Individual aggregate funding

[FAPInJax]

I seem to remember IRS commenting on the following situation but can not find the response. The comment was generally that IF there are NO active participants the deduction of the difference between the present value of benefits and the assets is NOT reasonable. Therefore, I will ask the learned members of this forum what they think.

We will make the first case 'simple'. This is a one man plan where the participant reached normal retirement age (sometime prior to the valuation date). The valuation is performed at the beginning the plan year.

Valuation 1/1/2005

Is it legal to value the participant's actuarial increased benefit and develop a contribution equal to the difference between the present value of this benefit and the assets at 1/1/2005??

Can this method be maintained when the participant does not retire but continues working and a new valuation is prepared at 1/1/2006?

Is the answer any different IF there is more than 1 participant (all inactive)??

Thanks for any and all responses.

[Blinky the 3-eyed Fish]

Pepi: Tell me more! I want to know ALL the constellations!

Homer: Well, there's... Jerry, the cowboy. And that big dipper-looking thing is Alan... the cowboy.

Pepi: Oh, Papa Homer, you are so learned.

Homer: Heh heh heh. `Learned', son. It's pronounced `learned'.

That being said, in either situation, one person at NRA or all term, you have no one that has a future working lifetime to spread the costs over. You are correct that the IRS has espoused (and I agree) that IA is not reasonable to use in either situation. Now you could assume the person at NRA will actually retire at the end of the year. That might be fine to use IA in the first year, but to continually use it and fund the unfunded amount would too be unreasonable, IMHO.

[SoCalActuary]

I guess I'm clueless here. I thought the Court opinion in the actuarial audit cases included the thought that the plan should be sufficient to pay the benefits when the participants have all retired.

To me, that implies that the final year of funding under IA should be the difference between assets & liabilities. So tell me again why that's unreasonable.

[Blinky the 3-eyed Fish]

By actuarial audit cases, are you talking about the small plan audits of the late 80's or so? That pre-dates my existence in this business by a long shot, so I can't comment. However, there are these from the Gray Book that I lifted from a post of pax's. How you weigh court cases versus current IRS thinking is up to you. I know that no other funding method offers the availability to fully fund benefits in a final year. Now PBGC covered plans have that option and there's UCL for those not PBGC covered, but of course, that is separate from the funding method.

QUESTION 93-10

Aggregate Funding Method -- No remaining actives

A sponsor has a defined benefit plan at a location where the company closes down the operation for economic reasons, effective as of the first day of a plan year. After the closing, the plan will cover only terminated vested and retired employees and their beneficiaries. The funding method is the aggregate cost method. Assume the following with respect to the plan year:

Present Value of Benefits: $20,000,000

Valuation assets: $19,000,000

Under these circumstances, is the average future working lifetime for each participant equal to 1 year and thus the normal cost and maximum deductible limit equal to $1,000,000? Would the answer be different if the closing did not occur until sometime during the plan year?

RESPONSE

No. It would not be acceptable to use one year as the average future working lifetime where there are no remaining actives. One acceptable approach is to use the expected retirement ages as if the employees were still active.

QUESTION 99-6

Funding: Spread Gain Method for Plan with No Future Benefit Accruals

Plan A uses the aggregate method (level percentage of compensation) to determine funding requirements. As of December 31, 1998, all active participants are terminated, but the plan remains in existence.

(a) How is the normal cost for 1999 determined?

(b) Would the answer be different if the plan was frozen, but participants continued in active service with the sponsor?

RESPONSE

(a) The funding method must be changed to an immediate gain method. Under any immediate gain method, the normal cost would be zero and the accrued liability would equal the present value of benefits. Any unfunded liability initially recognized should be amortized over ten years.

(b) No. The response in (a) would also apply to a frozen plan covering participants who are still employed by the plan sponsor.

[david rigby]

Gray Book Q&A’s have touched on this issue.

Q&A 93-10 and 99-6: http://benefitslink.com/boards/index.php?showtopic=19350

Q&A 2004-20: http://benefitslink.com/boards/index.php?showtopic=23970

[SoCalActuary]

After reading Frank's post again, here's a second opinion.

1. IA for a person at retirement age, it seems obvious you can fund in one year.

2. IA for a terminated plan for employees not at retirement age, I agree that you should convert to unit credit funding with 5 to 10 year amortization for 412 purposes but with immediate deduction for PBGC plans of UCL.

[Blinky the 3-eyed Fish]

I agree with that.

[FAPInJax]

Thanks for all the replies!

So, an actuarial valuation was performed at 1/1/2003 when the participant was 64 years old (due to retire 12/1/2003) and the final contribution was determined. I gather from the responses that no one has problem here.

The actuarial valuation as of 1/1/2004 finds the participant still alive and kicking BUT beyond normal retirement age.

I feel comfortable making an assumption regarding his anticipated retirement date (it may be one year or several years as far as assumptions go). Assuming a single year, then funding for the actuarial increase in benefit (or the formula if larger) works just fine. The only reservation I saw was Blinky felt that if he did not retire at the end of the 2005 year then it might be a bit much to keep presuming a 1 year future assumption.

The problem I ran into was that the participant is at the 415 compensation limit as of 1/1/2004 and does not project an actuarial increase. The client running the valuation is using 7% pre for funding BUT actuarial equivalent of 6%. Therefore, the valuation produces a contribution less than desired.

What about just calculating the PVB at 1/1/2004 and subtracting the assets was the question?? This does not appear to be reasonable because he can not get the money out of the plan now. His benefit is not increasing and the lump sum is dropping. The plan could be amended to go to a 5.5% actuarial equivalent to get the maximum lump sum (but let's not go there at this time)

SoCal has no problem with the one year funding while the person is at retirement age. What happens at 1/1/2005 (assuming the assets go down for example)?? Can the same method be employed? Can the method be maintained several years into the future - one year at a time??

I just hope to convince the gentleman to take his money and run<GGGG>.

[SoCalActuary]

The 415 compensation limit is the critical problem here. The participant must "USE IT OR LOSE IT" (screaming caps added intentionally :angry: )

If the participant is scheduled to take late retirement in your scenario, then the participant must make a decision about the benefits due.

1. Pay it out

2. Forfeit the benefit, if the plan allows it.

The valuation then reflects the same treatment you would give a retiree. The PVB at the beginning of the year is based on the actuary's best estimate of the future benefits to be paid. If the actuary assumes an annuity payment and is comfortable with 7% interest, then the PVB is an immediate annuity on the beginning of the plan year (valuation date) using funding assumptions. If the actuary assumes a lump sum is the intended benefit, then use the plan 6% or the 417(e) rate for the lump sum. This can be done with the post retirement interest assumption only if you value the participant as a retiree.

However, I understand that this can cause a problem for your temporary annuity calculation. The result you need is a temporary annuity value of 1 for this participant, and showing the person as a retiree in your valuation would not produce that result.

As another choice, you can assume a one year later retirement with an assumption that one year of benefits is due and payable, as in a manual adjustment of the plan assets. True asset of $800,000, with 60,000 benefit payable, so you manually adjust down to $740,000 for valuation purposes. Or enter a trust fund record showing the payment made one day before the valuation date.

[david rigby]

...if he did not retire at the end of the 2005 year then it might be a bit much to keep presuming a 1 year future assumption.

IMHO, we should not presume anything unreasonable just because the retirement assumption for an over-NRA employee is changed in multiple years.