PPA Fundamentals

[Gary]

Say a new one participant plan is implemented for 2008.

Say the assumed distribution form is a lump sum.

The participant is age 55 at plan inception and NRA is 65.

So the lump sum would be taken in 10 years based on the plan terms and assumptions.

When computing the funding target (subject to a subsequent thorough review of existing regulations):

My impression is that I would compute the PV of lump sum at age 65, where the basis is the 417e segment interest rates and the 417e unisex mortality table (assuming these are all finalized or at least proposed).

In order to do this the lump sum would be based on an annuity at 65 where the first ten years of payments (65 to 75) are computed using segment 2 rates and segment three rates are used beyond age 75.

So let's say the pv at 65 is 200,000.

Then the funding target would be the pv of a payment of 200,000 made in ten years and thus it would be v ^ 10 * 200k, using the segment 2 rate.

Does the above make sense? As opposed to if an annuiity were the form of payment and the pv would be computed using the funding segment rates and funding mortality tables (not unisex).

In conclusion my impression is that the mortality table and the segment rates for 417e purposes is different from 430 purposes.

Shortfall Amortization

Say a new one participant plan is implemented for 2008, where the participant is age 60 (with 5 years of past service counted) at inception and NRA is 65 & 5. As a result the employee would receive his pension in 5 years.

If the participant received a lump sum in 5 years, where the plan were terminated at that time, the question is:

Should (or must) the initial funding target be amortized over 7 years even though plan is only expected to be in existance for five years? If it is 7 years, then of course there may be a relatively higher final plan contribution to fully fund the pension.

Thanks.

[Andy the Actuary]

It would seem the funding target would be calculated by discounting to age 55 the expected payments from 65 to 75 using the second segment rate and discounting to age 55 the expected payments from 75 and thereafter using the third segment rate. This assumes your lump sum basis is 417(e). If not (say for example, '94 GAR and 5%), then as Sir Blinkeford, the triclopsian denizen of the deep argued (and I now concur), you would use the single second segment rate to discount the PV at 65 to age 55. See: http://benefitslink.com/boards/index.php?s...le+segment+rate

The regs. say 7 years for amortization re:430. Provided there was room in the max deductible, you could always amortize faster.

If anyone disagrees with these contentions, please refer to my disclaimer.

[mwyatt]

Is this a maxed out plan? What are you putting in for actuarial assumptions? Look to the proposed asset/liability regs for lump sums subject to 415 limits.

I have plans where AE is set @ 94GAR 5.5%, benefit is the 415 maximum. Comfortable within those guidelines using the PV calculated in that case @ NRD using those assumptions, and then discounting (interest only, no pre-ret mort) using the segment rates (based on conversations at the EA Dialogue meeting for small plans and the rational for ignoring pre-ret decrements in a small plan).

[Gary]

Ok, good points.

In conclusion it seems to me that I can compute the lump sum at age 65 using 417e basis, with it limited to 415. Of course with small plans this usually means 415 in many cases.

One key being that if computed under say 417e (or other non 415 rates that are in plan) then it is computed under segment rate 2 for ten years of payments (years 11 through 20) and segment 3 thereafter.

Thanks.

[Mike Preston]

The proposed regs require that the present value be determined for the lump sum ignorant of 417(e) interest rates and instead, appropriate use of the funding rates. 417(e) mortality is used, however.

Let's presume that the present value is being calculated 20 years before benefit commencement. This crystalizes the concepts, I think. One first determines the presumed lump sum to be paid 20 years hence. That amount is determined using the third segment rate and 417(e) mortality. Assuming a 6.09% third segment rate, that value is 131.41876.

One then discounts that amount to the present day (assuming no pre-retirement mortality) using that same third segment rate: 6.09%. So, the present value at age 45 of a benefit commencing at age 65, assumed to be payable as a lump sum at age 65 is 40.28732 for a 1/1/2008 valuation using funding segment rates with phase in.

If there is pre-retirement mortality in play, then the discount from age 65 to age 45 would involve that mortality discount (for example, the 2008 Male Combined table).

Have I just repeated what others have said above or is my description different?

[Andy the Actuary]

The proposed regs require that the present value be determined for the lump sum ignorant of 417(e) interest rates and instead, appropriate use of the funding rates. 417(e) mortality is used, however.

