C. B. Zeller

There is an exception in 1.401(a)(26)-1(b)(3) for underfunded frozen DB plans. If your plan meets the requirements of that section then it would be exempt from 401(a)(26) for the year. Otherwise, see 1.401(a)(26)-2(b).

This article on ASPPA's website (Redoing a Schedule SB on a Takeover When You Identify Issues) claims that "The IRS has indicated that it is not necessary to amend a Schedule SB when the changes have no impact on minimum funding" but does not provide a source.

Putting aside the fact that it is the plan administrator, not the trustee, is required to sign, what would you do if the former administrator resigned and moved to Tahiti? Chase them down and make them sign a form? Or what if they were hit by a bus? It's the administrator at the time the form is signed.

No. A key employee is a key employee regardless of whether they are eligible.

Regarding definition of comp for safe harbor nonelective purposes: 1.401(k)-3(b)(2) Safe harbor compensation defined. For purposes of this section, safe harbor compensation means compensation as defined in §1.401(k)-6 (which incorporates the definition of compensation in §1.414(s)-1); provided, however, that the rule in the last sentence of §1.414(s)-1(d)(2)(iii) (which generally permits a definition of compensation to exclude all compensation in excess of a specified dollar amount) does not apply in determining the safe harbor compensation of NHCEs. Thus, for example, the plan may limit the period used to determine safe harbor compensation to the eligible employee's period of participation.

You said "related" participating employers so I am going on the assumption that this is a controlled group or an affiliated service group. If they are still working for the member of the CG/ASG, they are under age 59-1/2, and there is no hardship, then no distributable event has occurred. All members of a CG/ASG are treated as a single employer for purposes of 416, so they are included in the top heavy test. They would be included in the coverage test as nonexcludable, not benefiting. This all changes if this were a MEP, i.e. if a CG or ASG does not exist. In that case top heavy and coverage are tested separately. However it would still not be a distributable event. It may be possible to spin off a new plan and terminate that plan, which would then be a distributable event.

I won't paste them here, but the relevant reg sections are 1.401(m)-2(a)(6)(ii) which says that you can include elective contributions in the ACP test so long as a) the plan is subject to the ADP test, and b) the plan would satisfy the ADP test including the shifted contributions; and 1.401(k)-2(a)(5)(iv) which says that elective contributions used to satisfy the ACP test are not included in the ADP test, except to the extent necessary to demonstrate satisfaction of the requirement in the (m) reg that the ADP test is satisfied when including those contributions. The way I'm reading this is that the ADP test including the shifted contributions must pass because the (m) reg says it must pass. The ADP test not including the shifted deferrals must pass because simply every 401(k) plan (exceptions notwithstanding) must pass the ADP test. Therefore all of the usual correction methods that apply to the ADP test, including recharacterization as catch-up, are available for the test performed not including the shifted deferrals. BG, to address your point, let's say you had a genuine scenario where the ADP test passed with some excess, but the ACP test fails. After shifting enough deferrals to the ACP test to make it pass, the ADP test results in some excess contributions which are eligible for recharacterization as catch-up. Would you say that you can not shift the deferrals in this case? Here is a quote from the DC-2 Study Guide, 7th ed., which suggests that recharacterization as catch-up is a valid outcome of an ADP test performed after shifting deferrals: It is possible that excess contributions under the ADP test might be recharacterized as catch-up contributions, if any of the HCEs who are catch-up eligible participants have not used up the full catch-up limit for the year involved. The plan may not apply the recharacterization rule first, and then determine how it wants to shift elective deferrals to produce different testing results. All testing must be completed, including the shifting of elective deferrals if desired, before determining whether there are any remaining excess contributions that would be distributable but are eligible for recharacterization as catch-up contributions. This paragraph cites the preamble to the final 414(v) regs as a reference, and says see 68 F.R. 40511.

