Non-Calendar Year Safe Harbor 401(k) Plan

[Guest s.c.semler]

An interesting Plan design possibility came up in discussion and I would like some feedback as to its legality. Employer has a safe harbor 401(k) plan with a 7/1 to 6/30 plan year. It seems the HCE could defer the 402(g) limit between July 1st and December 31st and then defer the 402(g) limit for the following year between January 1st and June 30th. Prior to the beginning of the next plan year, he could revoke the safe harbor election. The following year he would again go back to the safe harbor plan and defer the 402(g) limit ad infinitum.

Therefore the owner would achieve the maximum deferral each calendar year while only having to make the safe harbor contribution every other year.

Any thoughts?

[Guest pineapple]

what about Treas. Reg. 1.401(a)(4)-5(a) - series of amendments may not discriminate in favor or HCE.

[Tom Poje]

sounds clever, and then the hce puts in the catch up as well.

I suppose it makes more sense if the plan isn't top heavy, since you won't save much if you have to cover top heavy.

(except for vesting)

[Guest s.c.semler]

Tom:

It is our thinking that the top-heavy minimum contribution would not be required since the HCE would not be receiving an annual addition in the "off" years.

[Tom Poje]

so there will be no profit sharing contribution.

pineapple has a valid point when he says you can't keep amending if it significantly favors HCEs.

and technically, opting in and out of safe harbor every year is considered amendmens. on the other hand, every other plan year the HCE gets 'nothing' since you indicated there will be no profit sharing, so one could argue the amendments do not favor the HCE.

hmmm. I suspect the IRS would catch on eventually and put a stop to something like that with a technical correction of some type. In other words, I wouldn't recommend the scheme, it just seems like you are pushing it.

From another perspective, such a plan is a good way NOT to retain good help. but of course, a lot of companies don't think about that.

[Tom Poje]

as another possible strike against your idea

Notice 98-1

The notice also includes an "anti-abuse provision." If there are repeated changes in testing procedures or plan provisions that have the effect of distorting the ADP or ACP test to the benefit of the HCEs, the plan will be deemed to have failed the tests if the principal purpose of the changes was to achieve this result.

While this applies to switching testing methods from current to prior to current to prior etc, I would hold the concept or idea would hold true for safe harbor plans. yes, its true the regs don't say that...yet...but if you start fooling around it won't take the IRS long to step in.

[Guest pineapple]

As Tom and I have indicated, it's questionable whether the idea is allowable. (My personal opinion is that it's not.) Regardless, I'm positive that if the IRS discovers plans are abusing the safe harbor in that manner, they will shut the loophole down (i.e., only calendar year plans can use safe harbor).