MWeddell

The typical 401(k) participant loan agreement -- https://www.shrm.org/resourcesandtools/tools-and-samples/hr-forms/pages/401k_loanapplication.aspx -- includes a consent to repay the loan through payroll deductions. It is a contract so one party can't unilaterally change its terms. The participant agreed to it. State law is pre-empted by ERISA. So in response to the question posed by the original post, the plan sponsor should do nothing but explain to the participant that loan repayments will continue to be deducted from the participant's paychecks. I believe this is correct, but am not 100% confident in my conclusion. I don't believe the IRS or DOL has addressed the issue.

Does the "one bad apple" rule happen in practice? Suppose two plans merge. The smaller of the plans, which held 10% of the combined plan's assets, came from a recent acquisition administered by staff other than the current employer's h.r. department. The larger of the plans, which held 90% of the combined plan's assets, is the employer's ongoing plan, administered by the h.r. department for many years. The IRS detects a compliance problem with the smaller of the two plans from a plan year before the acquisition closed and before the plans were merged. Has anyone ever witnessed a situation where the IRS imposed greater sanctions because the plans had merged than the IRS would have imposed if the plans had not merged? I've had a long career and I have never witnessed such a situation. I am employed by a large consulting firm and have asked many co-workers this question and have never heard of a situation when the IRS was harsher because the plans were merged. Yet I've heard scare talk of "tainted asset" and "one bad apply" numerous times. I've started telling clients that merging the plans poses a theoretical risk that one has tainted a larger pool of assets but that I've never witnessed it making a difference in practice.

There are anti-trust concerns if competitors discuss their fees. I suggest not doing so.

Most plan documents should be interpreted so that elective deferrals do not cease when the participant's compensation reaches $285,000, the 401(a)(17) compensation limit. Whether that is true of the original poster's plan document, I don't know. If deferrals should not have ceased, then all you have to do is restart deferrals at the 5% level. No QNECs are required because (1) Rev. Proc. 2019-19, Appendix A, Section .05(9)(a) allows for no correction of deferrals if the deferral error was corrected prospectively after the earlier of 3 months after failure began or, if the employee notified the plan sponsor, the first payroll after the end of the month after the month of notification. This assumes that you will notify the employee within 45 days after the deferral rate has been corrected. (2) The employee didn't lose out on any matching contributions, as you are aware. If you want to suggest that the employee raise his deferral rate, that really has nothing to do with correcting the error.

Yes, my position is that only once all of the allocation conditions are fulfilled is the allocable share a protected benefit. Agree, but watch out that there often are exceptions to the "last day of the plan year" condition for those who retire, die, become disabled, go on military leave, etc. during the plan year. It complicates things if the allocable share is a 411(d)(6) protected benefit for some participants.

It is true that qualified plans in the Internal Revenue Code are classified as pension plans, profit-sharing plans and stock bonus plans. What is nicknamed a 401(k) plan typically is a type of profit-sharing plan. However, just because a plan officially is a profit-sharing plan does not mean that the plan document currently permits employer nonelective, nonmatching contributions to be made. In short, I think ESOP Guy is conflating two different issues.

Regarding whether the two plan solution works, the IRS declined to express an opinion on it in 2013. I don't think it changes anyone's mind, but I thought I'd mention it. See IRS Q&A-16 from the 2013 ASPPA conference at https://www.asppa.org/sites/asppa.org/files/PDFs/GAC/2013 DC QA.pdf

Unlike in that other thread, the employer would be stuck maintaining two DC retirement plans for a prolonged time period. Once the plans merge or there is a transfer of assets between them, Treas. Reg. Section 1.411(a)-5(b)(3)(ii) requires that vesting service be recognized from the earliest date of when any component plans were established.

It is a secret buried in the regulatory scheme that the definition of "plan" is very difficult to apply in a defined contribution plan context. It is not clear at all. In practice, we assume usually that a "plan" under the 414(l) regulations is the same as a "plan" for DOL purposes, but the regulation doesn't come any where close to saying that. Quite some time ago, I have seen legal counsel draft for an employer a defined contribution plan document that stated that there was more than one asset pool and hence more than one "plan" within the 414(l) meaning within a single plan document. It strikes me as very aggressive, an example of form over substance if one paragraph in a plan document can make it so, but I couldn't say that it was definitely wrong.

