Opinion
No. 94 C 5916.
December 11, 1995.
Miriam N. Geraghty, James R. Potter, Jesse Andrew Wing, Kinoy, Taren, Geraghty Potter, Chicago, IL, for plaintiffs.
Wilber H. Boies, Nancy G. Ross, McDermott, Will Emery, Chicago, IL, for defendants.
MEMORANDUM OPINION AND ORDER
Five former employees of Amoco Corporation ("Amoco") have brought this class action against the Employee Retirement Plan of Amoco Corporation and Participating Companies ("Plan") and Amoco (in its capacity as Plan Administrator) under Employee Retirement Income Security Act ("ERISA," which comprises 29 U.S.C. § 1001-1368) §§ 1132(a)(1)(B) and 1132(a)(3). On behalf of themselves and of all other Amoco ex-employees who have received pension benefits under the Plan, plaintiffs seek to force Amoco to recalculate their pensions to include pay in lieu of vacation within the "earnings" component of the calculation.
For convenience both defendants will be referred to collectively as Amoco, because it is Amoco's interpretation of the Plan (as Plan Administrator) that is at issue.
Both sides have now moved for summary judgment under Fed.R.Civ.P. ("Rule") 56. They have briefed the motions and have complied with this District Court's Rule ("GR") 12(M) and 12(N), so that the cross-motions are ready for decision. For the reasons set forth in this memorandum opinion and order, plaintiffs' motion is granted and Amoco's motion is denied.
This opinion will use "P." and "D." rather than the parties' names in citing to their respective Exhibits and their GR 12(M) and 12(N) statements, which will be identified as "12(M) ¶ ___" or "12(N) ¶ ___." Affidavits and depositions will be identified by the name of the affiant or deponent (e.g., "Brennan Dep. ___").
Summary Judgment Standard
Under familiar Rule 56 analysis, a party seeking summary judgment bears the burden of establishing the lack of a genuine issue of material fact ( Celotex Corp. v. Catrett, 477 U.S. 317, 324, 106 S.Ct. 2548, 2553, 91 L.Ed.2d 265 (1986)). This Court is called upon to draw inferences in the light most favorable to the non-moving party, but it is "not required to draw every conceivable inference from the record — only those inferences that are reasonable" ( Bank Leumi Le-Israel, B.M. v. Lee, 928 F.2d 232, 236 (7th Cir. 1991) and cases cited there). Where as here cross-motions for summary judgment are involved, these principles require the Court to take a dual perspective — one that this Court has frequently described as Janus-like. In this instance that problem does not exist, for the underlying facts are not in dispute — instead the parties are at odds only about whether the term "earnings," when used for purposes of calculating pension benefits, includes pay in lieu of vacation. Thus this case turns on a question of contract interpretation, making it ripe for final disposition at the summary judgment stage ( Bechtold v. Physicians Health Plan of N. Ind., Inc., 19 F.3d 322, 325 (7th Cir. 1994)).
Background
Under the Plan there are three alternative formulas for calculating benefits: the Annuity Benefit Formula, the Career Average Minimum Formula and the Service Minimum Formula. When an employee becomes eligible for a pension, he or she is entitled to a benefit based on the highest of the three calculations.
Unless otherwise noted, all references to the Plan are to its version "As Amended and Restated as of July 1, 1985." Although that version was later superseded by a version "As Amended and Restated Effective as of April 1, 1993," both sides agree that the two versions do not differ in any relevant aspect (P.Mem. 1-2 n. 1; D.Mem. 3 n. 1).
At the center of the present dispute is the Annuity Benefit Formula, which is based on the average annual earnings of an employee-participant during the highest consecutive 3 calendar years out of the last 10 years of employment (Plan § 5.01). Amoco calculates the pension benefit by multiplying an employee's average annual earnings by a constant fraction, then multiplying that figure by the employee's length of participation in the Plan ( id.). While the fraction remains constant, the size of an employee's benefit is thus related in linear fashion to the employee's average earnings figure and length of Plan participation.
1985 Plan § 5.01 refers to "the highest consecutive three calendar years . . . out of the last ten years," with a mechanism for rounding out the year of retirement on a proportional basis to make it the last year in the calculation. 1993 Plan § 2.28 uses the greater of (1) the last 36 months of employment and (2) the highest 3 consecutive calendar years out of the last 10 years. That distinction does not affect the current issue.
