MWeddell

Modifying a bit Tom Poje's idea, you may have to move any QNECs to HCEs that exceed 8.1% (that's 2.7% times 3) into the ADP test too. It'll hurt the ADP test a little bit but satisfy the 3x comparability gateway allowing you to run a 401(a)(4) cross-test on the portion of the employer nonmatching contribution that is not shifted into the ADP test. Even more so now, you'll have to check whether the plan document permits that.

It is a distribution from the trust, so it does not go through payroll at all. I know, I've seen it done through negative contributions processed by payroll, but there is no legal justification for doing the correction that way.

Just because your account in the 401(k) plan has a 17% rate of return during some recent time period does not mean it is likely to return that much in the future. As you may have heard, past performance is not a guarantee of future performance.

This is a matter of interpreting the plan document, so our ability to help from this distance is limited. You say that the employer stopped their match mid year, but did they amend their document. If so, how did the amendment address the true-up match issue? If the employer merely stopped funding the match but did not amend the document, this could be bad news for the employer: the remaining match may have to be funded. Yes, it matters that the match percentages are not uniform. Limits are applied based on the plan year, you told us, so it sounds like the employer has not followed the rules set forth in the written plan document. Now if there was an amendment and it clearly said that no match was to be allocated after a specified date, perhaps the plan document may be interpreted so that provision overrules the true-up match language.

You may include all or none of the matching contributions into the ACP test for the after-tax (non-Roth) employee contributions for a 401(k) / 401(m) safe harbor plan.

Yes, I do read the regulation to allow the plan document to provide that the contribution may not be suspended. Definitely read the regulation yourself though, because there is some facts and circumstances language in there that the client should consider.

It is permissible, although unusual, for a plan document to not allow an employee to change his or her qualified cash or deferred arrangement election other than once per plan year. See Treas. Reg. 1.401(k)-1(e)(2)(ii).

I think this is not just permitted but required in your circumstances -- QACA covers only those with a year of service but plan lets employees make deferrals before then. The automatic enrollment must take plan when participants enter the "plan," which in the testing regulations means the portion of the plan that satisfies the QACA safe harbor rules, not the otherwise excludable employee portion of the plan that is subject to ADP / ACP testing (if it has any HCEs). I usually tell clients that having a longer eligibility service condition for employer contributions than for elective deferrals doesn't work well for the QACA safe harbor design because of the delay in when the automatic enrollment must occur, but apparently you like this side effect in this instance.

You're right. Separate testing would then be needed to allow for cross-testing.

Joining you on this tangent ... This strikes me as a really bad plan design idea. (1) Given how infrequently participants tend to revisit contribution elections, this is likely to lead to a much less retirement savings because the contribution amount does not rise as the participant's compensation tends to increase over time. (2) If a plan has a matching formula based on % of pay, one would want to enable participants to make their contributions as a % of pay.

If you have access to the ERISA Outline Book, they cite about four informal pieces of IRS guidance, such as newsletters, etc., where the IRS has consistently said that basing deferrals on total compensation doesn't bother them as long as the 402(g) limit / 401(a)(17) limit results in a deferral rate no greater than what NHCEs must do. A trickier issue is whether doing this complies with your plan document.

Especially if Plan A and Plan B cover mostly the same employees, it sure seems like you will have a better chance of passing the 401(a)(4) general test for plan B if you aggregate the plans for testing purposes. However, if you want to keep the plans separate for testing purposes, then the answers to your questions are: 1) The Plan A contributions are ignored when performing a general 401(a)(4) test on just Plan B. Those employees are included in the denominators of the fractions when you test each rate group because they are part of the controlled group. 2) Average benefit test is performed for all plans (or at least all DC plans) and all employees in the controlled group. 3) Anticipating a question you didn't ask, you'll have to cross-test Plan B to have any chance of passing the 401(a)(4) test, so you'll want to check whether the 5% of pay contribution given to NHCEs uses a 415 definition of compensation in order to get through the comparability gateway.

As long as you follow what is in the written plan document, yes it should be fine. Treating one group of NHCEs different from another group of NHCEs is not a problem given that everyone becomes eligible within the time period specified by Code Section 410(a).

Rev. Ruling 2012-4 is another example where the IRS approved accepting rollovers from former employees. PLR 8816050 is another example. Still didn't find anything from the DOL though.

I had thought that there was guidance allowing a plan to accept rollovers from former employees. I didn't have much luck locating any authority for that just now though. Private letter ruling 200027058 permits incoming rollover contributions from former employees, but that's not very relevant: what we really want is something from the DOL indicating that it is not an exclusive benefit rule violation.

I believe that no missed earnings need to be restored to participant #1. The guidance states that the corrective QNEC must be adjusted for earnings, but when there is no corrective QNEC, there are no adjusted earnings. Yes, this is a different result than if we went back to the general principal of putting the plan and the participant in the position they would have been in had there been no operational faitlure, but the corrective actions such as that in the Appendix A provision you cited are deemed to be satisfactory.

Amend the plan too. Effective in 2015, the ability to change the match mid-year is more restricted.

The EPCRS does specficically permit you to not make corrective QNECs to the extent the participant's deferrals reached the 402(g) limit.

Seems to me that this would definitly violate the contingent benefit rule in the 401(k) plan (assuming that you are talking about having the whole 401(k) plan account balance, not just the employer nonmatching contribution aaccount, be the offset).

I'll let others answer. I work with larger employers and hence rarely perform top-heavy testing.

1. You will need to make sure the merger & acquisition transition rule does not end prematurely or perform and satisfy coverage testing for 2014. If you are relying on the m&a transition rule, then it is okay to amend Company B's 401(k) plan as long as the amendment is not a "significant change in the plan or in the coverage of the plan other than the acquisition or disposition." Treas. Reg. 1.410(b)-2(f). There is almost no elaboration in the regulations beyond the text I just quoted. I would think that amending the plan to reflect Company A as the sponsor is not a significant change, but that's just my opinion. I would think that changing the vesting schedule is a significant change and would prematurely end the m&a transition period. However, given that this is essentially a judgment issue, I would urge the client to consult with legal counsel. 2. There is no problem with that provided that the plan document after it is amended so provides.

Yes, if you aggregate for ADP / ACP testing, then you must also aggregate for 410(b) coverage testing and BRF testing. A tougher issue to resolve is whether you have to aggregate the employer nonmatching nonelective contributions portion of the plan or can test them separately. Seems to me that you may test them separately but be aware that the IRS has informally opined that you should consistently aggregate that portion of the plan too since you are aggregating the 401(k) and 401(m) arrangements.

If they should have been eligible to defer in 2013, then they are included in the ADP test. If they should not have been eligible to defer, then the plan was operated not in compliance with the plan document, so the situation should be corrected using the EPCRS.

To remain a 401(k) / 401(m) safe harbor plan, only 4% of pay may be provided as a discretionary matching contribution, but you could get the total match up to 4.5% of pay if part of the match was mandated by the plan document and not discretionary.

You have to administer the plan in accorance with the written plan document. If the document seems to be silent on the subject, then someone must interpret the plan document.