Matthews v. E.I. Dupont De Nemours & Co.
| Decision Date | 16 March 2017 |
| Docket Number | No. 16-3237,16-3237 |
| Parties | MARK MATTHEWS, Appellant v. E.I. DUPONT DE NEMOURS & COMPANY, a Delaware corporation; HEWITT ASSOCIATES LLC, an Illinois corporation |
| Court | U.S. Court of Appeals — Third Circuit |
NOT PRECEDENTIAL
On Appeal from the United States District Court for the District of Delaware
Submitted Under Third Circuit L.A.R. 34.1(a)
March 6, 2017
Before: SMITH, Chief Judge, HARDIMAN, and KRAUSE, Circuit Judges.
Mark Matthews appeals the District Court's summary judgment in favor of his former employer, E.I. DuPont De Nemours & Company, in a case arising under the Employee Retirement Income Security Act. The District Court held that the decision of DuPont's Plan Administrator to revise Matthews' pension-benefit calculation was entitled to deference under ERISA. Matthews argues that the District Court should have reviewed the Administrator's decision de novo because it was interpreting a state court order instead of the Plan. We agree with the parties that the standard of review dictates the result in this appeal and because the District Court should have applied de novo review, we will reverse.
Four undisputed facts set the stage for this dispute. First, Matthews earned a pension from DuPont while working there from 1973 to 2013. Second, Matthews married his ex-wife in 1973 and divorced in 1993, meaning they were together for about half of his tenure at DuPont. Third, Matthews' divorce decree, entered by the Delaware Family Court in 1995, included a Qualified Domestic Relations Order (QDRO) that required DuPont to pay a specified portion of Matthews' pension benefit directly to his ex-wife. And fourth, DuPont amended its Plan at the end of 2007, reducing benefits to pensioners for post-2007 work by two-thirds.
Although Matthews' pension benefit determination involved numerous byzantine calculations, this appeal turns on one number: the denominator of the "marital fraction." The marital fraction is a ratio that determines the percent of Matthews' pension to which his ex-wife is entitled. Contained in the QDRO, the marital fraction is calculated by dividing the "[n]umber of months [Matthews] was in the Plan during the marriage" by the total "[n]umber of months [Matthews] was in the Plan." App. 13. This basic formula, which has been approved by the Delaware Supreme Court and is often used in Delaware QDROs, is known as the Cooper Formula. See Simms v. Greene-Simms, 22 A.3d 727, 732 (Del. Fam. Ct. 2009).
DuPont altered the basic arithmetic of the Cooper Formula on the heels of its 2007 Plan amendment, which reduced Matthews' accrual of future benefits by two-thirds. To account for that change, DuPont began crediting each year after 2007 as only one-third of a year for marital-fraction purposes. This change "ensure[d] that an employee's months of post-2007 participation are not unfairly weighted given the reduced accrual rate for that period." DuPont Br. 6. Accordingly, Matthews' final six years of service (from 2008 to 2013) counted only as two years in his marital fraction, reducing his marital-fraction denominator from 487 to 439 months. That change resulted in Matthews' ex-wife receiving an additional $262.11 per month.
Matthews challenged this determination through two layers of internal appeals, but his claims were denied by the DuPont Benefit Determination Review Team. He then filedthis ERISA action in the U.S. District Court for the District of Delaware. The District Court reasoned that "[w]hat DuPont has done here occurs at the border between interpretation of the [QDRO] (to which I owe its decision no deference) and its interpretation of the Plan (to which I owe it substantial deference)." Mathews v. E.I. Du Pont De Nemours & Co., 2015 WL 8082315, at *2 (D. Del. Dec. 7, 2015). After determining that DuPont's interpretation was "within the umbrella of its discretionary decision-making," the District Court granted summary judgment to DuPont. Id. Matthews filed this appeal.
Matthews raises two related issues on appeal: (1) whether the District Court erred in reviewing DuPont's determination for abuse of discretion; and (2) whether DuPont's pension determination was correct. As we noted at the outset, the answer to the second question is dictated by the way the first question is resolved.
Matthews claims DuPont's determination was based on an interpretation of the QDRO, which we would review de novo. See Hullett v. Towers, Perrin, Forster & Crosby, Inc., 38 F.3d 107, 114 (3d Cir. 1994). DuPont counters that its determination was based on an interpretation of the Plan, which we review to determine whether the Administrator's decision was arbitrary, capricious, or an abuse of discretion. See Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115 (1989); Miller v. Am. Airlines, Inc., 632 F.3d 837, 844 (3d Cir. 2011).
We agree with Matthews that DuPont was interpreting the QDRO. At the most basic level, the dispute is about the meaning of "months [Matthews] was in the Plan" (i.e., does "months . . . in the Plan" refer to literal calendar months, or does it have some latent meaning that allows for DuPont's re-weighting?). See App. 13. That provision—and the entire marital fraction—is found in the QDRO entered by the Delaware state court.
DuPont's arguments against this straightforward conclusion fall short. Although DuPont asserts somewhat opaquely that it was interpreting the 2007 amendment, it neither identifies the Plan provision it claims to be interpreting nor quotes the Plan at all. In fact, DuPont indicated throughout the appeal process that it was interpreting the QDRO. For example, in a letter to Matthews by the Benefit Determination Review Team, DuPont explained that it "interprets the language in a [QDRO] using a marital fraction to mean 3/3 of a month for each month prior to January 1, 2008, and 1/3 of a month for eachmonth [thereafter]." App. 531. It also stated that the QDRO "is written in a certain way and that based on DuPont's procedures we will interpret it a certain way." App. 569.2 Thus, even had the 2007 amendment factored into DuPont's determination, it would have been as a backdrop for interpreting the language of the QDRO.
DuPont next claims it was merely "implementing" the QDRO. DuPont Br. 17. While it is true that a QDRO formula is not self-executing and will require some math, such implementation does not permit the discretionary alteration of the Cooper Formula that DuPont made here. See, e.g., Blue v. UAL Corp., 160 F.3d 383, 386 (7th Cir. 1998) ( ); Hullett, 38 F.3d at 114 ().
We turn now to the merits. Matthews argues almost exclusively that DuPont's benefit recalculation cannot withstand de novo review because it is contrary to the terms of the QDRO. For its part, DuPont argues exclusively that its determination was not an abuse of discretion. As the parties tacitly concede, the result of this appeal is dictated by the standard of review.
Having explained why de novo review applies, our analysis begins and ends with the text of the QDRO, which states that Matthews' ex-wife's benefit should be determined by taking the "[n]umber of months [Matthews] was in the Plan during the marriage" and dividing by the "[n]umber of months [Matthews] was in the Plan." App. 13. There is no indication in the QDRO that "[n]umber of months . . . in the Plan" should mean anything other than a tally of the number of months Matthews was a participant in the DuPont Plan.
The controlling state law supports this straightforward reading.3 Estate of Osborn ex rel. Osborn v. Kemp, 991 A.2d 1153, 1159-60 (Del. 2010) (internal quotation marks and citation omitted). As the District Court rightly noted, DuPont all but concedes that an objective reading of the QDRO supports Matthews' interpretation. Matthews, 2015 WL 8082315, at *2 (...
To continue reading
Request your trial