Safe Harbor 401(k) with Cross Tested PS

[Guest DTrom]

A plan uses the safe harbor matching contributions to pass ADP. The plan also includes a cross tested profit sharing provisions. The only eligibility for the 401(k) is the attainment of age 21. The PS requires age 21 and one year of service.

The Plan Sponsor would like to only offer the safe harbor match to the participants who have met statutory entry requirements.

I know that the Safe Harbor rules have some added complexity when it comes to disaggregating the statutory excludable employees; such as the ADP test must be run if an HCE is in the excludable group.

One question I had was, if the 401(k) portion of the plan is being tested by disaggregating the statutory excludable employees, that does not necessarily mean that the PS must be tested using disaggregation as well?

The Plan is not currently top heavy, but if the plan were top heavy and the otherwise excludables received the top heavy contribution would they then become part of the profit sharing plan? If the PS was then tested using disaggregation, they would not need to receive the gateway?

Any other complications to watch out for?

Thanks.

[AndyH]

You have 3 plans, k, m, and nonelective. You can test each other separately using separate 410(b)/401(a)(4) techniques since they are mandatorily dissagregated for 410(b) and 401(a)(4).

Regarding the question about "would they [excludables getting top heavy minimums] become part of the profit sharing plan" depends. Whether or not they get an allocation depends upon the plan provision. Whether or not they are included in the a(4) test depends upon your testing methodology. You could exlclude them, which actually means test them separately, which means nothing if there are no HCEs in that group.

[Guest DTrom]

Thanks Andy. If they were included in the 401(a) test then since they are benefitting (the top heavy contribution), would they then need to receive the gateway?

Would it uncomplicate things if the wait for all portions of the plan was one year?

[AndyH]

Yes, remember your 401(a)(4) test would fail if you included an NHCE who did not get the gateway, because then you would not be allowed to cross test.

The disaggregation rule was put in to allow early k participation, but if it is causing top heavy minimums then maybe a 1 year wait for all does makes sense.

P.S. DTROM, you understand that a 3% SHNEC is usually a better choice in a safe harbor k plan with a cross tested provision, don't you. If not Tom will fax you his outline which is supposed to explain that, right Tom! Or better still, sit in on his upcoming ASPA session.

[Tom Poje]

Andy:

I would generally agree, with a cross tested plan the SHNEC is better than the SHMAC, generally because you are forced into giving 3 % anyway.

one note of caution. you can simply 'give' the gateway unless the document allows.

e.g. If I recall correctly, the Corbel document has a snap on amendment to accomplish this.

not sure if my notes will contain everything. part 1 is a walk through of safe harbor plans (mainly because of the questions that keep coming up on this website) plus what the proposed regs includes

for safe harbor plans, and then I get into a discussion on the cross tested end, though I did lean more towards some 'tips'. I mean what can I say, yes, you include the SHNEC in rate group testing and it counts towards the gateway. No, you don't inlcude the SHMAC excet in the avg ben % test.

Survived hurricanes 1 through 4 down here. joy of joys!

[Guest DTrom]

Andy,

Thanks again. So far the plan is well below top heavy, and I am hoping it will continue that way. It is a good size group with only two owners (two of three HCEs and the only two keys). The third HCE is not included in the top tier. So far the cross testing with the safe harbor match has worked well. I am aware of the "triple dip" of the safe harbor nonelective though, and appreciate your pointing that out.

Tom,

Thank you as well; our document (DATAIR) does allow for a top heavy minimum only, in the event of cross testing.

[Guest DTrom]

Tom,

Not sure if this is the right place to ask this question, but...

Continuing with the scenario above, if I wanted to disaggregate the 401(k) portion of the plan, but not the 401(a) portion of the plan,

can Relius Admin handle that? It seems to me that with Relius it is all disaggregated or all not disaggregated. Would I need to run separate tests, and print separate reports, and probably have to edit the 401(k) reports to get rid of the 401(a) information?

Not sure if I explained that correctly, but thanks for any input.

[Tom Poje]

well, you can certainly code the plan specs so safe harbor only goes to those having met 1 year wait.

testing wise, you can tell non discrim system to include or exclude otherwise excludables.

(In fact you could do the same with the ADP/ACP test, but since it is safe harbor that shouldn't matter. But I dont think there is an easy way of performing the ADP including otherwise excludables and the ACP excluding them)

If you are printing the coverage reports, you would have to print the reports twice - once with and the other without and cross out the tests you dont want. I dont know of any other way. Again, since it is safe harbor , it would n't really matter since all benefit

[Guest DTrom]

Thanks Tom.