Pension plan termination strategy – part 1

While interest rates remain near historic lows and, thus, measured pension liabilities remain high, some plan sponsors, particularly sponsors of frozen plans, have decided on (or are considering) plan termination. In this article we discuss the financial factors affecting a decision to terminate or not terminate a frozen plan, comparing the cost of buying annuities with the ‘book value’ of the plan plus ongoing overhead costs. This is not going to be a ‘numbers-based’ article – we simply want to review the factors a sponsor will want to consider in deciding to terminate/not terminate.

In a follow-on article we will consider a factor that is sometimes left out of the termination analysis – the difference between an analysis based on yield curve-based valuations and one that considers long-run returns.

Basic plan termination analysis

We’re going to focus on plans that have undergone what some call a ‘hard freeze’ – that is, a cessation of all accruals for all participants.

Such a plan represents a pure legacy obligation, not an employee benefit, because employees are no longer accruing benefits. Rather, a hard frozen pension plan is, from an employee’s perspective, “something the company owes me for work I did in the past.” For the sponsor, the plan looks like any other debt, albeit one that is funded by a trust and subject to a complicated regulatory regime.

In that context, the question – should we terminate the plan? – becomes a question of numbers: Is termination cheaper or more expensive than continuing to maintain the plan? Is there enough money in the fund to terminate it, and if not does the sponsor have the cash to ‘fund up?’

The numbers for each plan will be different. Critically, the demographics of the group covered by the plan will determine the effect of interest rates and mortality assumptions on measured liabilities. As we have written recently (see, e.g., our article Impact of changes in interest rate and mortality assumptions on de-risking activity), most sponsors are reviewing, or have already reflected, recent improvements in mortality, which can increase measured liabilities as much as 10% in some cases.

First cut

We begin with a summary of the basic plan termination analysis. For this purpose, we start with a baseline of the current value of plan liabilities on the sponsor’s financial statements: the plan’s projected benefit obligation (PBO). In analyzing the cost of terminating the plan, we compare (1) the present value of the ongoing plan cost to (2) the annuity purchase premium.

1. The ongoing plan cost

As we said, our baseline is the plan’s PBO. If the sponsor is going to continue the plan, then, in addition to the PBO, it must pay (1) administrative costs and (2) Pension Benefit Guaranty Corporation premiums. The present value of those two items is the present value of the (marginal) cost of continuing the plan.

Administrative costs

Administrative costs will vary somewhat from plan to plan. Sponsors will want to review their cost of DB plan administration. They may also want to consider whether there are ways – short of plan termination – of reducing these costs.

PBGC premiums

The present value of future PBGC premiums will also vary from plan to plan. And there are two different premium regimes. The PBGC flat-rate premium is a per-head cost – it bears no relationship to the size of the PBO. But terminating a plan does eliminate this premium. The PBGC variable-rate premium is not a function of measured liabilities per se, but rather of unfunded liabilities. As we discuss below, we believe this premium is better analyzed as a credit issue than as a function of a plan termination.

The PBGC flat-rate premium

The PBGC flat-rate premium is a function of headcount. Generally, sponsors should consider the extent to which headcount can be reduced before plan termination. Small benefits (those worth less than $5,000) can generally be paid out without the participant’s consent. Lump sums may also be offered to terminated vested participants. This sort of ‘de-risking’ will reduce the per-capita PBGC flat-rate premium just as much as paying out a more expensive annuity in connection with a plan termination.

For purposes of what follows we will assume that the sponsor has already ‘picked’ this low-hanging fruit. We now want to calculate (or at least understand) the present value of the future PBGC flat-rate premiums the sponsor will owe for the remaining participants (generally, retirees, active employees, and terminated vesteds that turned down the offer of a lump sum).

The 2016 flat-rate premium is $64 per participant, increasing to $69 in 2017, $74 in 2018, $80 in 2019, and adjusted for wage growth after 2019. The main variable determining present value (other than the interest rate assumption) is how long the participant is expected to continue to be a participant. We calculate the present value of the typical flat-rate premium for a 65 year old – assuming the participant stays in the (continuing) plan and receives a retirement benefit (and remains a participant) until death – to be about $1,400.

Two points are worth (re)emphasizing about this calculation:

It is unrelated to the size of the PBO. Thus a sponsor saves more in flat-rate premiums (relative to the benefit the plan owes) by settling (either with an annuity or a lump sum) smaller benefits than by settling larger benefits.

It is related to the length of future participation. The total cost of flat-rate premiums depends on how long in the future the employee will remain a participant. Thus a sponsor saves more (in flat-rate premiums) by settling (either with an annuity or a lump sum) benefits for younger participants than by settling benefits for older participants.

Underfunded plans and the PBGC variable-rate premium

The analysis with respect to the PBGC variable-rate premium is fundamentally different. Oversimplifying somewhat, underfunded plans must (in addition to the flat-rate premium) pay a variable-rate premium of $30 per $1,000 of unfunded liability in 2016, subject to a $500 per participant cap. The variable premium is increasing to at least $33 in 2017, $37 in 2018, $41 in 2019, and adjusted for wage growth after 2019. These amounts are also increased after 2016 for national average wage increases, so the 2019 variable premium is more likely to be $43 or $44.

Generally, variable-rate premium costs are better viewed as a choice between borrowing-and-funding vs. paying premium, rather than as a function of the plan termination calculus. (See our article PBGC variable premium vs. borrow-and-fund: impact of higher premiums.) With this caveat: if the sponsor’s borrowing costs are so high that paying the PBGC premium is a ‘better deal’ than funding, that is one reason not to terminate the plan or fund the shortfall today.

2. The annuity purchase premium

The annuity purchase premium is the cost of annuities settling all plan liabilities minus the PBO. It’s the premium you pay the annuity company, over the book value of plan liabilities. Our ‘first cut’ plan termination analysis is simply a comparison of the annuity premium to the present value of ongoing costs.

As many observers have noted, the adoption of new mortality tables by sponsors for purposes of calculating their PBO has brought the cost of an annuity more in line with the PBO on most plans’ books. Estimates of the marginal increase cost (vs. the PBO) vary but are generally in the single digits. The annuity market is not, however, very transparent, and underwriting is plan-specific and will depend on the plan’s unique demographics.

Generally, the annuity premium is smallest for blocks of current retirees, larger for deferred participants, and largest of all for active participants. Longer time horizons and ‘optionality’ with respect to the form and timing of benefit commencement for these groups create much more uncertainty, and higher prices, from insurers.

In some cases, we understand insurers are declining to bid on active and deferred lives at all, particularly cash balance plans with difficult-to-hedge interest crediting provisions.

Bottom line

It’s not our purpose to provide a ‘mock’ calculation of how much it costs to terminate a plan. A sponsor serious about termination should consult with its actuary to get a feel for the ‘annuity market value’ of its plan and for the potential ongoing cost of PBGC premiums and other administrative overhead costs.

* * *

Thus far we have considered the cost of continuing or terminating a plan taking as our baseline the plan’s PBO. In that analysis the key variables are (1) the present value of future plan administrative costs, plus (2) the present value of future PBGC premiums (a function of how many participants remain, and how long they are expected to remain, in the plan) compared to (3) the annuity premium.

The PBO is, however, calculated using a discount rate based on a high quality corporate bond yield cure. What a ‘PBO analysis’ leaves out is the value (if any) of an equity risk premium. In our next article we are going to consider the cost of a plan termination vs. continuing the plan and continuing to take equity risk on plan investments. We will also briefly take up the issue of offering lump sums as part of a plan termination.