Three recent decisions illustrate the threats facing multiemployer plans as they struggle to collect withdrawal liability while trying to forestall insolvency.
In United Mine Workers of America 1974 Pension Plan v. Energy West Mining Co., 2020 WL 2615536 (D.D.C. 2020), appeal docketed, No. 20-7054 (D.C. Cir. June 24, 2020), the district court held that the plan could use Pension Benefit Guaranty Corporation (”PBGC”) plan termination interest assumptions to discount vested benefit liabilities for withdrawal liability purposes, producing a much higher claim than the plan’s funding assumptions would.
In Sofco Erectors v. Trustees of the Ohio Operating Engineers Pension Fund, 2020 WL 2541970 (S.D. Ohio May 19, 2020), appeals pending. No. 20-3639 and -3671 (6th Cir.), the district court held that the plan could not use the “Segal Blend” (a weighted average of PBGC and funding rates), which produced a claim somewhere in the middle.
History may not repeat itself, but it often rhymes. The District Courts for New Jersey and the Southern District of New York split on the validity of the Segal Blend two years ago (New York Times Co. v. Newspaper & Mail Deliverers’ – Publishers’ Pension Fund, 303 F.Supp. 3d 236 (S.D.N.Y. 2018), appeals voluntarily dismissed, Nos. 1801140 and 1801408 (2d Cir 2019); Manhattan Ford Lincoln v. UAW Local 250 Pension Fund, 331 F.Supp. 3d 365 (D.N.J.), appeal voluntarily dismissed, 2018 WL 10759132 (3d Cir. 2018).
Taking a different approach, the court of appeals in The National Retirement Fund v. Metz Culinary Management, 946 F.3d 146 (2d Cir. 2020), pet. for cert. pending, No. 19-1336, held that the plan’s actuary could not change from funding assumptions to PBGC factors except for withdrawals occurring after the change.
PBGC assumptions have been in effect since the dawn of ERISA time. PBGC uses periodic surveys of group annuity prices and a standard mortality table to derive a set of interest factors. The mortality and interest factors are used to determine the asset sufficiency of terminating single-employer and multiemployer plans and termination liability claims.
The Segal Blend was developed in the early 1980s, shortly after Congress enacted the Multiemployer Pension Plan Amendments Act of 1980 (“MPPAA”). MPPAA introduced withdrawal liability in its current form: a departing employer’s unconditional obligation to pay its share of the plan’s underfunding.
Withdrawal liability is based on the plan’s unfunded vested benefits (“UVB”). 29 U.SC. § 1391. UVB represents the value of vested benefits minus the value of plan assets. 29 U.S.C. § 1393(c). Both values are to be derived by using actuarial assumptions and methods “which, in the aggregate, are reasonable (taking into account the experience of the plan and reasonable expectations) and which, in combination, offer the actuary’s best estimate of anticipated experience under the plan.” 29 § 1393(a)(1). The actuary “may rely on the most recent complete actuarial valuation used for [minimum finding] purposes . . . and reasonable estimates for the interim years . . .” 29 U.S.C. § 1393(b)(1). The employer’s share of UVB is usually based on its proportion of total plan contributions for years preceding its withdrawal. 29 U.S.C. § 1391.
In the early 1980s, interest rates were extremely high. Even 30-year Treasury yields were consistently in the double digits. PBGC interest factors peaked at 11%. Plan funding assumptions, also subject to the reasonable/best estimate test, took a long-term view and generally reflected expected earnings on a prudent mix of equities, fixed income, and cash equivalents, generally in the 7 to 8% percent rage.
The higher the discount rate, the lower the present value of a liability. So using a blended rate instead of pure funding assumptions gave withdrawing employers a break.
For decades, though, the effect has been the opposite. PBGC interest factors have generally been lower than funding interest assumptions. Today, the PBGC interest factors are less than 2%. Thus, employers, once cosseted by the Segal Blend, now have an incentive to challenge withdrawal liability calculation using anything other than funding assumptions.
As a rule of thumb, a 100-basis point change in the interest rate can swing the liability by 10-20%, depending on the length of the payment stream. A 400- or 500-basis point decrease can have an enormous effect, as the cited cases illustrate.
As plans become less well funded and some slip toward insolvency, they may—and arguably should—shift to more conservative investments. That can explain the use of PBGC assumptions or other interest assumptions consistent with low-risk investments.
