker, 533 U.S. 167, 173, 121 S.Ct. 2120, 150 L.Ed.2d 251 (2001) (explaining that "[w]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion and exclusion") (citation and internal, quotation marks omitted); United States v. Nason, 269 F.3d 10, 16-17 (1st Cir.2001) (similar).

The appellant's argument boils down to the notion that, wherever ERISA mentions plans and employees, the drafters intended that reference should be made to the statutory definition of "participants." This notion is unfounded. The specific language used in section 1051(2)'s "participant" definition clearly encompasses employees who have not yet and may never become eligible for plan participation—yet the language of the top-hat. exemption does not countenance the prospect of mere future eligibility. One does not "maintain[]" a plan for employees who are not yet eligible for it.

The fact that the plan documents for the FRBP refer to the surgeons as "[p]articipant(s)" does not undermine this conclusion. While a plan's specific language can aid a court in determining whether that plan qualifies as a top-hat plan, see, e.g., Bakri v. Venture Mfg. Co., 473 F.3d 677, 678 (6th Cir.2007), a scrivener's choice of plan language cannot be allowed to control the more general interpretive inquiry into what the statute requires.

For these reasons, we reject the appellant's importuning that we broaden the "select group" analysis by engrafting onto the top-hat provision the unrelated statutory definition of "participant."^[8] While ERISA, as a broad remedial statute, should be liberally construed in favor of coverage, Kross v. W. Elec. Co., 701 F.2d 1238, 1242 (7th Cir.1983), that principle does not permit us to override the clear text of the statute.

The itinerary for this leg of our voyage lists one more port of call. Once we have determined the relevant subpopulation of employees, it remains to be seen whether those employees constitute a "select group of . . . highly compensated employees." In this instance, the answer is obvious.

The status analysis turns on both qualitative and quantitative dimensions. See Senior Exec. Ben. Plan Participants v. New Valley Corp., 89 F.3d 143, 148 (3d Cir.1996). As the highest-earning surgeons at BSG, the employees who contributed to the plans are qualitatively select; and because they comprise no more than 8.7% of BSG's workforce, the group is quantitatively select as well. See, e.g., Demery, 216 F.3d at 289 (describing 15.34% of the workforce as within "the acceptable size for a `select group'").

The question of whether the relevant universe of employees is "highly compensated" is even more open-and-shut. To come within the compass of the top-hat provision, the employer must be able to show a substantial disparity between the compensation paid to members of the tophat group and the compensation paid to all other workers. See Simpson v. Ernst & Young, 879 F.Supp. 802, 816 (S.D.Ohio 1994). Over the three years in question, the average income of contributors to the FRBP was roughly $440,000—more than five times the average income of BSG employees as a whole. During the same time frame, the average income of the UDCP contributors was even greater (and, thus, the gulf was even wider). Although we acknowledge that in some cases it will be difficult to determine the exact boundaries of what constitutes "high" compensation within the purview of the top-hat provision, the case at hand is nowhere near the gray area; the contributors to BSG's plans were highly compensated in both relative and absolute terms. The appellant's first assignment of error therefore fails.

D. Bargaining Power.

We turn now to the appellant's second argument. The district court found that, when considered collectively, both the plan contributors and the general population of teaching surgeons possessed appreciable power to influence the terms of their deferred compensation arrangements. Alexander, 467 F.Supp.2d at 147. Relatedly, the court found that the appellant, as an individual, lacked that power. Id. at 140.

The appellant asserts that the court erred in concluding that his lack of individual bargaining power was irrelevant to the top-hat determination. In his view, every member of the select group, for whose benefit a top-hat plan exists must possess bargaining power sufficient to influence the terms of the plan; elsewise, the plan cannot qualify for the exemption.

Once again, we look first to the text of the top-hat provision. In that proviso, Congress nowhere mentioned bargaining power. Indeed, the statutory language contains no indication that Congress contemplated that courts would consider employees' ability to bargain over the terms of their deferred compensation plans, either individually or collectively, when measuring the bona fides of a select group and determining the applicability of the. top-hat provision.

What authority exists for the appellant's position derives from a Department of Labor (DOL) opinion letter purporting to shed light on Congress's reasons for enacting the top-hat provision. In pertinent part, that letter states:

It is the view of the Department that in providing relief for "top-hat" plans from the broad remedial provisions of ERISA, Congress recognized that certain individuals, by virtue of their position or compensation level, have the ability to affect or substantially influence, through negotiation or otherwise, the design and operation of their deferred compensation plan, taking into consideration any risks attendant thereto, and, therefore, would not need the substantive rights and protections of Title I [of ERISA].

DOL Op. No. 90-14A.

The appellant leans heavily on the reasoning of this letter and chronicles a number of cases in which courts have cited the letter when discussing the individualized bargaining power of top-hat plan contributors. See, e.g., Duggan v. Hobbs, 99 F.3d 307, 312-13 (9th Cir.1996); Guiragoss, 444 F.Supp.2d at 658-59, 664; Carrabba, 38 F.Supp.2d at 477-78; see also Prior v. Innovative Comm. Corp., 360 F.Supp.2d 704, 713-14 (D.Vi.2005) (relying on Duggan). Insofar as these cases adumbrate a hard-and-fast requirement that individual top-hat plan beneficiaries must have bargaining power, they all ultimately derive that requirement from the DOL opinion letter.

We decline the appellant's invitation to depart from the plain language of the statute and jerry-build onto it a requirement of individual bargaining power. The DOL opinion letter speaks only to Congress's rationale for enacting the top-hat provision. It does not present itself as an interpretation of the provision's requirements, nor does it make any mention of the need for or propriety of demanding that employers demonstrate their employees' ability to negotiate the terms of deferred compensation plans.

