Rate Groups

[Archimage]

If I give one HCE an allocation rate of 0% and I give two other HCEs an allocation, do I have 2 or 3 rate groups for 401(a)(4) testing purposes?

[AndyH]

The term "rate group" is defined in 1.401(a)(4)-3©(1)

"For purposes of this paragraph ©(1), a rate group exists under a plan for each HCE and consists of the HCE and all other employees (both HCEs and NHCEs) who have a normal accrual rate greater than or equal to the HCE's normal accrual rate, and also have a most valuable accrual rate greater than or equal to the HCE's most valuable accrual rate".

Seems to me that if the 0% HCE is not excludable for purposes of the general test, you've got 3.

p.s. It is also defined in 1.401(a)(4)-2©(1) as existing for each HCE and consists of each HCE and all other employees in the plan who have an allocation rate greater or equal to that of the HCE.

[Archimage]

That is what I thought. When I was running the test in Relius, however, it would not make the 0% HCE its own rate group. Anyone have any ideas why or how to fix this?

[Tom Poje]

you are correct, Relius simply doesn't show HCEs who are at 0 on the overall report. (I am not sure why they chose to do things that way, but nonetheless that is what you get)

It doesn't change the testing results at all since someone at 0 automatically passes, so maybe that is why they do it that way- except for a possible reference, it doesn't really matter- well, except for the fact it is confusing and makes it hard to explain to others.

I never use the 401(a)(4) nondiscrimination summary because I don't like the looks of it, and I find it rather confusing, so I can't comment on that report. If that report says 3 groups out of 3 pass when there are really 5 with 2 HCEs at 0, then that report is just plain wrong.

[Archimage]

Tom, out of curiosity do you use a different software or your own spreadsheet creation?

[Tom Poje]

Hopefully you are not a cat, because my understanding is that curiosity killed the cat, and I would hate to lose an administrator.

for proposals I am using a spreadsheet, simply for the speed and ability to change allocations and groups, etc.

I do a lot of testing for others so I use Relius, mainly because that is the software the company I work for uses on the DC side. I had worked for Corbel from 92-97 and actally wrote up notes on using the system back in the old 'glory' days of Pentabs. I had revised the notes for use with Quantech (now Relius) and have done some training at user-group meetings as well as some 'rent a tom' training. I think the boss just likes to get rid of me at times. I haven't revised the notes for a while, though for the most part, they are still valid -close to 20 pages. The folks at Quantechland liked them so much they have been using, though I have made some changes to them to correct one or two things, I imagine they have modified them a little as well to suit their needs.

some options - such as using SSRA - I can not recomend using, since that is not the normal retirement date, and I haven't seen a document yet that says I can use that as my testing assumption.

rate banding, is weak (but better than it used to be), you really have to know what you are doing.

I could actually do the calculations without any software, possibility exists I will be designing something more to suit my needs in the future. it depends on how much time I can get free.