Some actuaries say that, regardless of a plan’s funding status, withdrawal represents a “settlement,” as the employer will no longer share in gains or be asked to make up losses as a contributor. Use of PBGC assumptions is consistent with this view, as is the Segal Blend at least insofar as it uses “settlement” rates as part of a weighted average. More generally, actuaries rely on Actuarial Standard of Practice 27, which requires that economic assumptions be suited to their purpose.
Withdrawal liability disputes are heard by arbitrators, and there is necessarily a battle of experts. The plan actuary’s determination of UVB “is presumed correct” unless the employer “shows by a preponderance of evidence that—. . . the actuarial assumptions and methods . . .were, in the aggregate, unreasonable (taking into account the experience of the plan and reasonable expectations). . . .” 29 U.S.C. § 1401(a)(3)(B).
The appellate case law generally supports the actuary’s exercise of judgment in developing withdrawal liability assumptions and methods. In Combs v. Classic Coal Co., 931 F.2d 96 (D.C. Cir. 1991), the D. C. Circuit upheld the actuary’s use of funding assumptions as reasonable despite the employer’s argument that mid-1980’s short-term rates or a blend would be more reasonable. In Chicago Truck Drivers, Helpers & Warehouse Workers Union (Indep.) Pension Fund v. CPC Logistics, 698 F.3d 346 (7th Cir. 2010), the Seventh Circuit overturned not the Segal Blend itself but the actuary’s allowing the plan trustees to choose that method.
In Concrete Pipe & Products v. Construction Laborers Pension Trust, 508 U.S. 602 (1993), the Supreme Court stated: “Using different assumptions [for different purposes] could very well be attacked as presumptively unreasonable both in arbitration and on judicial review.” Id. at 633 (bracketed material in original; citation omitted).
But the Court concluded that the employer's burden is simply “to show that the combination of methods and assumptions employed in the calculation would not have been acceptable to a reasonable actuary . . .” and “falls outside the range of reasonable actuarial practice.” [S]ince imprecision inheres in the choice of actuarial methods and assumptions, the resulting difficulty is simply in the nature of the beast. 508 U.S. at 635-36.
Energy West lines up with Manhattan Ford that actuaries may have best estimates that differ for withdrawal liability and funding purposes. Sofco lines up with New York Times. Unlike the situation in Times, in Sofco, the actuary clearly testified that the Segal Blend represented his best estimate for withdrawal liability purposes.
In Metz, the court of appeals reviewed the application of PBGC assumptions, but ultimately ruled on procedural grounds. It held that while a new actuary may replace her predecessor’s assumptions and methods, the plan cannot apply them to a withdrawal occurring earlier in the plan year. The court relied on a MPPAA provision meant to protect employers from retroactive changes:
(a) No plan rule or amendment adopted after January 31, 1981, under section 1389 or 1391(c) of this title may be applied without the employer’s consent with respect to liability for a withdrawal . . . before the date on which the rule or amendment was adopted.
(b) . . . The plan sponsor shall give notice to all employers who have an obligation to contribute under the plan . . . of any plan rules or amendments adopted pursuant to this section.
29 U.S.C. § 1394
The court acknowledged that “actuarial assumptions and methods” may differ from “plan rules or amendments.” And the statute does not speak to retroactivity of actuarial assumptions and methods. But, relying on legislative history, the court held that retroactive application of new assumptions was “inconsistent with Congress’s legislative intent.” 946 F.3d 150-51.
Moreover, the court said, allowing retroactivity “would create significant opportunity for manipulation and bias . . . . Actuaries unwilling to yield to trustees’ preferred interest rate assumptions can be replaced by others less reticent.” 846 F.3d at 151.
The response might be that the “best estimate” standard exists to police against interference with the actuary’s judgment. CPC Logistics; Huber v. Casablanca Indus., 916 F.2d 85 (3d Cir. 1990); see Concrete Pipe, 508 U.S. at 633 n. 19.
In any case, it seems to be open season on withdrawal liability assumptions and methods. Both plans and employers will need expert legal advice and well-chosen actuarial experts for planning and conduct of withdrawal liability disputes.
Parties involved in mergers and acquisitions and distressed investing would also do well to have expert legal advice and an actuarial expert. Both buyers and sellers can price transactions better if they know how to evaluate withdrawal liability exposure.
*Mr. Goldowitz was on the PBGC’s amicus briefs in Concrete Pipe and New York Times.
⸙This article is reprinted with permission from The Wagner Law Group.