Agency opinion letters command deference only to the extent that they possess the power to persuade. See Kalda v. Sioux Valley Physician Partners, Inc., 481 F.3d 639, 647 (8th Cir.2007); Johnson v. Watts Regulator Co., 63 F.3d 1129, 1134 n. 2 (1st Cir.1995). The DOL opinion letter here at issue falls within this taxonomy.

We have no quarrel with the letter's persuasiveness as a gloss on Congress's intentions in enacting the top-hat provision. But relying on that letter to justify a nascent requirement that every employee covered by a" top-hat plan possess the power to negotiate the terms of that plan is simply too much of a stretch. To our way of thinking, such a reading is both unwarranted and unpersuasive. Even without the additional requirement of individual bargaining power, Congress's enactment strikes us as a reasonable effectuation of its purpose. In any event, in limiting the top-hat provision to a "select group of management or highly compensated, employees," Congress ensured that employees' interests would be sufficiently protected.

If more were needed—and we doubt that it is—we are further counseled against the importation of a requirement of individual bargaining power by the bizarre consequences that would follow from it. Most important, that thesis implies that every top-hat plan can be rendered noncompliant by demonstrating that a single covered employee lacks individual bargaining power, no matter the overall characteristics of the "select group of management or highly compensated employees" to which he belongs.

Such an absolutist construction clashes with the essential nature of the top-hat provision, which has been interpreted more generally to mean that not every member of the select group need belong to the upper tier of management or fit within the highest stratum of compensation. See, e.g., Demery, 216 F.3d at 289; Guiragoss, 444 F.Supp.2d at 663-64; Belka v. Rowe Furniture Corp., 571 F.Supp. 1249, 1252-53 (D.Md.1983). These cases recognize the sensible proposition that it is the configuration of the group as a whole that controls.^[9] If, as the appellant suggests, Congress was singularly concerned with an individual's ability to fend for himself or herself, we think it unlikely that Congress would have framed the statute in terms of "groups" at all.

We add a coda. As said, the district court determined that the surgeons, as a group, enjoyed bargaining power. See Alexander, 467 F.Supp.2d at 147. Because neither party has either challenged that determination or questioned its necessity, we have assumed for argument's sake that collective bargaining power might conceivably be a prerequisite for a top-hat plan. We think it is wise, however, to note our grave doubts about the correctness of this assumption.

The two main reasons underpinning our holding that there is no requirement of individual bargaining power to qualify for the top-hat provision—(i) that neither the text nor the legislative history of the statute contains the slightest hint that Congress contemplated that courts would consider employees' ability to bargain over the terms of their deferred compensation plans and (ii) that the DOL opinion letter does not presume to interpret the statute—militate just as strongly against importing a requirement of collective bar gaining power into the top-hat provision. Although Congress's rationale for fashioning the exemption was that the members of a "select group of management or highly compensated employees" could fend for themselves, the statute, by its terms, does not purport to require proof of power of any sort. In such, circumstances, it would be highly unorthodox to convert a rationale (like the one set forth in the DOL opinion letter) into an independent statutory test.

III. CONCLUSION

For the reasons elucidated above, we uphold the judgment below. Consequently, the appellant's prayer for attorneys' fees under 29 U.S.C. § 1132(g) must be denied as moot.

Affirmed.

NOTES

[*] Of the District of Puerto Rico, sitting by designation.

[1] Even though the parties dispute certain facts, none of those disputes is relevant to our decision.

[2] The lower court's description of this aspect of the plan, see Alexander, 467 F.Supp.2d at 139, albeit stipulated to by the parties, is an inexact paraphrase. Although the stipulation varies in some respects from the actual language of the FRBP, the discrepancies are not material to our decision.

[3] There is, however, some evidence that, from the beginning, BSG anticipated that no more than 10% of its aggregate workforce would contribute to the UDCP.

[4] We note that, during the course of his career with BSG, the appellant's performance resulted in both positive and negative NPI figures, leading to a series of credits and debits to his deferred compensation plan, accounts.

[5] The appellant suggests that the infrequency of amendment is emblematic of a lack of bargaining power. That is sheer speculation; that circumstance may well have been due to widespread satisfaction with the plans as they stood.

[6] It is beyond cavil that the plans at issue here satisfy the other two statutory criteria: they are unfunded and maintained by BSG primarily to provide a deferred compensation benefit.

[7] It is an open question whether the statutory phrase "select group" modifies only the term "management" or also modifies the term "highly compensated employees." For the purposes at hand, we assume arguendo, without deciding, that both categories of employees are subject to the "select group" requirement.

[8] This holding defeats the appellant's reliance on Darden v. Nationwide Mut. Ins. Co., 717 F.Supp. 388 (E.D.N.C.1989), aff'd, 922 F.2d 203 (4th Cir.1991), rev'd on other grounds, 503 U.S. 318, 112 S.Ct. 1344, 117 L.Ed.2d 581 (1992). Darden rested squarely on an erroneous holding that the "participant" definition should be transplanted root and branch into the top-hat exemption. See id. at 396-97.

[9] None of the cases cited by the appellant purports to hold that when an otherwise valid top-hat plan is offered to a "select group of management or highly compensated employees," that plan will lose its exempt status if any one of the covered employees lacks individual bargaining power. To the extent that any of those cases rely on an erroneous interpretation of the DOL opinion letter, we respectfully decline to follow that interpretation.