Post Retirement Mortality?

[AndyH]

Small frozen DB plan provides a death benefit equal to the present value of the accrued benefit. Actuarial equivalence definition includes pre retirement mortality.

Participant defers payment past NRD. Is the deferred benefit from NRA to a later age (assume 68) required to be increased with mortality (using D's)? Document does not specify and past practice was no since there is no forfeiture on death.

I know there is past discussion of this, I have read it. The key word is required.

FWIW, I think the IRS clearly says no for purposes of 415 adjustments.

Opinions please.

[david rigby]

Participant defers payment.

Just to be clear, does this sentence mean "defers beyond NRD"?

[AndyH]

Participant defers payment.

Just to be clear, does this sentence mean "defers beyond NRD"?

Yes. I will edit to clarify. Thanks.

[David MacLennan]

Very bad idea to put pre-retirement mortality into a document. Who drafted the document? Please chastise them for me. ;-]

A complete answer to your question is beyond the scope of the bulletin board format, but NOT increasing with mortality is probably pretty close to the correct answer. So, just use interest.

[AndyH]

A PBGC (post termination) auditor says the benefit must be increased with mortality. Formally, in writing.

Looking for information for our counter-argument/appeal.

(I happen to strongly agree with your comment about pre-retirement mortality, BTW, but not everybody does).

[David MacLennan]

The auditor is confused. He/she is asking for something that is not allowed by the plan document. Just tell him/her you must adhere to the terms of the document.

He is confused because he thinks that the actuarial equivalence definition is a benefit provision. Remind him it is in Section 1 with the definitions and not Section 4 with the benefits. If the a.e. assumptions say to use pre-retirement mortality, that does not mean that the benefit is forfeitable. He is making a mathematical mistake, ignoring the value of the death benefit. In other words, he is only using one side of the probability.

You have an uphill battle though, because this auditor and all his/her colleagues at PBGC have probably done this before, so they may be be reluctant to admit an error if it has an impact on their earlier cases.

[AndyH]

Thanks.

Hoping this auditor is a renegade - but don't know yet. If not, we won't be the only ones facing surprises of this sort - after the fact.

What would your view be if the death benefit were a REA only benefit (not paid by employee)? Subjective?

[david rigby]

... increased with mortality (using D's)?

D's?

[SoCalActuary]

... increased with mortality (using D's)?

D's?

Dx/Dy adjustment for age instead of (1+i) adjustment. It may have been a while since you read the Jordan book.

[Draper55]

isn't really the N's instead of the D's.. the D's would adjust

a single payment for qx and i but not an annuity..

[david rigby]

isn't really the N's instead of the D's.. the D's would adjust

a single payment for qx and i but not an annuity..

Bingo. Sure, my copy of Jordan ("my precious") gathers dust, but it should be the ratio of N(12)'s.

[AndyH]

Is the

age 65 Annuity rate x (1+i)^3 / the Age 68 Annuity rate

method (assuming age 68), for a plan that provides for a full pvab "wrong", as the auditor says, is the question.

[SoCalActuary]

Is the

age 65 Annuity rate x (1+i)^3 / the Age 68 Annuity rate

method (assuming age 68), for a plan that provides for a full pvab "wrong", as the auditor says, is the question.

The auditor is wrong in my opinion, since there was no risk of forfeiture from age 65 on.

Try to look back thru the 415 regulations for an example of their reasons why late retirement adjustments ignore mortality after NRA.

Their reasoning should carry some authority here.

[AndyH]

Is the

age 65 Annuity rate x (1+i)^3 / the Age 68 Annuity rate

method (assuming age 68), for a plan that provides for a full pvab "wrong", as the auditor says, is the question.

The auditor is wrong in my opinion, since there was no risk of forfeiture from age 65 on.

Try to look back thru the 415 regulations for an example of their reasons why late retirement adjustments ignore mortality after NRA.

Their reasoning should carry some authority here.

Thank you. And, yes, the 415 regulation does support this method for exactly the "non-forfeiture" reason, directly on point, both in the preamble and in the formal regulation section, and we are citing it. But the PBGC seems to play by it's own rules, so they may consider the IRS 415 regulation to be irrelevant since 415 is not an issue. We'll see.

[david rigby]

This might not be relevant to anyone else: neither the PBGC nor the IRS has any authority to define generally accepted actuarial principles.

[SoCalActuary]

This might not be relevant to anyone else: neither the PBGC nor the IRS has any authority to define generally accepted actuarial principles.

They have authority to interpret laws. That is a different issue.

[Mike Preston]

Could this whole thing be a misunderstanding? Nobody increases with interest and mortality after normal retirement age. However, to those who don't speak fluid "generally accepted actuarial principles" it may be possible that the PBGC auditor is looking at the monthly benefit and saying that the monthly benefit must go up more than just the interest rate adjustment would have it go up.

To that I would think we would all agree.

So, if the interest rate is 5% and the initial benefit is $1,000/mo, make it clear that the monthly benefit one year later is something greater than $1,050 at age 66 ($1,077.47 using 94GAR/5%).

Maybe that will make the issue go away.

[SoCalActuary]

Mike, it appears that the issue is whether you use the mortality decrement as well.

N(12) 65 / n(12) 66 is the description of the PBGC request.

1+i x a 65 / a 66 is what the original poster wants.

The math is different only on the issue of lx 65 / lx 66

The deeper argument is whether the age 65 benefit is equivalent to the sum of the death benefit from 65 to 66 plus the annuity from age 66.

Since death did not happen, the PBGC wants to throw it out of the balancing equation.

[Effen]

Nobody increases with interest and mortality after normal retirement age.

Really? Why do you say that Mike? I actually think it is pretty common to increase w/ mort post NRD using Nx(12). Wouldn’t a J&50 death benefit be deemed forfeiture upon death?

[AndyH]

No misunderstanding, unfortunately.

The auditor is using a program, and we are given screen shots so we understand the difference in the numbers. There appears to be no option in the program not to use post retirement mortality, which has the effect of the l65/l68 adjustment as SoCal describes.

We have not been able to elevate the discussion yet, unfortunately.

Both approaches are widely used, I have learned.

[AndyH]

Oh, and Effen's question I expect to be phase II of this issue in the future.

Does a REA-only death benefit create a forfeiture upon death which requires the use of post retirement mortality? Who gets to decide?

[Mike Preston]

Nobody increases with interest and mortality after normal retirement age.

Really? Why do you say that Mike? I actually think it is pretty common to increase w/ mort post NRD using Nx(12). Wouldn’t a J&50 death benefit be deemed forfeiture upon death?

It is a small plan. Therefore the plan was written so that there is no forfeiture beyond NRA for any reason. Therefore it is inappropriate to use mortality increases.

Anything else is malpractice.

I don't have time to look but I have a gut feeling that there is an ASOP that says as much.

[Effen]

I agree that for this particular question using mortality might be inappropriate, however I read your "Nobody increases with interest and mortality after normal retirement age" as a more universal statement.

I think there are times when interest and mortality my be appropriate, however not with this particular fact pattern.

[Mike Preston]

I meant it solely with respect to the fact pattern being discussed. Sorry if I was too general.