C. B. Zeller

The requirement for the 60-day notice of intent to terminate is found under ERISA sec. 4041(a)(2). This section only applies to plans subject to Title IV, in other words, PBGC-covered DB plans.

A corrective amendment under 1.401(a)(26)-7(c) is subject to the same requirements as a corrective amendment under 1.401(a)(4)-11(g)—in particular, the amendment must satisfy coverage and nondiscrimination testing on its own. If the individual(s) you are looking to bring in to the plan under the amendment is/are HCE, and the employer has any non-excludable non-HCEs, the amendment would not be allowed. The same is true if you are looking to satisfy the meaningful benefit portion of the test by increasing an HCE who is already benefiting at a lower level—you could not increase them on their own without also benefiting some non-HCEs. Note this is only true for a corrective amendment adopted after the end of the year. Assuming a calendar year plan, it is currently too late to fix under -7(c) for 2022, but if you are looking at 2023, then you could expand the group of participating employees in any way you like before the end of the year without the additional restrictions of -11(g). In addition, although there is currently some ambiguity around when this becomes effective, you will (eventually) be able to adopt an amendment retroactively to fix this under 401(b)(3) (as added by SECURE 2.0 sec. 316) without the additional restrictions of -11(g). One last thing—does your plan document include a 401(a)(26) fail-safe? If so, follow its terms before you start looking at corrective amendments.

Careful here - what happens if the employee (presumably these employees are partners or otherwise individuals significantly contributing to the production of the business) chooses not to participate or chooses a smaller contribution? Do they get that amount in cash (or in the case of a partner, earned income) instead? In other words, does making an election to receive a contribution result in a corresponding reduction in their compensation (and if it doesn't, why would anyone choose not to participate or choose a contribution level less than the maximum legal limit)? This sort of arrangement could result in a deemed CODA, and could disqualify the DB plan.

Even if the plan document allows a distribution, the distribution could be restricted by the plan's funded status under sec. 436, and if the participant taking the distribution is an HCE, 1.401(a)(4)-5(b). If the participant is at or near their 415 limit (or will be in the future), be aware of issues that can occur when there are multiple annuity starting dates. The distribution taken now still counts against the eventual 415 limit at retirement.

The rule is that it is based on the year in which the employee retires. The IRS has never given a concrete definition of "retires" for this purpose. If you asked the employee when they retired, would they say they retired in 2022 or in 2023? I have a feeling they would say they retired in 2022. Not that this is necessarily determinative, but it is probably indicative of the common understanding of what it means to retire in a given year. If I can hazard a guess, it sounds like RMDs should have started on 4/1/2023 but weren't, and you're trying to find a way to avoid the failure and associated penalties. I'm certainly sympathetic, but I would caution you (and your client) against taking a position that stretches reasonable interpretation. I'd also remind you (and your client) that the penalty for missed RMDs was reduced significantly by SECURE 2.0 and it may help everyone rest easier at night to simply admit to the failure and pay the penalty. Or better yet, file a Form 5329 and request a waiver of the penalty entirely.

So date of birth is 1947, which means he uses the pre-SECURE Act age of 70½. The required beginning date is April 1 of the year following the year in which the participant attains age 70½ or retires, whichever is later. The participant attained age 70½ in 2017 or 2018, and if I'm understanding the facts correctly, retired in 2022. So the required beginning date was 4/1/2023.

What is their date of birth?

For Cycle 3, the IRS required that the plan document explicitly specify the determination period for calculating matching contributions, including safe harbor match. Take a look at item C.18 in the adoption agreement. If the adoption agreement says the determination period is annual, and the employer calculates and deposits the match each pay period, then a true-up will be required. If the adoption agreement says the determination period is per pay period, then a true up would not be allowed unless the plan were amended, and then the rules for mid-year changes to safe harbor plans would come into play. If memory serves me right, FT had a FAQ sheet about this back when Cycle 3 came out. It is probably still on their website somewhere. Or I'm sure they would be happy to send it to you if you contact them, as Bill suggested.

There are two failures here: the missed deferral opportunity, and the failed ADP test. The QNEC used to correct the MDO is limited to the 402(g) limit. The cite on that is rev. proc. 2021-30 appendix B.02(1)(a)(ii)(B)(1) A QNEC included in the ADP test under the 401(k) regs does not have a similar limitation. However, this is a very unique situation, and as you have explained the numbers give a result that is wildly disproportionate to what most of us would consider to be a reasonable outcome. If this were my client, I might try to apply under VCP to amend the plan for 2020 to a 4% safe harbor non-elective contribution, and then use the 3% QNEC on top of that for the MDO.