The other thread on creative ways to handle a top heavy plan reminded me of something I thought up a while back but never went forward with because I felt it was too aggressive. However I can not find any reason why it would be actually disallowed. I'm hoping someone here can poke a hole in this scheme and teach me something. This is a 401(k) plan with a single 100% owner who is the only HCE and the only key and is over 50 years old. The only contributions for the plan year are deferrals, although discretionary matching and nonelective contributions are permitted in the document. The plan passes the ADP test for the current year. The plan is top-heavy for the current year. The HCE's deferral contributions are equal to $6,000. Can we do the following: Shift 100% of the NHCEs' deferrals to the ACP test. The ACP test passes because there are no HCEs included in the test. Recharacterize the HCE's $6,000 deferral as catch-up as it now exceeds the limit of the ADP test, when testing on only the un-shifted deferrals. Since all of the NHCE deferrals were shifted, the ADR for the NHCE in this test is 0% and therefore the ADP limit for the HCE is 0%. There is no top heavy minimum for the current year, since the key did not have any contributions other than catch-up contributions, and catch-up contributions for the current year are not taken into account for purposes of section 416. My thinking is that this falls apart on step 1, that you can not shift deferrals into the ACP test just for fun, that there has to be an actual failure of the ACP test first. Is that actually written somewhere, or just accepted practice?

One more possible trick, if the keys are over age 50, is to amend the plan to impose a limit on elective deferrals for key employees equal to $0. Then they can defer up to $6,000 and it would be reclassified as catch-up due to exceeding a plan-imposed limit, and catch-up contributions are not included when determining the minimum allocation rate for the keys. If you want to play it safe, instead of a $0 limit you can do a $1 limit. Since you have to be eligible to defer in order to be eligible for catch-up, and you could make the argument that someone with a $0 limit is not eligible to defer, this avoids that possible interpretation. However there will be a (small) top heavy minimum required for the non-keys. I also do not know off the top of my head if there are any complications with amending a plan mid-year to add a limit. If the keys have not deferred anything to date then my feeling is it wouldn't be a problem. But I might be forgetting something.

Don't forget about the deduction limit, if you're giving large contributions to people with lower comp it can sneak up on you.

1.416 T-10, as Kevin pointed out, raises an interesting scenario. Consider an employee who is eligible for a 401(k) plan and a separate profit sharing plan which are part of the same required aggregation group. The reg states that "In the case of non-key employees who do not participate in more than one plan, each plan must separately provide the applicable minimum contribution or benefit with respect to each such employee." If you take "participate" to mean "contribute" then would an employee who does not contribute have to receive the top heavy minimum in both plans? Anecdotal, but as always a reminder to check your plan document: I have a document which defines Required Aggregation Group as "each plan of the Employer in which a Key Employee is a participant in the Plan Year containing the "determination date" or any of the four preceding Plan Years (regardless of whether the plan has terminated), and each other plan of the Employer which enables any plan in which a Key Employee participates to meet the requirements of Code Sections 401(a)(4) or 410." In this case I think it's clear that the use of the term "participant" means that anyone eligible in the past 4 years would be included, regardless of how you interpret the word "participates" in the regs.

You didn't say why he is unable to benefit in the CB plan, but I'm assuming the younger HCE was hired in the second half of 2016, since he would have had to have 2016 income >$120k in order to be HCE for 2017, and with 1 year of service plus Jan 1/July 1 entry dates he wouldn't be eligible for the CB plan until 2018. If this is the case, then for 2017 you can disaggregate him as otherwise excludable. Assuming he is the only employee who is otherwise excludable, then you have separate coverage and nondiscrimination tests covering just him, and you can do pretty much whatever you want. If this is not the case, for example the CB plan just excludes non-owner HCEs as a class, then you could try to restructure the aggregated test into component plans. Have one component with the partners and the younger NHCEs, and test on accrual rates. Have another component with the younger HCE and the older NHCE and test on allocation rates. This should let you get the younger HCE an amount on the DC side equal to at least the minimum gateway rate for the NHCEs, plus a little bit based on their equivalent accruals in the CB plan.

It wasn't my intention to take the discussion off the rails. My point was just that another paragraph of 416 clearly uses the term "participant" to include "person who is eligible regardless of whether they choose to defer" so it seems like a stretch to claim that the same term could mean something else in the same section of code.

416(c)(2)(A) states that every "participant" who is a non-key employee must receive a minimum contribution. I think the consensus (although I don't have a cite handy) is that in a 401(k) plan, non-key employees who satisfied the plan's minimum age and service requirements as of the end of the plan year, e.g. the 410 definition of participant, are entitled to the top heavy minimum as long as they are still employed at the end of the year, regardless of whether they contributed any elective deferrals. It seems reasonable to assume that the same definition of "participant" applies in 416(g) as in 416(c).