Under the Career Average Minimum Formula the benefit is calculated by multiplying a constant fraction by the employee's average annual earnings throughout the employee's career (Plan § 8.02). As with the Annuity Benefit Formula, the size of a participant's pension is directly affected by the average annual earnings amount.
Under the Service Minimum Formula the benefit is calculated by multiplying a constant dollar amount by the number of years of service (Plan § 8.03). Because that calculation is wholly unrelated to earnings, it is not relevant to this action.
Amoco's vacation policy can be described generally as "use it or lose it": Employees must take their vacation time each year or else forfeit it (D. 12(N) ¶ 33; Summary Plan Description ("SPD") Book 1 at 15-16). One critical exception, and the one that gives rise to this lawsuit, is that persons who retire or who otherwise leave the company may receive pay in lieu of earned but unused vacation — but solely for the year in which employment is terminated (all unused vacation time for earlier years having been forfeited (D. 12(N) ¶ 33)). Amoco claims that it gives each retiring employee the option of either taking pay in lieu of vacation or using the vacation time to extend his or her retirement date out beyond the date when he or she actually stops working (Brennan Aff. ¶ 13), but plaintiffs contend that they were not given that option (Complaint ¶ 49; Gallo Aff. ¶¶ 3, 9). At any rate Amoco reserves the right to determine when vacations are taken (SPD Book 1 at 16) and "typically" controls an employee's date of termination (D. 12(N) ¶ 30).
That dispute is not outcome-determinative and hence is not material for Rule 56 purposes.
When employees leaving the company take pay in lieu of vacation, it is Amoco's policy to exclude that pay from earnings for purposes of the Annuity Benefit Formula (D. 12(N) ¶ 24). However, Amoco does use pay in lieu of vacation to calculate earnings for the Career Average Minimum Formula (D. 12(N) ¶ 25). Finally, when a retiring employee instead uses vacation time to extend his or her retirement date beyond the date on which he or she last works, Amoco counts the vacation time in making the "length of participation in the plan" calculation (Gallo Aff.Ex. E).
Plaintiffs contend that pay in lieu of vacation must be considered part of an employee's "earnings" in calculating the pension benefit under the Annuity Benefit Formula. Amoco disagrees, pointing to what it says is a long-standing (but concededly unwritten) interpretation of the Plan that pay in lieu of vacation is not considered as earnings for purposes of the Annuity Benefit Formula (Brennan Aff. 36).
Neither the Plan nor the SPD expressly addresses pay in lieu of vacation. 1985 Plan § 1.23 originally defined "earnings" this way:
"Earnings" means salary, wages, overtime, commissions, bonus payments and other compensation of a similar nature (all to the extent includable under the terms of the plan) received by an employee for his personal services as determined by the Company.
In January 1989 that definition was amended by the Plan's Eighth Amendment, so that revised Plan § 1.23 read:
"Earnings" with respect to any Employee-Participant means total salary, wages and commissions, including forms of base pay delivered in alternate manners such as piecework and payment by mileage for drivers; overtime; shift differentials; and bonuses in the year received, including bonuses in the form of premium pay for services rendered outside of normal working hours or conditions; provided payment is received by an Employee-Participant for personal services. For purposes of this subsection the determination of Earnings shall be made without regard to salary reductions contributions made on behalf of an Employee-Participant to a plan maintained under the Internal Revenue Code of 1986 "Code" Sections 125 and 402(a)(8).
Then 1993 Plan § 2.12 changed the term "earnings" to "compensation" and made other minor changes:
"Compensation" means total salary, wages and commissions, including forms of base pay delivered in alternate manners such as piecework and payment by mileage for drivers; overtime; shift differentials; and annual bonuses in the year awarded and received, including bonuses in the form of premium pay for services rendered outside of normal working hours or conditions; provided payment is received by the Participant for personal services. For purposes of this subsection the determination of Compensation shall include salary reduction contributions made on behalf of a Participant to a plan maintained under sections 125, 129 and 401(k) of the Code.
