On May 21, 2026, the Supreme Court handed down a unanimous opinion in M & K Employee Solutions v. Trustees of the IAM National Pension Fund, and if you’re an employer who’s ever thought about walking away from an underfunded multiemployer pension plan, you should be paying attention.

The question was deceptively simple: when a plan actuary picks the assumptions used to calculate how much a withdrawing employer owes, does that decision have to be made before the last day of the prior plan year, which is the date as of which the amount the employer owes will be determined? Or can the actuary set key assumptions after that measurement date, even after becoming aware of the employer’s withdrawal, perhaps armed with the benefit of hindsight?

Justice Jackson, writing for all nine Justices, settled the matter: no deadline exists. The actuary can pick assumptions after the measurement date, as long as the assumptions are based on information that existed on the measurement date.  While this is a loss for employers that rely on multiemployer pension funds’ withdrawal liability estimates when planning for potential withdrawal liability, the decision did have a silver lining in that the Court recognized that actuaries are capable of bias when setting assumptions and emphasized that hindsight is not permitted. 

Calculation of Unfunded Vested Benefits and the Measurement Date

As described in our earlier alert, when a contributing employer in a multiemployer pension plan ceases to have an obligation to contribute to the plan (e.g. in the case of a cessation of union operations or bargaining out of the obligation to contribute to the plan), the employer “withdraws” from the plan.  At a high level, this withdrawal gives rise to a liability – called “withdrawal liability” – if the withdrawing employer is obligated to pay a share of the plan’s underfunding due to the withdrawal. 

Under ERISA, the plan’s underfunding is defined in section 4211 (29 U.S.C. § 1391) as the plan’s “unfunded vested benefits” (i.e., the value of the plan’s nonforfeitable benefits less the value of the plan’s assets) as of the end of the plan year preceding the date the employer withdraws.  Accordingly, the withdrawal date will always be after the date commonly referred to as the “measurement date”, as of which the plan’s unfunded vested benefits for that withdrawal are determined.  ERISA section 4213 (29 U.S.C. § 1393) requires the plan’s actuary to select an interest rate and certain other assumptions when calculating a plan’s unfunded vested benefits.  Other assumptions being equal, the lower the assumed interest rate, the higher the liability. 

Employers have the right under ERISA to request a withdrawal liability estimate annually.  Employers contemplating a withdrawal due to transactional or bargaining activity generally rely on these estimates to plan for potential withdrawal liability. 

Case Background

In this case, the employer group (M&K) withdrew from the IAM National Pension Fund in 2018 meaning the measurement date for determining M&K’s withdrawal liability was December 31, 2017.  The discount rate used to value the plan’s unfunded vested benefits that was in effect on that date was 7.5%.  However, after that date, the plan’s actuary adopted a 6.5% discount rate and used that rate to value the plan’s unfunded vested benefits.  That one-point change was not academic. It ballooned the Fund’s unfunded vested benefits from roughly $500 million to over $3 billion – a sixfold increase. For M&K, withdrawal liability jumped from approximately $1.8 million to around $6.2 million.

M&K challenged the plan’s calculation using assumptions adopted after the measurement date through the arbitration process in accordance with ERISA, scoring a victory.  The arbitrator concluded that the plan must use assumptions in effect on the measurement date.  The plan appealed to the U.S. District Court for the District of Columbia, which held that the plan actuary can use assumptions adopted after the measurement date as long as those assumptions are based on information as of the measurement date.  The D.C. Circuit Court affirmed. This created a circuit split because the Second Circuit in National Retirement Fund v. Metz Culinary Mgmt., Inc., 946 F. 3d 146, 152 (2020), held that the plan actuary was required to use assumptions in effect on the measurement date.

On May 9, 2024, M&K filed a petition for writ of certiorari with the Supreme Court.  The petition was opposed by the pension fund.  On May 30, 2025, the Supreme Court granted the petition as to the narrow issue of “[w]hether 29 U. S. C. §1391’s instruction to compute withdrawal liability ‘as of the end of the plan year’ requires the plan to base the computation on the actuarial assumptions to which its actuary subscribed at the end of the year, or allows the plan to use different actuarial assumptions that were adopted after the end of the year.”

Supreme Court Holds “As Of” Doesn’t Mean “Locked In”

M&K argued that because withdrawal liability must be calculated based on the plan’s unfunded vested benefits “as of” the measurement date, every input – including actuarial assumptions – must be locked in by that date. The Supreme Court sided with the pension fund, holding that the actuarial assumptions for calculating a plan’s unfunded vested benefits can be set after the measurement date, as long as those assumptions reflect the actuary’s knowledge as of the measurement date.  In so holding, the court reasoned that actuarial assumptions aren’t observable facts or hard data that are frozen as of a particular date, but tools actuaries use to predict pension funding outcomes. Facts get frozen on the measurement date; tools do not.

The Court further observed that unlike other ERISA provisions, the provision that specifies the assumptions to be used in calculating withdrawal liability, section 4213, does not require that those assumptions be set on any particular date.  Rather, the Court determined that the only limiting factor Congress put on the actuary’s assumptions is the requirement that such assumptions be reasonable and reflect the actuary’s best estimate of anticipated experience under the plan.  In fact, if the actuary could not use assumptions that are developed after the measurement date, the Court reasoned that the actuary might be forced to use assumptions that do not reflect the actuary’s best estimate.

The Court rejected the argument that permitting an actuary to set assumptions after the measurement date could result in manipulation by the actuary to inflate the plan liabilities to increase an exiting employer’s withdrawal liability.  Unlike in the Court’s most recent significant decision in which it addressed withdrawal liability issues, the 1993 Concrete Pipe & Prods. v. Constr. Laborers Pension Tr., 508 U.S. 602 (1993) decision, the Court recognized that a plan’s actuary can be biased and improperly manipulate assumptions but reasoned that was the case without regard to whether those assumptions were adopted after the measurement date. 

Thompson Hine’s Takeaways

Following the Supreme Court’s decision, employers are more likely to be blindsided by changes to actuarial assumptions used to calculate withdrawal liability. But employers are not without recourse, and an assumption that looks more like post-hoc money-grab than an actuary’s “best estimate” is still challengeable. The Supreme Court noted that actuaries are obligated to select reasonable assumptions that reflect the actuary’s best estimate of anticipated experience of the plan based on information known to the actuary as of the measurement date.  An employer seeking to challenge actuarial assumptions can still request review and demand arbitration, at which point it is now even more critical to press the actuary on the information relied upon to develop the assumptions used to calculate the pension fund’s unfunded vested benefits.  To the extent an actuary relies on information that becomes known only after the measurement date in developing an assumption, or the assumption is otherwise unreasonable, that assumption is impermissible. 

Additionally, employers should be wary of relying solely on withdrawal liability estimates provided by the pension fund when engaging in corporate planning. Best practice is to stress test the pension fund’s estimates with alternative interest rates and in some cases other assumptions that might be used in the future.  This stress-testing is even more important following this decision because the estimate may be completely unreliable as provided.

Please contact the authors or your regular Thompson Hine attorney with any questions.