Nondiscriminatory Classification Test

[Gary]

Is a ratio greater than the midpoint between the safe harbor and unsafe harbor a reasonable and prudent target for passing the nondscriminatory classification test when testing for non discrimination in amount of benefits?

And when the ABP ratio is close to 90%?

And when many of the employees excluded from profit sharing allocations make over $100,000 but are not considered HCEs due to top paid group testing and thus bring down the NHCE ratios?

Curious to hear thoughts.

Thanks.

[SoCalActuary]

You check each of the required tests, and if it passes, you did well.

It sounds from your report that you achieved the proper result.

By the way, this discussion belongs in another forum - Cross-tested Plans.

However, you may be more likely to get answers here

[AndyH]

Gary, if the question is do most plans pass by getting all rate groups above 70% then I would say no, that is not a necessary result. I do not think that most plans, db or dc, are designed to that standard. I just would not design a plan that tests with an ABPT close to 70%, which you have acknowledged. It sounds to me that you have set a good goal. But I would also make sure that none of the rate groups are close to the mid point.

[Gary]

Thanks.

I guess another way of making my point is let's say a plan has an ABP of close to 90% and many employees not receving an allocation are NHCEs earning over 100k, then do those facts support a facts and cicumstance criteria to support using the midpoint between safe and unsafe harbors as a level in which each rate group for ND in allocations must meet? Instead of going the whole nine yards up to and over the elusive higher safe harbor percent.

Regarding permitted disparity.

For a DC plan must PD be performed on the allocations and then such rates converted to cross tested accruals? Or instead, could the allocations be cross tested to accruals and then apply PD on the accruals?

That is, is there a requirement regarding the order of things or is it essentially an option?

Thanks

[AndyH]

1. I assumed you were doing a 401(a)(4) test, in which case the safe harbor percentage is irrelevant. If you instead are doing a 410(b) test for coverage purposes, then I cannot answer your question. I have no idea what would pass facts and circumstances.

2. Re pd, if you test a dc plan on a benefits basis, then you convert the ebar first and apply pd to that according to the regulations. If you test a dc plan on a contributions basis, you use those separate rules that pertain to general testing a dc plan on a contribs basis, or if it is a db you convert to allocations first.

[Gary]

Regarding Andy H's responses.

His item 2. makes total sense. However, is there a cite that explicitly says this? I didn't come across anything regarding the order of calculations in the regs related to either imputation of permitted disparity (1.401(a)(4)-7) or cross-testing (1.401(a)(4)-8).

Regarding item 1. I am referring to the rate groups passing the coverage tests.

Of course if the top-paid group election is eliminated then there are many NHCEs (i.e. Associate attorneys) not getting an allocation, who are currently hurting the rate group ratios for the NHCEs who would be moved to the HCEs, thus lowering the rate group ratios of the HCEs and thus enabling the rate groups to easily pass non discrimination. Given that the plan year ended 3/31/2006 can the election for top paid group be removed after the plan year ends? Perhaps IRS 97-45 can help.

Thanks.

[AndyH]

It is clear in my mind that -7(b)(1) tells you how to do it if you are testing under -2© (allocation basis) and -7©(1) tells you how to do it if you are testing on a benefits basis.

I don't see any ambiguity.

I'll leave the TPG election question to someone else.