Mike is correct of course, but I will just add that the required aggregation group consists of all plans in which a key employee participated during the determination year or the previous four plan years - see 1.416-1 T-6. So if you exclude the key employees from the 401(k) starting in 2019, the first year that it would not have to be included in the top heavy aggregation group would be 2024.

Compliance with top heavy rules is a qualification requirement. It should be corrected under EPCRS.

"Middle click" - i.e., push in the scroll wheel - is a shortcut for open in new tab. I agree this is the best way to browse multiple threads without losing your place in the search results.

If I'm understanding you correctly, there is an employee who needs to receive an allocation to satisfy the gateway, but isn't eligible to receive an allocation under the plan's formula. If you can't pass coverage and nondiscrimination under the plan document as written, then you can adopt a corrective amendment that satisfies the requirements of 1.401(a)(4)-11(g) which provides for the additional benefits required to pass the test.

What would be the right way to account for sole proprietorship? I can see an argument that says since a sole proprietor is considered to be indistinguishable from the business, they are always considered to be a 5% owner, even before business activity begins.

Yes, please, that would be great if you could share the PDF.

Thanks @figure 8. I'll see if I can look up the webcast. Do you recall by any chance if there was a rev ruling or anything I could reference even for informal guidance?

I recently received a cash balance plan that defines the accrued benefit as the actuarial equivalent of the hypothetical account balance (where the hypothetical account balance is the usual sum of principal and interest credits), reduced by the actuarial equivalent of the balance of the "hypothetical offset account." The plan states that the hypothetical offset account for each participant is credited with an allocation equal to the lowest allocation rate from [Sponsor Name] Profit Sharing Plan plus the actual rate of return from the participant's account in the profit sharing plan. The profit sharing plan uses a new comparability allocation formula with each participant in their own group. In past years, the owner received a contribution equal to the 415(c) limit, and the non-HCE received the minimum gateway, let's say it was 6%. The profit sharing and cash balance plans satisfied the numerical tests for coverage and nondiscrimination when tested together. The idea behind this design seems to be that although the participants are not getting a uniform allocation in the DC plan, they are getting a hypothetical uniform allocation in the form of the hypothetical offset account, and it is that account balance which is being used to offset the accrued benefit. In other words, even though the owner is really getting a 20% allocation in the DC plan, his balance for purposes of the offset is based on only a 6% contribution since that is the lowest allocation rate of any participant in the DC plan. My concern is whether this arrangement satisfies the minimum participation requirements. 1.401(a)(26)-5(a)(2)(iii)(A)(2) states that the formula meets the requirements if "The employees who benefit under the formula being tested also benefit under the other plan on a reasonable and uniform basis." The word "plan" here concerns me as I do not think that a hypothetical offset account constitutes a plan. Nor do I think there is any permissive disaggregation rule that would allow that portion of the employer nonelective source which is attributable to contributions not in excess of a certain allocation rate to be considered its own plan. This plan does not have a determination letter. Has anyone ever encountered this type of plan design before? Does this seem permissible?

Can you return the check and have them re-issue it to the correct party?

This, I think, is the exact opposite of the ruling in the Microsoft decision. That said, post-Microsoft, many plans include a clause that excludes "employees treated as independent contractors" (or similar language) as a class. OP, you should check your document to see if it includes this language. If it does then you should be perfectly fine to exclude the former owner. If the former owner wants to continue to participate, then you should have him (as a sole proprietorship) adopt the plan as an adopting employer. That way, whether he is an employee or not, the exclusive benefit rule is satisfied.

I don't know if there are any clear guidelines on this. I would say that if they are expected to return to work before the end of the 1 year suspension then it is probably ok. If they are not expected to return within that timeframe, then either a) you are expecting them to repay the entire balance by check before the end of the 1 year period, which might be seen as a violation of the level amortization requirement, or b) you are expecting the loan to be deemed (and possibly offset, depending on the determination of disability) which would make it not a bona fide loan. However you have to be careful if you do not allow the refinancing, assuming the participant is NHCE, you could run afoul of nondiscrimination with respect to availability of benefits, rights and features. This is a sticky situation and is a good example of why plan sponsors might want to consider not allowing refinancing in their loan programs. I am curious to hear how others have addressed similar situations.