Compensation in excess of $200,000 (as adjusted for increases in the cost of living pursuant to section 415(d) of the Code) shall not be taken into account for any purposes under the Plan. For Plan years beginning December 31, 1993, $150,000 "shall be substituted for $200,000" in the preceding sentence.
As for the SPD, it refers to "earnings" in this fashion (SPD Book 4 at 51):
The annual earnings used in any calendar year for retirement plan purposes may be different from the earnings reported on Form W-2 for filing your income taxes. This chart shows some of the more common types of income that are and are not considered earnings for the plan's annuity benefit calculation:
Earnings Used for Earnings Not Used Retirement Calculations for Retirement Calculations
• Regular Wages • Moving Expenses • Additional Straight • Educational Assistance Time Pay • Overtime Pay/Bonuses • Taxable Life Awards • Shift Differential • Prizes and Awards • Tax Deferred Savings • Severance Payments (TDS)
Two other employee benefit plans that form part of Amoco's benefits package (also included in SPD Book 4) also bear substantial relevance here: its Employee Savings Plan and its Variable Incentive Plan (colloquially termed its "V.I.P. Bonus Plan"). As for the first of those, it has contained a definition of "earnings" — changed in 1993 to "applicable compensation," in the same way that the Plan then shifted from "earnings" to "compensation" — that has consistently been identical to the corresponding Plan definition (see P.Ex. 10 at 2-3). And the V.I.P. Bonus Plan has defined "pay (earnings)" in essentially equivalent fashion (Brennan Dep. Ex. 10 at 2). Importantly, Amoco has consistently treated and continues to treat employees' pay in lieu of vacation as part of their "earnings" in each of those plans, just as it always has in calculating the pension benefit under the Career Average Minimum Formula. Thus the only exception to that uniform treatment by Amoco has been in its calculation of the Annuity Benefit Formula under the Plan.
All five of the named class representatives either voluntarily terminated their employment (i.e., retired) or were terminated via a reduction in force, and all five received pay in lieu of vacation. On February 21, 1995 this Court certified the plaintiff class to which the parties had stipulated on February 17:
All past, present and future participants in the Employee Retirement Plan of Amoco Corporation and Participating Companies who either (a) have received benefits under the Plan since September 27, 1984 or (b) may be eligible to receive benefits under the Plan in the future; and who received or may elect in the future to receive pay in lieu of vacation accrued as of December 31 of the year prior to termination of employment, and their surviving spouses as plan beneficiaries.
Named plaintiff Matthew Gallo ("Gallo") estimates that if his "earnings" had included the pay that he received for the six weeks of vacation that he had due but did not take — or more accurately that he was not allowed to take because Amoco required him to work without vacation until his retirement date (Gallo Aff. ¶ 3) — that would have increased his lump sum annuity by $40,000 (Gallo Aff. ¶ 14). None of the other named plaintiffs has quantified his losses in that fashion: Each of Thomas E. Gray, Elijah Lewis, Alexander Dumas and Neal Goldberg estimates his losses at "many hundreds of dollars" (Gray Aff. ¶ 11; Lewis Aff. ¶ 11; Dumas Aff. ¶ 10; Goldberg Aff. ¶ 10). Amoco admits that the aggregate claims of all the class members would have a de minimis effect on the Plan's solvency (D. 12(N) ¶ 26).
Standard of Review and Framework of Analysis
Federal common law rules of contract interpretation guide this Court's interpretation of an ERISA-governed plan (see Hickey v. A.E. Staley Mfg., 995 F.2d 1385, 1389 (7th Cir. 1993)). First this Court must determine as a matter of law whether the term at issue is ambiguous or unambiguous ( id.). Ambiguity exists when a term "is subject to reasonable alternative interpretations" ( id., quoting Taylor v. Continental Group Change in Control Severance Pay Plan, 933 F.2d 1227, 1232 (3d Cir. 1991)).
Here the defined term "earnings" is unquestionably ambiguous as to the inclusion or exclusion of pay in lieu of vacation — none of the already-quoted Plan or SPD provisions provides a clear-cut answer to that question. Amoco's own reading of the Plan buttresses that conclusion: It interprets the selfsame term "earnings" both to include pay in lieu of vacation (in the context of the Career Average Minimum Formula) and to exclude it (in the context of the Annuity Benefit Formula). Thus Amoco's argument that "earnings" unambiguously excludes pay in lieu of vacation borders on the frivolous. Nor can this Court accept plaintiffs' counterargument that "earnings" unambiguously includes pay in lieu of vacation.
