Imputed disparity in cross-tested plans

[Belgarath]

Imputed disparity in a DC plan tested on an allocations basis seems relatively straightforward. But it seems rather more complicated if it is cross-tested. I want to see if I'm getting the basic idea right.

Let's say it is just a PS plan, no 401(k), no other DB plan, never has been a DB plan, never has been any integrated formula. Plain vanilla.

You have obtained the EBAR's for all employees. Now you are going to adjust those EBAR's for imputed disparity.

When calculating the "permitted disparity factor" - it is apparently the sum of the annual disparity factors for each year included in the measurement period for determining the accrual rates, divided by the participant's rtesting service in that period.

When doing this for a DC plan, can you use a "measurement period" of the current plan year only, as well as the current plan year only for "testing service?" In other words, if the average annual compensation is less than or equal to covered compensation (and good luck obtaining that data) and the unadjusted accrual rate is 1%, can you simply use the lesser of (2 x 1%) = 2%, or (1% + .75%) = 1.75%, therefore 1.75%? Or is it more complex than that and you have to go back and add up the sum of the disparity factors for all prior years, etc., and perform various voodoo and sorcery?

If you have been a participant for more than 35 years in such plan, is imputed disparity not allowed (probably dependent upon answer to above?)

Thanks.

[rcline46]

1.401(a)(4)-7(b)(2) for allocations rate, -7b3 for accrual rates.

[Belgarath]

Did you perhaps mean (b)(2) and (3) for allocations based, and ©(2) and (3) for accrual based?

Sorry, I should have been more specific in my original question. What I was trying to get at is in ©(4)(iii), they say you use the "measurement period" and "testing service" as in 1.401(a)(4)-3(d)(1). And under -3(d)(1)(iii) it appears that the measurement period may be the current plan year, and under (d)(1)(iv)(B)(2), it appears that if the measurement period is the current year, testing service is 1 year. Hence my original question, can I actually have the calculation as relatively painless as in my question, or am I totally misunderstanding it is a lot more complex?

[Tom Poje]

actually look at 1.401(a)(4)-7(b)(4)(iii)(B)(2) annual disparity factor after 35 years

of course, using .75 in your example makes little sense as there are other adjustments.

.75 is used for someone who has SSRA = 65, so for most people the factor is .65 at NRA = 65.

I suppose you could have someone born in 1938. they would be age 73, but then there is an adjustment for being past NRA (as I recall)

plus to reach 35 years implies a plan has been in existence since 1976. I think I worked on one plan in my life that is that old.

[Belgarath]

Yes, you are right - so if you modify my example to use a .65 factor instead, is it that simple, or am I getting it wrong?

Modified original paragraph taking into account .65.

When doing this for a DC plan, can you use a "measurement period" of the current plan year only, as well as the current plan year only for "testing service?" In other words, if the average annual compensation is less than or equal to covered compensation (and good luck obtaining that data) and the unadjusted accrual rate is 1%, can you simply use the lesser of (2 x 1%) = 2%, or (1% + .65%) = 1.65%, therefore 1.65%? Or is it more complex than that and you have to go back and add up the sum of the disparity factors for all prior years, etc., and perform various voodoo and sorcery?

[AndyH]

Yes, you have it right for the current year measurement period pd methodology (@ .65%). Not particularly complicated.

[Guest GeerTom]

I am going to make an assumption here, that the plan does not have a rigid allocation formula. In other words, the plan says group 1 gets an allocation rate, group 2 gets another, and so on. And that these rates are broadly limited by the nondiscrimination rules.

If so, this problem may be simpler than you think.

First, the plan can test on the basis of the current year only or on the basis of the current year and all prior years. The regulations are very clear about this, as cited above.

The next step is that I have never seen a plan where all years testing made the test harder to pass. In fact, all years testing advantages usually outweigh the advantages of permitted disparity by a huge margin. This is so because in normal circumstances (prior years without cross-testing or prior year cross-testing limited by other factors) there is usually a huge underallocation for prior years when comparing the actual formula to a hypothetical cross-tested. So, figure out what the allocation would be on an all years basis, apply the other limitations (ABP, etc.) and the limiting factor will almost always be another limitation. Voila, you don't have to worry about 35 years or any other permitted disparity issues. Try it.

Tom Geer