Can multiple group avoid the attribution problem?

[Guest]

I have a profit sharing plan with three participants: the owner, his daughter and a NHCE. He wants to maximize contributions for himself with single plan. I'm considering a cross-tested plan, and the EBR's are 3.62 for Owner, 4.45 for NHCE and 11.86 for Daughter. Daughter and NHCE will receive 5% contributions. Obviously, if Daughter is brought into the HCE group through attribution, I have a problem. Can I have three groups: (1)Owners, (2)Owners Through Attribution, and (3)Others? When doing the Nondiscriminatory classification test, is Daughter a HCE or NHCE? Is the average benefits test done once comparing the HCE to Groups 2 and 3, or is it done twice, once for each group? Thanks.

[Guest FredReilly]

The daughter is definitely an HCE due to attribution per IRC 318. So you definitely need to get her accrual rate at or below that of the sole NHCE or the rate group testing will never pass. As you suggest, she needs to be put in a separate allocation group from father to accomplish this based upon what you seem to be saying.

I always find the jargon concering the ABT confusing. The 70% percent test must be done once including all eligible participants of all plans of the employer. Each HCE is a member of a rate group and as long as the ratio of percentages of HCE's and NHCE's included in that rate group equals or exceeds the mid-point between the safe and unsafe harbors and the average accrual rate for NHCE's is at least 70% of the average accrual rate for HCE's, you've passed. To use the ABT, however it is my understanding that you must use groups which meet the definition of a non-discriminatory classification per the 410(B) regs. So you need to be careful how you define the allocation groups in the plan. For instance groups that consist of named individuals would not seem to meet that criteria.

[AndyH]

In many cross tested designs with sons or daughters of the owners, the most effective formula for the owner will be to "shaft" an HCE or two, i.e. bring them into a contribution rate lower or equal to most NHCEs.

With this situation and a larger group, it is sometimes effective to break it into component plans, with one component being the owner and some NHCEs, tested on a benefits basis, and the other being the son/daughter and other NHCEs, which your typically would test on a contributions basis. After you've done it a couple of times, it isn't so hard to figure out.

[Guest]

I'm still confused regarding testing. If the nondiscriminatory classification and average benefit tests are done one using Daughter as a HCE, we fail both tests. What am I missing?

[AndyH]

You have to give the daughter the lowest contribution rate, i.e. the same as Dad will always fail with this size group.

If Dad gets 5%, give the daughter 1%. That's the only way this will work as a cross tested plan. The daughter's EBAR must get below the NHCE's.

[Guest]

we try to make sure our documents define owners as

'direct owners' or 'owners other than by attribution'

and then possibly have an additional class 'owners by attribution'

[Guest FredReilly]

You need to get daughter's EBR to at or below the 4.45 that the sole NHCE has. That is the only way you can pass the Non-discriminatory Classification Test. In that situation you will always pass the ABT given the circumstance you posed.

This means the daughter's contribution rate will have to be below 2%. I presume that the main purpose of the plan design is to maximize dad while minimizing NHCE cost and to use the 5% gateway. In that case daughter must always have a very low contribution rate.

[Guest]

Are you saying that my formula should be something like:

Group #1 (Dad): 17.65%

Group #2 (Daughter): 2.00% or lower, whatever it takes to pass

Group #3 (NHCE): 5.00%

[Guest FredReilly]

I assume that he has $170k comp and gets $30,000 contribution. If so, that is correct. You didn't say what your acutarial equivalence assumpitons are (8.5% and UP-84?)

Some other assumption may get a better balance of EBR's between dad and daughter so that daughter could get a little better, but probably not much.

To sum it up, the combination of contribution percentages and assumptions must produce EBR for daughter which is no greater than for NHCE. When that happens there will never be an issue with the ABT.

Just curious, are you with Grant Thornton? If so please give my regards to Joan. You can send an e-mail per my profile.

[Guest]

Yes, I used the 8.5% UP-84. No, I'm with Virchow Krause, another accounting firm. I've discovered that Daughter may reduce hours to under 1,000 per plan year. Based on her contribution, it might be better to drop her under 1,000 hours (which are required for a contribution-no 410 problem since she is HCE) and contribute for Owner and NHCE only. What do others think?

[Guest]

watch out for possible top heavy issue - usually cross tested are top heavy, and if your document doesn't say keys excluded you may be stuck. I would go with 1983 IAF mortality table, it produces slightly better results if things are real tight. (Unless of course your document lists what 'assumptions' to use.

[Guest]

Thanks Tom, I had forgotten the top-heavy issue. I'm usuing a volumne submitter, so I'll make sure key's excluded.