Before this opinion turns to a resolution of that ambiguity, something should be said about the effect to be given to Amoco's reading. Unlike a typical contract case in which this Court would approach this question de novo, it is well established that "a deferential standard of review is appropriate when the plan administrator is authorized to exercise discretionary power to . . . construe terms of the plan" ( Krawczyk v. Harnischfeger Corp., 41 F.3d 276, 278 (7th Cir. 1994), citing Firestone Tire Rubber Co. v. Bruch, 489 U.S. 101, 115, 109 S.Ct. 948, 956-57, 103 L.Ed.2d 80 (1989)). As P.Mem. 6 concedes, Plan § 15.03 vests the Plan Administrator (here Amoco) with such discretion:
The Plan Administrator shall have such duties and powers as may be necessary in order to comply with ERISA and governmental regulations issued thereunder, and to discharge its duties hereunder including:
* * * * * *
(d) to construe and interpret the Plan, decide questions of eligibility and determine the amount, manner and time of payment of any benefits in accordance with provisions of the Plan, but subject to the authority of the named fiduciary for review of desired claims.
That language reflects the Firestone Tire teaching, as have other plans that have since been held to grant discretion to their administrators (e.g., Krawczyk, 41 F.3d at 278; Halpin v. W.W. Grainger, Inc., 962 F.2d 685, 688 (7th Cir. 1992)), so that a deferential standard of review is appropriate.
Plaintiffs contend otherwise, urging that the Plan Administrator did not exercise any discretion in interpreting "earnings," but instead rubber-stamped the company's decision not to include pay in lieu of vacation. On that score plaintiffs rely exclusively on Soule v. Retirement Income Plan for Salaried Employees of Rexham Corp., 723 F. Supp. 1138 (W.D.N.C. 1989). But this Court need not decide whether to sign on to the Soule approach, under which such a rubber-stamping should be rejected as a failure to exercise the vested discretion, for the record is silent as to how Amoco's decision to exclude vacation time was made, or even as to who made the decision. Where as here each side would obviously have a difficult task in demonstrating that the decision was or was not rubber-stamped, because of the absence of any evidence about how, when or by whom the decision was made (see the discussion below), that burden is more properly placed on plaintiffs because of the discretion vested in the Plan Administrator.
That degree of deference is most commonly referred to by the familiar but amorphous locution that the administrator's reading will be upheld unless it is "arbitrary and capricious." Those words have been fleshed out further in a few of the cases. Pokratz v. Jones Dairy Farm, 771 F.2d 206, 209 (7th Cir. 1985) has said that the standard calls for "the least demanding form of judicial review of administrative action":
Although it is an overstatement to say that a decision is not arbitrary or capricious whenever a court can review the reasons stated for the decision without a loud guffaw, it is not much of an overstatement.Cuddington v. Northern Ind. Pub. Serv. Co., 33 F.3d 813, 817 (7th Cir. 1994) has quoted from earlier case law a less colorful but more helpful standard:
If the trustee makes an informed judgment and articulates an explanation for it that is satisfactory in light of the relevant facts, i.e., one that makes a "rational connection" between the issue to be decided, the evidence in the case, the text under consideration, and the conclusion reached, then the trustee's decision is final.
Amoco's Interpretation
Typically, as reflected in the Cuddington formulation of the arbitrary and capricious standard, a court can look at the factors that influenced the plan administrator's interpretation. Here, however, Amoco has offered nothing whatever as to how or why — or even when or by whom — Amoco's method of calculating "earnings" in the now-critical respect was arrived at. All that has been submitted is that the method is "long-standing" (Brennan Dep. 36) — indeed, it predates the employment of the person (Michael Brennan ("Brennan"), Amoco's Director of Qualified Benefit Plans) who now attempts to set out some purely post-hoc justifications for its use ( id. 37).Amoco's reading is not memorialized in any way other than in a computer matrix that codes pay in lieu of vacation as an item to be excluded from the Annuity Benefit Formula — obviously a mathematical statement of a conclusion, not at all a statement of the reason underpinning that conclusion. Indeed, when Gallo's claim was rejected (thus precipitating this suit) he was given no rationale as to why pay in lieu of vacation is excluded, only a statement that it is (Exs. B and E of P.Ex. 6). As already indicated, Amoco now seeks to buttress what it has done by post-hoc rationalizations from Brennan and by some abstract theorizing in the affidavit of Amoco's expert. As part of this Court's non-computerized matrix, its decision on the arbitrary-and-capricious issue can surely take into account that Amoco seems to be making up the justifications for its treatment as it goes along.
