Lump Sums, Pre-Retirement Mortality

[Rball4]

Prior to PPA, you used to calculate a lump sum using the same method as actuarial equivalence, 1.417(e)-1(d)(1). So if actuarial equivalence had no pre-retirement mortality, you would use interest only for lump sum determination as well. But did PPA change this to always use pre-retirement mortality? Reading through PPA, it seems to not fully address this. What is being done in practice?

[Effen]

No change. You can use, or not use, pre-retirement mortality based on the provisions of the plan document.

[Andy the Actuary]

Even if you're adamant that THE MINIMUM calls for the use of pre-retirement mortality, ignoring pre-retirement mortality produces a greater amount so you should be okay. Consistency is more important.

[My 2 cents]

Amounts payable as lump sums are subject to 417(e) and the terms of the plan. The plan needs to explicitly say if there is to be no pre-retirement mortality in the definition of actuarial equivalence if that is what is intended. If the plan does not call for there being no pre-retirement mortality for actuarial equivalence purposes, pre-retirement mortality should be used for that purpose. Plans that do call for no pre-retirement mortality for actuarial equivalence purposes can and should use no pre-retirement mortality for that purpose.

What one can or should do for minimum funding requirement calculations is the subject for a different discussion.

[AndyH]

The plan needs to explicitly say if there is to be no pre-retirement mortality in the definition of actuarial equivalence if that is what is intended. If the plan does not call for there being no pre-retirement mortality for actuarial equivalence purposes, pre-retirement mortality should be used for that purpose.

No disagreement; everyone I know adheres to this in practice, but people I have asked cannot cite this authoritatively in print. Is there some pronouncement or standard that actually says this?

[Andy the Actuary]

Benefits must be definitely determinable. If the Plan says applicable mortality table without specifying post-mortality only, then it is implicit that APM applies at every age.

If we said, the a.e. is the 1994 GAR and 5% interest, would you need further clarification to stipulate that the 5% applies for both pre-and post- retirement?

[AndyH]

Fair point, but couldn't a counter-argument be made that the lump sum calculation requires both an assumption and a methodology? Should the death benefit be considered, i.e. the mortality factor applied to the amount of benefit subject to forfeiture upon death, which might be zero?

I don't disagree with you, just wondering if there is something in print on this.

Ran into a similar issue with late retirement..

[Effen]

I agree with AtA, but if the death benefit is 100% of the present value of the accrued benefit, you could argue that the use of pre-retirement mortality would not be reasonable because there is no forfeiture on death.

Therefore, even though it is not explicit in the document, an argument could be made that any other interpretation would not be reasonable in that specific situation.

[Andy the Actuary]

This won't give you the answer, because again, it's a default position, but the IRS 417(e) regs only state that PV must be as great as PV accrued benefit payable at normal retirement benefit. You could get into why e.r. subsidies and other ancillaries as well as the actual lump sum calculation methodology aren't included if you truly want to stir the pot. You're lucky if the plan documents describes the determination of the lump sum the way it is actually calculated!

[My 2 cents]

If the plan has a fully subsidized pre-retirement death benefit equal to the present value of the accrued benefit, would it be unreasonable to say that the value of that death benefit should be excluded from the lump sum payment (i.e., by discounting for pre-retirement mortality and interest)? If the employer chooses to generously provide full pre-retirement death benefit coverage, at its expense (which it is not required to do), why should the participant who is taking a lump sum derive an extra benefit just because they are foregoing that coverage? It is not part of the accrued benefit, after all.

[Calavera]

I believe the legality of not using the pre-retirement mortality when the death benefit is 100% of the present value of the accrued benefit is coming from 415 regulations. See 1.415(b)-1(d)(2).

[My 2 cents]

The following is based on my understanding of the various rules:

415 regulations do not allow a full actuarial increase when there are no forfeitures under the plan due to death before commencement (since the theoretical underpinning of an actuarial increase takes into account presumed reallocation among the living members of the theoretical cohort of the benefits/accounts of the members of the theoretical cohort presumed to have died).

This really does not directly impact the calculation of lump sums for people being paid lump sums prior to normal retirement age unless their lump sums are being restricted under Section 415, and then only in the calculation of the 415 limit itself. The mortality basis to be used to calculate the plan's lump sum amount is specified in the plan's definition of actuarial equivalence, which either limits assumed mortality to the period after the benefit commencement date or doesn't limit assumed mortality to that period. If the plan provides for actuarial increases and for application of mortality before and after retirement, the actuarially increased benefit being cashed out factors in those theoretical reallocations even if the calculation of the 415 limitations cannot.

[AndyH]

My 2 cents, why do you state that the death benefit is not part of the accrued benefit?

(BTW, the 415 cite does illustrate the concept we are discussing, but certainly it does not mandate what needs to happen with non-415 limited benefits)

[My 2 cents]

It is my understanding that death benefits are considered incidental, and that (for example) a plan providing a death benefit equal to the full lump sum value of the accrued benefit can be prospectively amended to reduce the death benefit to a standard QPSA without there being an impermissible cutback (in the same way that a plan can be amended to prospectively eliminate special benefits payable upon a participant's becoming disabled after the amendment).

AndyH, did you mean to say that the 415 cite certainly does not mandate what needs to happen with non-415 limited benefits?

[AndyH]

yes, thanks, edit made.

[Calavera]

I agree that 415 cite does not mandate what needs to happen with non-415 limited benefits. I have seen it was used as guidance when the plan document does not make clear distinction between pre- and post- retirement mortality.

[AndyH]

I agree is it a useful addition to the discussion.

Nobody had cited a "book" on the how the calc must be done. I have seen a 50% mortality discount in the case of a REA death benefit. I know a very prominent actuary who argues that a partial mortality adjustment is theoretically correct based upon the amount of the death benefit, so if there is a full pvab death benefit this would have the effect of no mortality discount even if the assumption of mortality was stated. This argues for a distinction between assumption and method.

[My 2 cents]

But in the context of a lump sum equivalent of the accrued benefit, the value of the subsidized pre-retirement death benefit would not necessarily be an appropriate part of the lump sum owed to the participant any more than the potential value of a subsidized disability benefit would be. Whether to discount for pre-retirement mortality would be governed by the definition of actuarial equivalence, not by the pre-retirement death benefit. It seems reasonable to me that by electing an immediate lump sum prior to normal retirement age, the participant would be foregoing the value of any further pre-retirement death benefit coverage offered under the plan, just as a participant who meets the service requirements for early retirement but not the age requirements and who is to be paid an immediate lump sum is almost surely not entitled to have the value of any early retirement subsidy (applicable to people who have met both the age and service requirements) included in their lump sum calculations. If that person were to be offered an immediate QJSA benefit prior to eligibility for early retirement (to satisfy the requirements for proper spousal coverage), it would be actuarially equivalent to the accrued benefit payable starting at normal retirement age, not to the subsidized benefit that would have become payable had the participant held off until early retirement eligibility.

[AndyH]

Good points, as yours always are.