profit sharing Any way to fix this profit sharing spread this year?

[Mleech]

A plan came on with us earlier this year, this is our first time doing testing for them. Owner wants a projection of what it'd look like to max out profit sharing with new comp (they've never done profit sharing before). Right now their plan doc has 3 month wait, no hours or age requirement, and monthly entry for all sources, including safe harbor. Owner has two kids, 12 and 14, which get a small paycheck, defer some, and get safe harbor money. This causes some wild numbers in 401(a)(4) testing because of their age; the $330 of safe harbor received by one kid means I'd need to get 5 NHCEs up to ~27 EBAR. Essentially, there's no way to max out the owner without giving wild contributions to everyone else because of those two kids.

Our plan is to amend their document for next year to either have an age requirement or exclude HCEs from the safe harbor contribution, along with some allocation conditions and other small provision changes to make this much smoother next year. That said, is there anything at all we can do for this year to make this spread better? I've seen conflicting information about the use of statutory exclusions for 401(a) rate group testing & struggling a bit to wrap my head around if there's any way we can make this work. Any input would be much appreciated!

[CuseFan]

You can parse otherwise excludable employees but must test that group separately and you don't get a free pass because that group has HCEs. You may want to try restructuring and parse the young HCEs with older NHCEs and test that group on contributions. The question then is whether your restructured plans can pass using ratio percentage or if you can pass average benefits with those young HCEs (might need to calculate AB% on contributions as well).

[justanotheradmin]

CuseFan has the best suggestion. Restructuring is sometimes also known as component testing. both testing groups would have to meet minimum gateway. Generally the youngest HCE + older NHCE are put in a group and tested on a contribution allocation basis, the older HCE and the younger NHCE are tested on future basis. 

For next  year - I would not suggest adding in allocation conditions - if you do , it handcuffs who can receive an discretionary employer contribution. Your plan document might waive allocation conditions for purposes of meeting gateway, but what if a younger NHCE left partway way through the year, and it would be advantageous to testing to give that person a larger contribution? you would not be able to if the plan has a last day employment condition. That person would be limited to the Safe harbor, and perhaps gateway. 

I do suggest that safe harbor nonelective go to NHCE ONLY in plans that are cross-tested. If it works out to give the HCE 3% and not skew testing, that can always be accomplished with a discretionary contribution. 

Alternatively - for some future year - if the plan is small, owner comp is high, and general participation is low - sometimes it works out better for the plan to use safe harbor match. The owners defer the maximum, if their comp is high they can receive a large match, and then make up the difference in discretionary employer. Depending on the specifics, it might get the owners to the maximum overall limit with less minimum to the NHCE to pass testing. May not work as well if the plan is top heavy. But something to consider sometimes. 

[justanotheradmin]

If the kids are not deferring the maximum - perhaps their tax advisor could educate them on IRAs. If they are eligible to make IRA contributions, might be better than messing up the 401(k) testing with deferrals. They could still be eligible for the plan, and help testing, but a way for them to still get tax savings, but not skew testing. 

[CuseFan]

@justanotheradmin for the win!

@Mleech this is excellent advice in each of those posts.

[BG5150]

23 hours ago, justanotheradmin said:

both testing groups would have to meet minimum gateway

Is this true?  Often in component testing, one group (the one with the young HCE) is tested on a contributions basis and the other one(s) tested on accrual basis.

Does the contributions basis 'plan' have to satisfy the gateway?  It's its own 'plan' after all....

'Tis been a few years since I last did one of these 'boutique' calculations as a former employer used to put it.

(And I kind of remember that the combined sub-plans, if you will, must satisfy the ABT?

I'd have to dredge up my notes....

[Bri]

Yes. both sub-plans still need to satisfy the overall gateway.  

[justanotheradmin]

if the two testing groups are in the same plan, yes overall gateway must be met if one is tested on an accrual basis. Answer might be different it it is actually two separate plans being permissively tested together, and each plan covers two different sets of people (not the same people). Like a plan for division or Company A, and a different plan for division or Company B, assuming A and B are a control group or some such. i don't know for sure the answer in that scenario. 

[C. B. Zeller]

The gateway may not be tested using component plans. See 1.401(a)(4)-9(c)(3)(ii)

[John Feldt ERPA CPC QPA]

Meaning, you evaluate the plan as a whole for determining if all the NHCEs got a gateway.

Keep in mind if the OEE group itself is not included with the others in testing, then no gateway applies to the OEE group - unless for some very strange reason the OEE group had an HCE getting more as a percent of pay than the younger NHCEs in that OEE group and you cross-tested the OEEs - but I’ve never seen that happen.