For DC plans, 415 limits how much can go into a plan in a given year. There is no limit on how much can come out; the participant gets whatever was contributed plus any earnings. For DB plans, 415 limits how much can come out of the plan at retirement. There are no limits on what can go in*, but generally it wouldn't make sense to put in more than would be allowed to be paid out. The actuary will help you determine a contribution formula that will get you to the desired retirement benefit. *There is a limit on how much can be deducted, but that is not really the question here.

See the rule published at 88 FR 12048 (page 65 of the pdf here: https://www.govinfo.gov/content/pkg/FR-2023-02-24/pdf/2023-02653.pdf) DC plans use the number of participants with a balance as of the first day of the year to determine if they are a small plan or a large plan, unless they check the box for first return/report, in which case they use the number of participants with a balance as of the last day of the year.

A plan is exempt from the automatic enrollment requirements of IRC 414A if it is a church plan as defined in IRC 414(e). IRC 414(e)(1) defines a church plan as "a plan established and maintained (to the extent required in paragraph (2)(B)) for its employees (or their beneficiaries) by a church or by a convention or association of churches which is exempt from tax under section 501." There are some exceptions in 414(e)(2) and other requirements as well so I would recommend going and reading the whole subsection. If your plan meets the definition in 414(e) then it would be exempt from the automatic enrollment requirement. "Who's the Employer" by Derrin Watson has a chapter on church and governmental employers - you may find that to be a helpful resource in analyzing this question, if you have access to it.

https://www.dol.gov/agencies/ebsa/about-ebsa/our-activities/resource-center/faqs/efast2-credentials#q17

When evaluating whether there are related employers, keep in mind the reduced ownership threshold for controlled groups under 415(h) and the special rule for 403(b) plans under 415(k)(4).

With the changes to the top heavy minimum in SECURE 2.0, plus the changes to the way that participants are counted to determine if a plan is exempt from the audit requirement, most of the reasons for keeping employees out of a plan are gone. It would be much simpler, administratively, to allow all employees in immediately, or after some short period of service, less than 500 hours in a plan/calendar year. I think that approach will probably be best for most employers. For an employer who doesn't fall into that category though, and who does have a reason to keep employees out of the plan for a longer period of time, they are going to be strongly disadvantaged if they switch to the plan year after the first eligibility computation period. For example, say an employer does switch to the plan (calendar) year. An employee who was hired in December 2023, and who works 500 hours in a year, will most likely enter the plan January 2025 - only 13 months after their date of hire, and the same date they would have entered if the plan had only a 1 year/500 hours requirement. Thus the LTPT rule is essentially just requiring this employer to define a year of service as 500 hours instead of 1000 hours for eligibility; it removes the "LT" from the "LTPT." I think it's a good idea to switch to the plan year most of the time, for the reasons already discussed. However, when you have 2-year eligibility (and this includes the 100% vesting rule for PS and DB plans, not just for LTPT), switching to plan year utterly undermines it. What might be an even better idea, and I have not looked into the regs to see if this would be permissible, would be for a plan to switch to the plan year only after the second eligibility computation period. That would still preserve the two-year requirement in a meaningful way, but also reduce the recordkeeping burden after the first two years.

A few points: 1. The rule that automatically creates a controlled group between spouses' otherwise-independent companies in a community property state is going away starting in 2024, thanks to section 315 of the SECURE 2.0 Act. 2. There is nothing that says companies in a controlled group can't have separate plans, the plans just have to be tested together. This is only an issue if either of your companies have any employees. 2a. It's possible that the plan documents you are using may automatically adopt the plan on behalf of all controlled group members, but that is an issue with the document, not with any law or regulation. If that's not what you want to happen, find a new document provider. 3. You can't terminate a 401(k) plan while maintaining another defined contribution plan (such as a 401(k) plan) within the same controlled group. This is known as the successor plan rule and is designed to prevent people from skirting the age 59½ distribution restriction on 401(k) plans. You will have to merge the plans instead, which is a little different from a standard trustee-to-trustee rollover that you might be thinking of. My suggestion at this point: pick one of your two existing plans to be the surviving plan, and merge the other plan into it. Execute a participating employer agreement (or joinder agreement, there are other names for it as well) to adopt the plan on behalf of all three employers (your company, your wife's company, and the joint venture). Optionally re-name the plan, but that is largely an aesthetic choice. One more thing that just came to mind: Have you been filing Form 5500-EZ? If not, is it because the assets in each plan are below $250,000? If the assets were above $250,000 combined you were likely required to file.