Thank you, Luke and Bird, for the thought discussion. I did not recall the TAM. The TAM largely supports my interpretation of the 411(d)(6) regulations, that the right to share of the allocation (but not any specific dollar amount given that a contribution might not be made) becomes a protected benefit when a participant has satisfied all of the allocation conditions. The conclusion of the TAM makes clear that the allocable share of the contribution, if any, that was to made became protected on December 31, 1992 given that the plan had a "last day of the plan year" allocation condition. It did not become a protected benefit on January 7, 1993, when the contribution was made to the trust but not yet allocated. The facts would have become harder if the plan was amended to add a second allocation method on or after December 31, 1992 but before the contribution had been made, but it seems that the TAM's conclusion would remain unchanged. I don't (yet?) agree with Luke's observation that it is a close question. I thought it was clear based on the regulation and even more so with the TAM. However, I do agree with Luke that an employer may be able to circumvent the regulation. Besides Bill Presson's two plan workaround, it is possible that there is a work around within one plan. Given the plan document language, I don't think that adding an allocation date would necessarily work, but having a short plan year might work. The employer could amend the plan document on or before October 31 to state that the plan year ended October 31 and that another short plan year would exist from November 1 through December 31. The amendment would also state that the 415 limitation year and the vesting computation period remains on a calendar year basis throughout. The amendment could also include a different allocation formula for the November 1 - December 31 two-month plan year that uses compensation for the whole plan year. The employer could then decide to allocate $0 for the 10-month plan year ended October 31 and substantial dollars for the 2-month plan year ended December 31. The plan year would have an rather interesting 401(a)(4) general test to run for the 2-month plan year and of course there would be an extra Form 5500 / extra audit needed, but I think that might work.

I wouldn't perform any BRF testing on the available rate of matching contributions (assuming one matching formula applies to all eligible employees). It's not just a reasonable definition of compensation but (assuming that 414(s) test passes) it is one that satisfies 414(s). Just because there are exclusions doesn't mean that BRF testing is needed.

Based on Treasury Regulation Section 1.411(d)-4, Q&A-1(d)(8), I agree with Bill Presson. The conditions for receiving an allocation of contributions or forfeitures for a plan year after such conditions have been satisfied are a 411(d)(6) protected benefit. They cannot be amended retroactively to the beginning of the current plan year. Having a separate plan merged in is a viable work-around in my view, but it is aggressive enough solution that I would suggest the client check with its legal counsel.

Discussed here: I eventually came around to the view that what you suggest is permissible. You would have to perform a 414(s) compensation test on the definition of compensation used to compute the safe harbor matching contributions (and make sure the client is prepared for the possibility that it might fail one of these plan years).

I was wondering the same thing yesterday, but don't have an info to add for you, Austin.

Not the case that I was trying to recall, but Toomey v DeMoulas Super Markets, Inc. is in the process of being settled. Plaintiffs sued, claiming that fiduciaries breached ERISA's fiduciary duties by requiring all participants to invest in a single fund that earned investment returns that were too low because it was invested primarily in fixed income.

You all work regularly in a market with different size plans that I do. In the large and jumbo DC plan market, one seldom sees plans pay a percentage of assets to an investment advisor.

Neither the Roth contributions nor the pre-tax elective deferrals can be used as the safe harbor employer nonelective contributions.

Here it is.

Joe was a greater than 5% owner in the look-back year. That makes him an HCE. Electing the top-paid 20% group rule won't change the result.

Ha! Well, you might get someone else interested enough to look at the regulations too.

It doesn't work. https://www.law.cornell.edu/cfr/text/26/1.401(k)-2 -- look for the phrase "first day of the plan year" at the end of Treas. Reg. Section 1.401(k)-2(b)(3)(iii)(B). When it comes to technical questions, there just isn't any substitute for looking carefully at the relevant regulations.

ldr, I agree with your latest post. However, the facts do sound a wee bit dodgy. The plan has to not just officially allow traditional employee after-tax contributions but they have to be effectively available to NHCEs during the plan year being tested. Is that feature mentioned in an SPD? Mentioned in the enrollment materials (either hard copy or on the web)? One can't just discover an after-tax contribution feature now. I am less skeptical than BG5150 is about having both Roth and traditional after-tax contributions and about payroll handling it well. However, I typically work with large and jumbo employers, so my practical experience may not be as relevant to your situation.

I just remember reading about the complaint. I don't know how the case was disposed but the probability is that it was settled. Certainly what is considered to be prudent can change over time. I would not be 100% confident that there is no fiduciary liability associated with offering just one fund and no investment choice in a defined contribution retirement plan. I would be more cautious when advising employers. The main problem with judicial-made law is that it often isn't known until afterwards.

Plus recharacterization is available as an ADP test remedy only if the plan generally allows NHCEs to make traditional employee after-tax contributions.

Quite a while ago (2009?), I recall a lawsuit by an older participant who sued when the market fell because the participant claimed that he/she would have chosen a more conservative allocation but the plan had just one investment choice. I doubt that I could find it and just because there was a lawsuit doesn't tell us how much the case settled for.