DB funding questions

[Guest lerieleech]

1. A small plan has a 12/31/08 val date. The participant is projected to receive the maximum benefit as limited by the 415 dollar limit at NRD. He/she is accruing over 15 years on a participation/participation basis. This is the 5th year. He/she will take a lump sum upon retirement, so we assume that. AEQ is 5%, but of course the max LS is based on 5.5%.

Obviously, the participant has not accrued the current 415 max, but the AB he/she has accrued projects to be part of a benefit that will be limited to 415. So, do we use 5% or 5.5% for post-retirement funding?

2. Same idea. In a plan such as the one described above, in determining the portion of the accrued benefit that is used to determine funding target (and thus is based on prior years), should the 2008 415 dollar limit be applied to prior service? Or should the 12/31/07 AB be used without applying the 2008 increase?

[Mike Preston]

1. Seems like a trick question. Isn't the answer neither (ignoring the at risk portion of the calculation for 404 purposes)? Last I heard, there were these things called segment rates that we had to use for most purposes. However, if you do get to a point in your calculations where the AE under the terms of the plan is factored in, your facts argue for 5%. It is all about the accrued benefits. BOY accrued benefit and EOY accrued benefit are nowhere close to the 415 limit, so the 5.5% factor is irrelevant.

2. Another trick question. I haven't seen a plan that artificially limits the projected benefit to the 415 limit when determining the accrued benefit in a long time. However, assuming you have one of them (they aren't illegal, just not something I've chosen to use as a design for a long time), it still depends. In most cases, you would ignore prior year benefit limits completely and just use the new year's limit. However, if you are dealing with an HCE and determining the amount that is subject to the 150% multiplier for 404 purposes, there have been conflicting statements from the IRS as to whether you ignore the amount attributable to the 2008, 2007 and 2006 COLA increases when determining the portion of the FT you can multiply by 150%. Anybody think this issue has guidance?

[Guest lerieleech]

Thanks.

As for the first question, the proposed regulations (1.430(d)-1(f)(4)) say that you must take into account the probability that a certain form of benefit is elected.

In a plan like this, the lump sum is just about always elected, so you need to assume this participant will elect a lump sum. Therefore, the "stream of payments" notion at NRD becomes moot, and you need to calculate the projected value of the lump sum at NRD. Then the segment rates (or rate, as there is only one assumed payment) is used to discount the value of the lump sum back to the valuation date.

[Mike Preston]

In this day and age, I'm having a lot more discussions about annuitization than I've ever had in the past. I don't think the IRS could object to a valuation that considered the likelihood of annuitization being greater than zero percent, even for the smallest plans. I agree with you that a presumption of lump sum leaves you with a valuation of the greater of two values: a) lump sum under the plan at actuarial equivalence rates; b) lump sum value at 417(e) rates where the value is this very strange combination of current segment rates and 417(e) mortality. There is still disagreement amongst actuaries about what to do with the maturity value determined above, but most agree that it (either one) is discounted to the valuation date from date of assumed payment based on the single segment rate. So, I still say your question was a trick question because it doesn't contemplate 417(e) rates or segment rates of any kind. Perhaps if you threw a couple of examples together with numbers the above would be clearer.

[Guest lerieleech]

As for examples, maybe I will do so later.

Most small plan actuaries I know are using 100% probability of lump sum election for a plan like this. I am not saying you're wrong-- just telling you what I am aware of, and what I would do, in general.

I agree that 417(e) rates must be considered, in general. But in this case, as 417(e) rates would project to come out higher than 5% in any event (and would not matter if we are at 415), I didn't make reference to them in my question.

No tricks here... Now if I would have asked if the answer were 5% or 5.5%, and the answer was "segment rtaes," now THAT would have been a trick. But asking a trick question also requires that one already knows the answer, which I didn't.

[Mike Preston]

I think you've made my point. If 417(e) would project out to be higher than AE, doesn't that mean you should be using it? Or did you mean higher rates (which would mean a lower present value)?

[Guest lerieleech]

I think you've made my point. If 417(e) would project out to be higher than AE, doesn't that mean you should be using it? Or did you mean higher rates (which would mean a lower present value)?

Higher rates, ergo lower present values.

[Mike Preston]

Which months are those?