Pending any future guidance to the contrary, I do not believe you can just count calendar years (or plan years) for determining LTPT eligibility. IRC 401(k)(15)(D)(ii) and ERISA 202(c)(4) (as added by SECURE 2.0 sec. 125) both indicate that the 12-month period used to determine LTPT eligibility is determined "in the same manner" as for standard eligibility, meaning the 12-month period commencing on the employee's date of hire, and presumably with the option to switch to the plan year only after the first 12-month period. What I would like to see document providers offer - and I don't know if anyone is planning on doing this yet - is the option to keep the anniversary date measurement period for purposes of determining LTPT eligibility, but switch to plan year for purposes of standard eligibility.

The cite is 1.401(a)(9)-2 (which has not yet been updated for the changes in RMD ages made by SECURE and SECURE 2.0, so mentally insert other ages as appropriate)

Agree with you 100%. The lifetime income illustration is a Title I requirement. Unless the CPA believes the plan is subject to Title I - in which case, bonding, disclosures, etc. all apply - then there is no requirement to provide the lifetime income illustration. That said, there is nothing saying you can't provide one, and if the CPA really wants to see it, I'm sure you'd be happy to prepare one for him, for a modest fee....

What does the plan document say? I have seen pre-approved documents with a checkbox option to limit the beneficiary to the participant's spouse. I don't know that I've ever seen that option used, but strictly speaking a DB plan isn't required to offer any forms of benefit other than what's required under the QJSA rules, and QJSA means spouse.

IRC 414A as added by SECURE 2.0 sec. 101 applies to any cash or deferred arrangement established on or after 12/29/2022. While a profit sharing plan could have a retroactive effective date, a CODA which is part of a profit sharing plan can not. In other words, the effective date of the 401(k) feature can't be earlier than the date on which the plan document was signed, and the effective date of the 401(k) feature is what controls whether mandatory auto-enrollment applies (with the caveat that this is my best reading of the law as written, since IRS has published no guidance on this yet). Does that answer your question?

I agree with DavidO. You only count compensation for employees who actually receive an allocation of the profit sharing contribution. The IRS's position is that an employee who only benefits under the 401(k) portion of the plan doesn't count for 404, because of 404(n). See PLR 201229012.

No need to apologize - we all had to start somewhere. Although, given the seriousness of this situation - the employer apparently held on to all of the employee's contributions for 4+ years, if I am understanding you right - you might want to work with an ERISA attorney, or at least a plan professional who has experience in complex plan corrections on this. Casual advice offered on a message board might not be the best fit. Others may disagree, but I am thinking that the fact the employer just kept all of the employee's money for this long of a time may cause this to rise to the level of an egregious failure, which is not self-correctable. Late deposit of employee contributions is a prohibited transaction, which is subject to an excise tax under IRC sec. 4975, and is also a breach of the employer's fiduciary duty under ERISA sec. 406. On top of this, it is also a plan qualification failure. Some practitioners will correct the qualification failure by using the IRS self-correction program, and paying the excise tax to the IRS, then considering the fiduciary breach to be solved. Others prefer to formally correct the fiduciary breach by filing with the DOL, which is also deemed to satisfy the IRS's correction requirements. Both methods require the participant to be credited with lost earnings; the DOL method allows the use of the calculator on their website whereas the IRS requires earnings to be credited at the plan's actual rate of return (in other words, you need to calculate what the contributions would have gained if they had actually been deposited on time). Late deposit of the matching contributions is not a prohibited transaction, so there is no excise tax, but it also falls solely under the IRS's corrective regime, which means you have to use the actual rate of return.

I have heard (although I do not have first hand experience with this) that you can request a copy of Form 5500-EZ using Form 4506. You might need a Form 2848 to request it on your client's behalf.

The VFCP calculator is used to calculate the correction under the DOL's Voluntary Fiduciary Correction Program. Is the sponsor actually filing under VFCP? If not, the VFCP calculator should not be used. Late deposit of employee contributions is a prohibited transaction, but late deposit of employer matching contributions is not (although it may be a self-correctable operational failure). So you would only include the deferrals in the VFCP calculator, not the match. The recovery date is the date that the principal was deposited. The final payment date is the date that the lost earnings will be deposited. The final payment date takes into account the earnings on the earnings, so it should be a date in the future when the amount determined by the calculator will be deposited.