Removing Safe Harbor Provisions

[justatester]

Need some help...

Plan has been safe harbor since 2002. Prior to the start of 2011, plan decides to remove safe harbor provision. They also elect to use the prior year testing method for 2011. Can they do that? Which I think they can...but what average would you use for the NHCEs in the test. Would you have to go back and "run" a test for 2010 to get the NHCE average?

Any thoughts would be greatly appreciated.

[Tom Poje]

well 1.401(k)-2©(1)(i) last sentence says "...a plan that uses the safe harbor method....is treated as using the current year testing method..." [this it appears to be anticiapted a safe harbor may go non safe harbor]

(ii)(A) says you can switch if the plan is not a result of aggregation (unknown from your description) AND the current year testing method was used under the plan for the preceding 5 years (True based on your comments)

therefore it appears you could switch. you can't use any safe harbor contributions from the prior year (because that would be double counting)

if the situation was the same except the plan was not safe harbor, you would do the same thing. run last year's test to find out the NHCE average.

[justatester]

ok...so I can "run" the test to get the ADP averages...Are you saying for ACP, I can not include the safe harbor match...so if that is the only match my prior year average for acp would be 0.0%? Is that what you mean by double counting? Wouldn't that then be the case for all plans that switch from prior to current?

[Tom Poje]

ah, you hadn't indicated which safe harbor was used.

so this gets into some strange things:

fact: safe harbor contributions are considered to be QNECs and QMACs

Notice 98-1 said the following:

Thus, in determining the ADP for NHCEs for the prior year, the following contributions made for the prior testing year are disregarded: QNCs used to satisfy either the ADP or ACP test under the current year testing method for the prior testing year, elective contributions taken into account for purposes of the ACP test, and all QMACs. Similarly, in determining the ACP for NHCEs for the prior year, the following contributions made for the prior testing year are disregarded: QNCs used to satisfy either the ADP or ACP test under the current year testing method for the prior testing year, QMACs taken into account for purposes of the ADP test, and all elective contributions.

when the final regs came out it says

1.401(k)-2(a)(6)(vi) Contributions only used once. Qualified nonelective

contributions and qualified matching contributions cannot be taken into

account under this paragraph (a)(6) to the extent such contributions are

taken into account for purposes of satisfying any other ADP test, any

ACP test, or the requirements of Sec. 1.401(k)-3, 1.401(m)-3 or

1.401(k)-4. Thus, for example, matching contributions that are made

pursuant to Sec. 1.401(k)-3© cannot be taken into account under the

ADP test. Similarly, if a plan switches from the current year testing

method to the prior year testing method pursuant to Sec. 1.401(k)-2©,

qualified nonelective contributions that are taken into account

under the current year testing method for a year may not be taken into

account under the prior year testing method for the next year.

Now, I take that for what it says. the title of the paragraph says you can only use them once. the next sentence says You can't use them to the extent they were taken into account to satisfy any other test, including 1.401(k)-3 which is safe harbor.

I think the ERISA Outline Book looks at the example that follows in the rest of the paragraph and applies it broadly. thus since the example says when a plan switches QNECs that were used in the current year can not be used when testing under prior. but because it doesn't mention QMACs in the example it is viewed as meaning you are permitted to do so. I'm guessing that is the logic. I don't disagree often with the ERISA Outline Book, but in the case I do.