Amoco unpersuasively seeks to place weight on the evidentiary value of the coding of pay in lieu of vacation in the payroll system, as reflecting a uniformly-applied policy decision to exclude pay in lieu of vacation from the earnings calculation. But Amoco prefers to forget that the very same coding is used to include pay in lieu of vacation in calculating the Career Average Minimum Formula. Computers do not yet make policy, and the question remains whether the decision to cause the computer program to be written in that manner was arbitrary and capricious.
Even more surely, plaintiffs persuasively urge that the arbitrariness and capriciousness of Amoco's handling is really best evidenced by the fact that it includes pay in lieu of vacation as part of "earnings" for purposes of applying the Career Average Minimum Formula but not the Annuity Benefit Formula, even though both utilize exactly the same definition of "earnings." And as indicated earlier, Amoco's inclusion of pay in lieu of vacation as part of "earnings" is carried through into the other two facets of Amoco's employee benefit package (its Employee Savings Plan and its V.I.P. Bonus Plan) that contain identical (in the first instance) or essentially equivalent (in the second instance) definitions of "earnings" — only the Annuity Benefit Formula calculation is a maverick. Plaintiffs' position that the calculation now at issue is indeed arbitrary and capricious is particularly powerful in light of the teaching of Egert v. Connecticut Gen. Life Ins. Co., 900 F.2d 1032, 1037 (7th Cir. 1990) that uniformity of interpretation is an important factor to be considered in the arbitrary-and-capricious analysis.
Egert involved the denial of a claim under an ERISA-governed health plan. While the Court of Appeals said that "[i]n the abstract, given the arbitrary and capricious standard of review, it would be reluctant to disturb" defendant's reading of the plan ( 900 F.2d at 1037), it found particularly significant the conflict between other parts of the plan documents and defendant's proffered reading of the plan. In consequence the Court held ( id. at 1038 (citation omitted)):
At the very least, provisions of the CCP [plan summary] — upon which [defendant] relies so heavily to construe the terms of the Plan — are substantially inconsistent and lead to contradictory dispositions of similarly situated claims. . . . [W]e regard [defendant's] denial of [plaintiff's] claim to be arbitrary and capricious.
This case actually provides a stronger example of the principle relied on in Egert: Here Amoco not only gives two different readings to the very same defined term in different sections of the same document, but it has itself chosen to give one of those two readings — the one contended for by plaintiffs — to two other documents containing the selfsame or much the same language. Amoco attempts to counter by arguing that Egert is not on point because, it contends, this is not a case in which the Plan treats similarly situated claims differently. It says that all calculations under the Annuity Benefit Formula exclude pay in lieu of vacation from the calculation, while all calculations under the Career Average Minimum Formula include pay in lieu of vacation.
While Amoco is accurate in what it says in that respect, the point that it makes is really irrelevant to the current inquiry. What is significant is rather that Amoco clearly — and admittedly (D.R.Mem. 18) — interprets "earnings" inconsistently in different parts of the Plan. And that inconsistency undercuts Amoco's claim that its handling does not "lead to contradictory dispositions of similarly situated claims." Two persons who are given pay in lieu of the last year's unused vacation but who have their pensions calculated under different formulas are treated differently: One whose benefit is based on the Career Average Minimum Formula will have such pay included in the calculation, while one whose benefit is calculated under the Annuity Benefit Formula will not.
If what was before this Court were simply Amoco's two variant readings of the identical language, perhaps it might be viewed as obvious that Amoco was being arbitrary and capricious without its being equally obvious as to just which aspect of Amoco's conduct merited that characterization. But here, where it is the Annuity Benefit Formula calculation that stands out like a sore thumb from the uniform treatment that Amoco applies everywhere else, there is no room for doubt on that score: It is the now-challenged reading by Amoco that fails the test.
