How much does it cost to withdraw from a multi-employer pension plan?

Next week’s argument in M&K Employee Solutions v. Trustees of the IAM National Pension Fund presents a technical question of great monetary significance – how to calculate what a business owes if it withdraws from a multi-employer pension plan.

The case involves provisions of the Multiemployer Pension Plan Amendments Act of 1980, which upgraded the provisions of the Employee Retirement Income Security Act of 1974 to protect multiemployer pension plans. The basic concern the statute addresses is that any single employer has a strong incentive to leave a multiemployer pension plan as soon as the plan experiences financial difficulty, lest the employer be left as the last one “holding the bag” and so responsible for paying unfunded benefits.

The intuition of the MPPAA is to remove the incentive to race for the exits by requiring a departing employer to make a one-time cash payment equal to its share of any shortfall in the plan’s assets when the employer departs. Assessing the amount of that shortfall is not a simple task, because it depends on numerous things that are not entirely clear at the time of the withdrawal. For example, the amount of benefits that the plan will pay out in the future will depend, among other things, on how long the employees ultimately live. Similarly, the amount it would cost in present assets to fund those benefits would depend on the level of interest rates during the years between the employer’s withdrawal and the ultimate payment of the benefits.

The statute calls for the departing employer’s obligation to be determined “as of” the last day of the year before the employer leaves. So, for example, if the employer leaves in 2026, and if the plan’s fiscal year matches the calendar year, the obligation would depend on the assets and liabilities of the plan “as of” December 31, 2025. The dispute in the case turns on what it means to calculate the shortfall “as of” that date. In this case, for example, the actuaries changed their actuarial assumptions in January 2018, before M&K’s withdrawal in 2018, but after December 31, 2017, the last day of the fiscal year before M&K’s withdrawal. Because the January 2018 actuarial assumptions led to a higher withdrawal payout than the assumptions in place during 2017, M&K sued and sought application of the earlier actuarial assumptions.

This is a subtle case. The employer argues that to calculate liability “as of” the withdrawal date requires use of the interest rate and other actuarial assumptions in place on that date, not some after-adopted assumptions. For the employer, using the later assumptions results in fudging the calculation to take account of later occurrences, things that happened after the valuation date. Moreover, it argues, letting the fund benefit from later changes in actuarial assumptions gives the fund an incentive to switch actuaries when it learns of an impending withdrawal, with the intent that the newly selected actuary will select assumptions that raise the withdrawal liability for the departing employer.

For its part, the fund emphasizes the statute’s command that the actuary always make calculations that reflect its best estimate of the financial position of the plan. And an assessment of the plan’s position “as of” a past date necessarily contemplates the actuary’s best view on the date of calculation of the plan’s financial position on that past date. For the actuary to use stale financial assumptions necessarily is to make a calculation that, on the date it is made, is not the actuary’s best estimate of the plan’s financial position on the relevant date. It would be as if an actuary in June 2020 (after the outbreak of COVID-19) made financial calculations for the plan based on the interest rates it had chosen in June 2019, knowing full well on the date of the calculation in 2020 that those interest rates were poor choices for the plan’s position as of January 2020.

It is hard to say what the justices will think of this, but I certainly see the fund as getting the better of the arguments “on paper,” as it were. The most persuasive point to me is that the amount of withdrawal liability is not like the Dow Jones – for which there is a specific number that we can see at any time. The amount of withdrawal liability for a particular date is to a large degree unknowable until an actuary sits down and calculates it using the assets and obligations that the plan had on the relevant date. And it seems likely, in the frame of the statute, that what Congress would have wanted was for the actuary – on the day he or she sat down to make that calculation – to do the best job the actuary could do on that day, which would mean using his or her best estimate, as of that day, of interest rates and life expectancies on the date of calculation. I’ll be interested to see how the justices react next week.