Shifting if ADP Fails

[austin3515]

ADP test fails and just $2,000 is reclassed as catch-ups (leaving $3,500 "available"). The IRS came out once before and said that if the test is failing you can still use shifting, but now you must shift for both HCE's and NHCE's, because the "plan is passing right at the passing percentage." See questions 15 and 16 of the attached.

The question is, can we shift elective deferrals over to the ACP Test in order to increase the amount reclassified as catch-ups, thus avoiding the need to shift any HCE deferrals?

Shifting___ADP_Test_Failed_IRS_Q_A.pdf

[rcline46]

First, let me say I was shocked, SHOCKED at the IRS response, but pleasantly surprised. However, because of the rule that you must pass the ADP test both before and after shifting, I do not think you can shift anything before first having a passing ADP. In your question, you do not have a passing grade before the shift.

I also hope there is only one HCE in your example. Due to the giveback rules, if you have more than one HCE, and the one causing the ADP failure is not the one getting the giveback (catch-up), you may not actually have a passing ADP test to start the shifting - that is the mechanical issue the mention in the question.

[austin3515]

However, because of the rule that you must pass the ADP test both before and after shifting, I do not think you can shift anything before first having a passing ADP. In your question, you do not have a passing grade before the shift.

You don;'t think the IRS Q&A gives the green light for shifting even a failed ADP test, provided you shift for both HCE's and NHCE's? I thought that was pretty plainly what they said.

And yes there is only one HCE. Funny you should mention that because here in the office we had that same conversation about how this clearly backfires with more than on HCE.

[rcline46]

Their example is after correction, where the ADP is perfectly passing. This is expressed in both questions. At least that is how I read the answers.

[austin3515]

The quesiton references that the shifting is being done AFTER recharacterizing excess deferrals as catch-ups. Do you want to read through one more time? I just think it veyr clearly states that this is the case.

Also, the question that follows clearly suggests that a passing grade is not a requirement to use shifting.

[Tom Poje]

having submitted the question myself, the intent of the question was if it was possible to shift after corrections were made to pass the ADP test, not

could you shift first to get out of a failed ADP test (or create catch ups or whatever). I never took the IRS response to be other than that.

1.401(m)-2(a)(6)(ii) clearly states you can use deferrals in the ACP..only to the extent that the deferral arrangenent would satisfy that test [the ADP test] for the plan year beforehand.

[austin3515]

So what you are saying is that in my example, we cannot shift deferrals to the ACP test for only the NHCE's to make the [edit: ADP]test fail by more, resulting in additional deferrals being reclassified as catch-ups, correct?

Put another way, even though there are unused catch-ups, there is no way to recapture the catch-ups by shifting NHCE deferrals?

Thanks Tom!!

[Tom Poje]

I have never heard that method proposed.

Based on what I see in the regs (or at least how I've seen them interpreted) that would not be possible.

your suggestion put another way would be:

plan fails test initially, create some catch up, now plan passes. so now I can shift.

shift some NHCE deferrals.

Plan now fails ADP again, so create even more catch up and plan still passes which is what the requirement for shifting requires.

The problem with that:

Suppose you had a 2nd HCE (not catch up eligible) who had a refund.

You never rerun an ADP test after calculating refunds because the top-down method of refund is based on $ deferred not percent, but you are, in a round about way suggesting that.

Interesting suggestion though.

Though what might work (never tried it) run the ADP test on comp less deferrals, failing the testing even more, creating a larger catch up, and then shift. I wouldn't think that would work, but without seeing actual numbers it might work.