More about MVAR calculations with Lump Sum

[Guest pension222]

I've been reading the posting from late last year regarding the calculation of MVARs when the plan pays a lump sum.

A posting on 10/31/02 from AndyH contains the following methodology.

**************************

So, assuming that the plan's QJSA is Joint & 50%, and using the above as a guide, I think the process is as follows:

1. Take the actual lump sum, divide that by the APR for an immediate Joint & 50% survivor annuity assuming the spouse is the same age. Then multiply that by the J&50% at the same age using testing assumptions. This I would call a "normalized" lump sum.

2. Bring that forward from AA to TA at the testing interest rate.

3. Then divide that by the life annuity APR at testing age using testing assumptions. Then divide that by testing comp and then by testing service. That is your MVAR.

This assumes of course that the lump sum is the most valuable accrual, which would be true unless there is a subsidized retirement benefit of some sort.

**************************

If the 50% J&S benefit is not subsidized, i.e. it is determined as the actuarial equivalent of the normal form life annuity, do we need to consider this benefit in calculation of the MVAR at all? It seems to me that using the plans conversion factors, the 50% J&S and life annuity are of equivalent value.

If the plan does not pay any early retirement benefits or lump sums, don't we only need to test the normal form benefit, and if the normal form is a single life annuity, there would be no conversion?

As a corollary to my first question, why does the above process for calculating the MVAR from a lump sum that has presumable been "subsidized" by the low 417(e) rates involve conversion through a 50% J&S annuity?

Does anyone have any guidance regarding why this is the accepted methodology (which from my reading, it appears to be).

[Guest archon]

By arguing that the plan rates are "reasonable" in the definition of "normalize" in 1.401(a)(4)-12?

{Note - please read this as very much an additional clarifying QUESTION from someone who does not by any means know!}

[Guest archon]

Forget my last comment above. I'm off in left field. Now I see what you are saying. The reason you suggest you needn't do the MVAR calculation in your situation is the same as the reason we don't need to do so in traditional DB plans. That is, the "most valuable QJSA" is the one at normal retirement age since all those at earlier ages are calculated off the benefit payable at normal retirement instead of the current cash balance account.

If I'm following, I hope you're right!

Anybody who knows listening?

[Guest pension222]

Thanks.

This is a traditional DB plan. It does not pay an early retirement pension, death and disablity benefits are fully vested PVAB, and the 50% QJSA is the actuarial equivalent of the normal form benefit which is life only.

My contention is that there is no need to normalize the 50% QJSA because (under the plan's equivalence rules) it is not more valuable than the life only normal form benefit.

Now if a plan pays a lump sum then I can get behind the methodology of calculating the lump sum payable as of the testing date using 417(e) rates if this produces the largest lump sum, projecting this to testing age using standard interest and converting to a life only annuity using standard mortality.

However, if the largest lump sum is calculated using the plan's actuarial equivalence factors (remember, 417(e) sets a minimum), then I would argue that this lump sum is no more valuable than the accrued benefit payable at testing age upon which it was based since it is the actuarial equivalent of the benefit accrual.

I put this forward because I have never had to think about lump sums when peforming this test. Overall it just makes sense to me that unless a benefit form is subsidized under the plan (for example 417 lump sums that are greater than that determine using plan equivalence, or early retirement benefits or 50% QJSAs that are greater than that determined as the actuarial equivalence of a single life annuity, assuming a SLA is the normal form) then it is not more valuable than the normal accural and does not need to be normalized.

[AndyH]

pension222, I had the same question and reaction, but I've been convinced that what you suggest makes sense but does not comply with the regulations.

I don't think that the regulations anticipated the 417(e) rate as being the most valuable accrual rate because interest rates were so high back then.

If you have access to ASPA exam reading material (C4) or conference summaries you will see that it is not commonly (if at all) done your way.

[Guest pension222]

Thanks Andy,

I cannot tell from your email if you think that we should ignore 417(e) lump sums or not. I did notice you just made a follow-up posting on another thread about this topic and it suggests to me that you might be willing to ignore the subsidization of the lump sums by the very low 417 rates. What is your opinion on this?

If you do think we need to include them then if my methodology is not correct, can you tell me how to do it and why this is the correct method? I've seen an example in one of the prior posting but want someone to tell me why it is correct. This one runs the calculations through a 50% J&S payment on the way to normalizaton.

Also, do you agree that if early retirement benefits and the QJSA is not subsidized (i.e. it is the actuarial equivalence of the normal form benefit which I will assume for the sake of discussion is a single life annuity), that these are not more valuable than the single life normal form and can be ignored in calculating the MVAR?

Does anyone have access to the ASPA materials that Andy mentioned? I'd love to see a copy.

[AndyH]

Pension222, I need to backtrack on this before I say something incorrect. I am no expert on the MVAR calculation.

There is room for interpretation and I have over the last few years collected opinions on this. There are others here who have worked with this much more than me. The opinions used to be 50% say yes, use 417(e), 50% no. I thought it was now 100% yes, but I stumbled on that thread that you referenced where a couple of experts here who's opinions are not to be dismissed said no.

But that was before Jim Holland supposedly said yes, 417(e) must be included, so they may have changed their minds. And Jim Holland's opinion on this is good enough for me.

Regarding the J&S conversion, I don't know enough about the background to the regs to debate that subject, nor do I particularly wish to. That is just the way that I've read (in several places) is correct.

There were a couple of relevant ASPA session handouts that were on their website archives, but I haven't been able to figure out how to access them since they redesigned the website.

[flosfur]

For what it's worth, Datair's system uses the same methodology i.e. converting current lump sum to QJSA etc. However, Datair's system ignores the S417(e) lump sum for this purpose.

Pension222: Here is my take on this:

For most (if not all) small plans, the testing assumptions are generally weaker than the plan's actuarial eq. assumptions. Thus, even without the S417(e), the lump sum is more valuable than a life annuity (testing interest has to between between 7.50% to 8.50% whereas in small plan's the interest rate for Act. Eg. is generally much lower and then there may be differences in mortality tables used).

So if the testing assumptions were the same as plan's Act. Eq. assumptions and S417(e) was ignored, then there will be no need to go through this conversion computations.

[Guest pension222]

Assuming that the plan's normal form is a single life annuity, I agree that if one were to "normalize" all of the optional forms of benefit that are, according to the plan, are of (actuarially) equivalent value to the normal form that some of the optional forms (such as 50% J&S and a lump sum payment) may turn out to be of greater dollar value as a normalized SLA than the original SLA.

My question is, are they really more valuable than the original benefit? For sure under the plan, unless they are somehow subsidized (say by application of the 417 lump sum factors) they are not more valuable than the original SLA on which they are based.

So if the original unsubsidized optional form of benefit is actuarially equivalent to the normal form SLA, is it really more valuable just because when it is "normalized" the resulting SLA is of greater dollar value than the original benefit?

My understanding of the concept of the most valuable accrual rate and resulting normalization is that they are to take into account any subsidization of benefits such as subsidized early retirement benefits or a subsidized QJSA, neither of which specifically actuarially equivalent to the underlying normal form SLA. If I am wrong (and I admit that I might be) can someone please point me in the correct direction? Any citations would be greatly appreciated.