Obviously recognizing the need to explain that level of inconsistency, Amoco seeks to suggest some policy reasons why its inconsistent treatment of "earnings" does not warrant the "arbitrary and capricious" label. It argues strenuously that the intent of the Annuity Benefit Formula is to provide a pension benefit commensurate with an employee's salary at the time of termination, and that including pay in lieu of vacation in the calculation gives a distorted picture of the employee's actual final salary. That contention rests on the notion that pay in lieu of vacation is a one-time payment, like severance pay, that does not reflect the employee's income. Amoco asserts that the Career Average Minimum Formula is different because the distortion is less severe when annual earnings are averaged over an employee's entire career instead of using a three-year average.
Before this opinion addresses that subject, it is worth spending a moment or two on Amoco's characterization of the Plan's "intent" that retirement income should precisely mirror the retiree's final level of compensation. To begin with, no such "intent" is stated in the Plan or SPD. Second, Amoco itself has demonstrably departed from that characterization: In 1993 it changed the Plan by "de-coupling" bonuses from salary for purposes of calculating the three highest years for use in the Annuity Benefit Calculation (P.R.Mem.Ex. 1). Under the 1985 Plan salary and bonuses in each year had been added together, and the three highest years were determined from those combined figures. Now on implementation of the 1993 changes, the three highest salary years are determined separately from the three highest bonus years. That de-coupling provides employees in the discretionary bonus group (employees who are most likely highly-placed executives) with the benefit of any high bonuses that did not coincide with the three highest salary years. If it were really the Plan's intent simply to reflect an employee's compensation at retirement, salaries and bonuses would have remained coupled. As with Amoco's other contentions, this one smacks of opportunism rather than consistency: Say whatever must be said to support a position that is now sought to be advanced.
To return to Amoco's claimed justification for its disparate treatment of the identical item (pay in lieu of vacation) in its alternative calculations of the retirement formula, as a matter of general principle it makes no sense to argue that inconsistent interpretations of "earnings" under the two formulas is somehow justified because one causes a tolerable level of distortion while the other does not. Distortion is after all distortion. Including pay in lieu of vacation as part of "earnings" either provides an accurate picture of an employee's earnings record or it does not, and if the sole goal of the Plan were as Amoco urges and as this opinion has discussed earlier, Amoco's inconsistency of treatment would really be insupportable.
What plainly seems to have been lost sight of by Amoco, and what helps to explain the invalidity of its arguments, is the reason that such alternative pension calculations have come to be included in pension plans to begin with. Because the long-term trend in this country has been inflationary, pensions calculated (as was once common) only on an average of an employee's total historical earnings — without adjustments for cost of living or otherwise — tended to leave loyal employees who had served their employers over extended periods with fixed incomes that were totally inadequate to meet their retirement needs. Hence the alternative provisions that average only a few years' compensation — typically the highest and typically the most recent — were inserted into pension plans, both to take account of inflation and to take account of the employees' normally greater worth to the company as a result of longer service (something that tends to be reflected in increases in their compensation beyond the effects of inflation alone). Under those circumstances, the amount that is represented by an employee's final-year payment in lieu of that final year's vacation should be — absent some rational explanation — equally includable in any calculation that takes account of the employee's "earnings" in that final year.
That also helps to explain why plaintiffs' position does not really involve a "distortion" at all. Amoco seeks to analogize pay in lieu of vacation to severance payments, which Krawczyk, 41 F.3d at 279 has held may properly be excluded from a calculation of earnings. But that analogy is flawed — severance payments are predicated on an employee's entire past period of service, and in economic terms they may fairly be viewed as having been earned ratably over those years, so that realizing such a payment in cash in the employee's final year can seriously skew any calculation that is intended even generally to reflect the employee's real level of compensation in that year. By sharp contrast, under Amoco's "use it or lose it" arrangement the only vacation time that may come into the calculation is for a vacation that has been earned in the year of termination — or perhaps, from a different perspective, earned in both that year and the preceding year. Because both of those years are included in the three highest years for the Annuity Benefit Formula calculation, no economic distortion is introduced by including benefits that are directly attributable to services rendered in those years as part of the formulation.
