Can ADP QNEC correction exceed 402(g)?

Strange situation. Consider a self-employed husband and wife, each deferring $19,500 (2020) Their compensation turns out to be $65,000. Oh, and they have an employee that should have been eligible 7/1/20, but was overlooked an that was the only NHCE. It's too late for a refund, so a QNEC is required. The HCE ADP percentage is 64.93%:NHCE is 0%. 62.93% QNEC would be over $28,000. This would be higher than the §402(g) limit.

My inclination would be to limit the QNEC to $19,500, but I can't be sure that this is right. A seemingly paradoxical situation is that had the NHCE deferred $19,500, it still would not pass ADP.

Is there a way out of this dilemma?

I don't think a QNEC to pass ADP/ACP is limited by 402(g), only 415. But I'm not sure what an equitable solution would be is you took this to VCP.

Thanks, Corey. I'm aware of the MDO. I'm focused on the failed ADP test. It may be that the VCP retro amendment is the only way to get a good resolution at a price of $3,000 or less. Either that or a $28,000 QNEC.

It seems incongruous that you could have a situation where the NHCEs could defer the maximum and yet still fail ADP (understanding that this fact situation is not that).

ADP is a bright line test, you pass the math or you don't. Whether or not that is reasonable or equitable isn't really considered by the math. I'm sure the IRS would say - if you have a low pay owner deferring the max, SH is an option.

You also might consider testing on comp from date of entry which might cut your QNEC roughly in half.

The plan is now SH. Only dealing with 2020 and the cost of correction. Maybe a suggestion to revise §401(k)(3) to the effect that if the ADP of the NHCEs are greater than say 6.5%, then the ADP test is satisfied. [Typically, the 402(g) limit divided by the 401(a)(17) limit is roughly 6.7%]

This may be the 401(k) equivalent of the Rule Against Perpetuities.

The ADP testing would not need to include the NHCE if they weren't offered the plan. 

See page 86 on Rev Proc 2021-30

https://www.irs.gov/pub/irs-drop/rp-21-30.pdf

"(g) The methods for correcting the failures described in this section .05(2) do not apply until after the correction of other qualification failures. Thus, for example, if, in addition to the failure of excluding an eligible employee, the plan also failed the ADP or ACP test, the correction methods described in section .05(2)(b) through (f) cannot be used until after correction of the ADP or ACP test failures. For purposes of this section .05(2), in order to determine whether the plan passed the ADP or ACP test, the plan may rely on a test performed with respect to those eligible employees who were provided with the opportunity to make elective deferrals or after-tax employee contributions and receive an allocation of employer matching contributions, in accordance with the terms of the plan, and may disregard the employees who were improperly excluded."

And if that were the only NHCE? So we have to correct the ADP test first. This would be, potentially, a QNEC of 64% of applicable compensation. Same if none of the eligible NHCEs deferred.

57 minutes ago, Dalai Pookah said:

And if that were the only NHCE? So we have to correct the ADP test first. This would be, potentially, a QNEC of 64% of applicable compensation. Same if none of the eligible NHCEs deferred.

I think you are mis-understanding. The ADP test would not include the NCHE at all. If the only two people who were given the opportunity to defer are the H/W owners, then the ADP test will pass. You will have 2 in the HCE group, and 0 in the NHCE group. No ADP test failure at all. 

Then you would move on to the MOD failure. 

So you are saying it is better to not tell folks about the 401(k) plan and just give them missed deferral opportunity than run the correct test in the first place?

For a plan year that has already ended, and the error is discovered and correct now, I see no issue in using a testing method specifically allowed by EPCRS. 

They should also correct the ongoing MOD if there is one, which yes, offering the plan now, and means that the testing for the current year will include the NHCE. At which point that's on them if they choose not to tell their employees about the plan. But typically we strongly suggest they add a safe harbor provision so that the ADP testing is moot, and often the Top Heavy minimum becomes irrelevant as well. 

I have refused to work on corrections for prior years where I know the sponsor is having the error continue. Part of the principles of EPCRS involve fully correcting and changing processes and procedures to reduce future issues. Why would I want to work with a plan sponsor that has no intention of doing it right moving forward? (this is rhetorical - I don't need anyone to answer).

While it may not feel right to give the participant nothing for the ADP test, consider the participant will get the QNEC for the missed opportunity to defer, and likely a top heavy minimum if employed at year end. If there was any sort of match or other employer contribution that would be in the mix for correction as well. 

Maybe that's not enough for some, but still can be a decent percentage of compensation, and almost certainly more than what a safe harbor contribution to that NHCE would have been. 

Oh I'm sure the participant will be happy with the "found" money. I'm just surprised the IRS would allow a correction that does not require re-running the test with the employee in it allowing the owners to not make refunds, not make one-to-one corrections for the failed test, and not make a QNEC required to pass testing. I mean great if that's the end result, just seems at odds with the IRS correction programs in general.

The HCE ADP of 64.93% and the NHCE ADP of 62.93% you provided indicates you are using the +2% part of the ADP test.  At these percentage levels, you should be using the 125% part of the ADP test so the NHCE would need to get to 52.944% to pass.

Could you share the employee's annual compensation for the year in question so we can explore the net impact of the proposed corrections?

Is the $65,000 the total of the compensation of the husband and the compensation of the wife, or is did the husband have compensation of $65,000 and the wife have compensation of $65,000?  Is the $65,000 before or after reductions for payroll taxes?

The employee's QNEC for the MDO will be 3%.  This like would be 3% of compensation earned after the 7/1/2020 entry date. 

You can also use that as the testing compensation which would leverage the QNEC as a % of pay.  If the employee's compensation is low relative to the owners, the net cost may be tolerable after considering the time and cost involved with a VCP or retroactive plan amendments.

Don't forget about a top-heavy contribution which will be separate and apart from the ADP debacle.  This will add 3% of the employee's annual compensation to the price tag.

Does a plan that excludes 100% of its eligible NHCEs have a "processes and procedures" problem which might cost the ability to even USE EPCRS here?