Cross Tested ABR

[perkinsran]

We have a client with 401k safe harbor cross tested plan and a cash balance plan. All statutorily eligible employees are eligible for both plans. When we run the ABRs for the 401(a)(4) tests for the cross tested plan, do we have to look at the cash balance plan or just the cross tested dc plan?

[Tom Poje]

There is a provision in the regs that says you can test all DBs on an accrual basis and all DC on an allocation basis, but since it sounds like the DC plan 'is cross tested', that probably wouldn't work.

therefore you will most likely test everything together on an accrual basis.

by the way, you can't simply say "they received a 10% cash balance credit, the gateway is satisfied" Becasue the cash balance uses a different interest rate, you have to convert the credit into an equivalent DC contribution rate.

[Mike Preston]

You can certainly test them separately but there is a "gotcha" waiting for you if you do: if you impute permitted disparity you must limit the imputation in the plans so that it is not double counted. Obviously, it is far easier to do this if you just run one test, since most software is programmed correctly with respect to imputation on a combined plan basis.

[AndyH]

Mike, how do you test them separately? Do you mean compute them separately and add the results together for the benefit percentage? This is an average benefits test, so I think Tom cites the rule correctly, with the added option of summing separately calculated results. Agree?

[Mike Preston]

Maybe I'm not understanding. What makes you think it is merely an average benefits test that is being discussed? The average benefits test is usually discussed in the context of a 410(b) test, not a 401(a)(4) test.

Maybe we need clarification as to just what test is being run here and what, exactly, ABR means.

What I had surmised is that the plans passed 410(b) and the test that is being run is merely the 401(a)(4) rate group analysis. In that context you run each plan separately and ignore the benefits in the other plan entirely when testing a given plan.

But you still can't impute permitted disparity twice!

[AndyH]

I took ABR to be Average Benefits Rate, whatever the technical term is. So I'm sure that is all this is - we're talking about different tests.

perkinsran, please clarify.

[perkinsran]

I took ABR to be Average Benefits Rate, whatever the technical term is. So I'm sure that is all this is - we're talking about different tests.

perkinsran, please clarify.

[perkinsran]

Sorry for the delay and thanks for the response. Here is the scoop.

We just picked up the dc only portion of this plan. It is a 401k safe harbor cross-tested plan.

Employer contributions to this plan are about 5.5% per year for nonowners.

When we run the a4 test, each of our rate groups do not pass the 70% test so we have to rely on the

average benefits % tests. The prior TPA ignored the Cash Balance plans in the calculation of the

latter for a4 testing.

The two plans each pass general 410 b coverage tests since they cover all eligibles.

Does that change the answer?

[AndyH]

So you are testing a DC plan on a benefits basis, I will assume, and you don't seem to want or need to permissively aggregate the CB plan.

If each of your rate groups are at or above 70%, I believe that you do not need to consider the CB plan at all.

If one of your rate groups is < 70% then you need the Average Benefits Test and I believe that you must include benefits from all plans under this particular set of facts.

And I think the same applies to the CB plan, that they need the DC plan data if all of their rate groups are not at 70%.

If instead you were testing the DC plan on a contributions basis and the CB plan were being tested on a benefits basis then neither plan is being cross tested and you could in fact exclude the other type of plan entirely, as both Tom and Mike alluded to.

[perkinsran]

So you are testing a DC plan on a benefits basis, I will assume, and you don't seem to want or need to permissively aggregate the CB plan.

If each of your rate groups are at or above 70%, I believe that you do not need to consider the CB plan at all.

If one of your rate groups is < 70% then you need the Average Benefits Test and I believe that you must include benefits from all plans under this particular set of facts.

And I think the same applies to the CB plan, that they need the DC plan data if all of their rate groups are not at 70%.

If instead you were testing the DC plan on a contributions basis and the CB plan were being tested on a benefits basis then neither plan is being cross tested and you could in fact exclude the other type of plan entirely, as both Tom and Mike alluded to.

Andy, That is exactly correct. We are testing a dc plan on a benefits basis. It is not that i don't want to aggregate but we are not the actuary for the db cash balance plan. Our dc cross tested rate groups fail the 70% test, so I am trying to determine if I need to get benefit rate from the actuary for the CB plan to determine if the plan passes the (a) 4 test. So before I ask the actuary for

the equivalent rates in the CB plan and he says I don't need them, I just wanted to post this inquiry.