Let's presume that the present value is being calculated 20 years before benefit commencement. This crystalizes the concepts, I think. One first determines the presumed lump sum to be paid 20 years hence. That amount is determined using the third segment rate and 417(e) mortality. Assuming a 6.09% third segment rate, that value is 131.41876.

One then discounts that amount to the present day (assuming no pre-retirement mortality) using that same third segment rate: 6.09%. So, the present value at age 45 of a benefit commencing at age 65, assumed to be payable as a lump sum at age 65 is 40.28732 for a 1/1/2008 valuation using funding segment rates with phase in.

If there is pre-retirement mortality in play, then the discount from age 65 to age 45 would involve that mortality discount (for example, the 2008 Male Combined table).

Have I just repeated what others have said above or is my description different?

In these times when sifting through answers is not so easy, your example helps. Modifying your example, if the Plan's lump sum basis was '94GAR and 5%, you would calculate the lump sum at 65 and then discount to age 45 using the 6.09% rate. if the lump sum were discounted to age 55, you would discount using the second segment rate.

[AndyH]

So in Mike Preston's example combined with a.t.a.'s extension, the lump sum at age 65 is the greater of the 5% amount and the 6.09% amount (but not more then 415), and this is discounted to age 45 at 6.09%, right?

[Andy the Actuary]

So in Mike Preston's example combined with a.t.a.'s extension, the lump sum at age 65 is the greater of the 5% amount and the 6.09% amount (but not more then 415), and this is discounted to age 45 at 6.09%, right?

I would suggest the greater of two pv's: (a) 417(e) of 415 limited benefit at 65 using segment rates and 417(e) mortality post retirement -- this is Mr. P's example and (b) PV of 415 limited benefit lump sum at 65 using Plan Rates and discounted to 45 using 3rd segment rate.

Where I believe the distinction would come in more pronounced is if you were valuing the PV at age 63, then under calculation (a) all 3 segment rates would be used where as under calculation (b) only segment rate one would be used. In short, in (a) you're going to be valuing a PV at 65 not using seg 1, 2, and 3 for <5,5-20, and 20+, but for x (years), 5 to 20, and 20+ years; or 20-x and 20+ years; or 20+ years

I suspect there are other reasonable approaches.

[mwyatt]

The pertinent stuff I see is from the proposed regs on measuring assets and liabilities, namely 1.430(d)-1(f)(iii)(B) and ©.

My reading of © is that if you have a lump sum being paid using the greater of 417 and some plan AE, that you need to take into account the fact that the lump sum could be higher than calculated using 417 (which makes sense). If lump sum will only be paid using 417 rates, then you use the regular funding segment rates as your "estimate" of future 417 rates for valuation purposes.

This of course begs the logical conclusion that the segment rates are reflecting a yield curve of various maturity dates, and that the current 30-year rate yield curve data point isn't the market's interpretation of what short term rates will be 30 years from now. The logical thing if you wanted a more accurate assessment at projected retirement would be to value the benefit at NRA using the segment rates, then discount that back to now using the segment rates. BWDIK...

[Mike Preston]

In these times when sifting through answers is not so easy, your example helps. Modifying your example, if the Plan's lump sum basis was '94GAR and 5%, you would calculate the lump sum at 65 and then discount to age 45 using the 6.09% rate. if the lump sum were discounted to age 55, you would discount using the second segment rate.

Correct on both counts. I did not discuss plan rates or the interaction of 415 limits. Both still apply. So, the real value at age 65 in my example is:

1) Take the greater of the 417(e)/6.09% value (where you would use a combination of the second segment rate and 6.09% as the third segment rate if using your 10 year example) and the plan rates actuarial equivalence value

2) Limit that to the 415 lump sum (5.5%/correct mortality table)

And then discount that for the 20 years (or 10 in your example).

[Gary]

I think the concepts and application of the concepts are clear.

I know I need to read all the available PPA regs to have the necessary background.

In terms of actuarial basis, my impression is the following:

417e - is based on the funding segment rates (I believe the transitional segment rates and not the segment rates for a newly implemented 2008 plan?) and the new 417e unisex mortality table

415 - it appears may still be GAR94 and 5.5%? Wasn't sure if it changed for PYB in 2008

Thanks.