That skewing is obviously even more pronounced where a plan follows the typical pattern of providing substantial severance pay for employees with a great deal of longevity, while at the same time choosing only a few years (including the severance year) to make the pension calculation for the reason already explained.
Thus vacations at Amoco are not provided until after an employee has completed one year of service (SPD Book 1 at 14), and absences from work during one year can reduce the amount of vacation time allowed in the following year ( id. at 16). It is worth adding that Amoco's current argument (D.Mem. 23-24) that vacation pay does not represent compensation for personal services — a clear lawyers' afterthought, for even its often-evasive witness Brennan was unwilling to subscribe to that bizarre notion (his Dep. 38-39) — is extraordinarily lacking in force. Does Amoco really believe that an employee who works (say) 50 weeks and is paid for 52 is not receiving the added two weeks' pay as compensation for his or her personal services? Does Amoco view that as a matter of corporate benevolence — perhaps a gift to its employees?
That also tends to scotch Amoco's further argument that its exclusion prevents employees from manipulating the size of their pension benefits by choosing pay in lieu of vacation. For one thing, Amoco specifically countenances such purported "manipulation" both in the Career Average Minimum Formula and in the other two employee benefit plans — another illustration of its inconsistency of argument as well as its inconsistency of treatment. For another thing (and more significantly), Amoco expressly holds out to employees who are leaving the company the right to take pay in lieu of vacation. SPD Book 1 at 16 states:
If you leave the company for any reason before you have taken all of your vacation for the year, payment for vacation you haven't taken will be included in your final paycheck.
Amoco can scarcely cry "manipulation" when an employee exercises a right that Amoco itself sets out in the Plan documents.
Finally, Amoco offers the affidavit of Ira Cohen, who has many years of experience in working with qualified benefit plans, including a stint with the IRS when he analyzed and assisted in the drafting of ERISA. Cohen's affidavit states that it is industry practice and good plan design to exclude pay in lieu of vacation from the Annuity Benefit Formula, and that it is consistent with good plan design to include pay in lieu of vacation in the Career Average Minimum Formula calculation. Moreover, Cohen asserts that it is not theoretically inconsistent to treat pay in lieu of vacation differently under the two formulas.
Amoco's problem is that expertise of that nature might be useful is this Court were faced with a very different question: whether an "arbitrary and capricious" characterization should be attached to some other plan that, after having taken such considerations into account, had chosen to build two different statements of its alternative calculation methods into its provisions. But that is totally different from what Amoco did, and so any opinion by Cohen in that respect would be wholly irrelevant to the present controversy. Indeed, Cohen's Aff. ¶ 10 expressly acknowledges that some other employers' plans do include pay in lieu of vacation (what Cohen labels "vacation buyouts") in their calculations of earnings under their formulas that correspond to the Annuity Benefit Formula, and of course he does not suggest that such a choice is "arbitrary and capricious." But that point, which would tend to undercut Amoco's position, would candidly be just as irrelevant as Amoco's attempted use of Cohen's opinions.
What controls here instead is that Cohen's explanations of what might have been cannot rescue Amoco's entirely inconsistent application of its single definition of "earnings" to plaintiffs' detriment. Cohen does not of course opine as to why industry practice or any good plan design would give a single plan definition such dramatically different applications in different parts of the Plan — it would frankly be astonishing if someone with his credentials were to do so. And it would be at least as astonishing if someone with Cohen's credentials were to opine that one of those dramatically different applications, which flies in the face of the same employer's uniform applications everywhere else, would conform to industry practice or to good plan design.
But neither of those scenarios is before this Court. Both such precedents as Egert and the teachings of common sense require a consistent interpretation of "earnings," and Cohen's affidavit does not justify Amoco's treatment to the contrary.
Conclusion
There is no genuine issue of material fact, and plaintiffs are entitled to a judgment as a matter of law. Amoco's exclusion of pay in lieu of vacation from the calculation of the Annuity Benefit Formula as to plaintiff class members is arbitrary and capricious. Amoco is ordered to recalculate the pension benefits of all class members to include pay in lieu of vacation within the "earnings" component and to make appropriate payments to plaintiffs to reflect those calculations.
This ruling involves the granting of plaintiffs' motion (Docket No. 42-1) and the denial of Amoco's (Docket No. 35-1).