EBECatty

I don't see the quoted language in the preamble to the proposed RMD regulations, but here's an excerpt from a comment letter from the SPARK Institute summarizing the issue as I understand it: Application of 10-Year Rule to Non-Governmental 457(b) Plans. There has been some confusion as to whether or not the SECURE Act’s 10-year rule applies to section 457(b) plans of non-governmental entities. The reason for this confusion is that Code section 401(a)(9)(H)(vi) states: “For purposes of applying the provisions of this subparagraph in determining amounts required to be distributed pursuant to this paragraph, all eligible retirement plans (as defined in section 402(c)(8)(B), other than a defined benefit plan described in clause (iv) or (v) thereof or a qualified trust which is a part of a defined benefit plan) shall be treated as a defined contribution plan.” The cross reference to Code section 402(c)(8)(B) suggests that, perhaps, section 457(b) plans of non-governmental entities would not be treated as a defined contribution plan, and thus may not be subject to the 10-year rule. We believe most have read the quoted language above as not exclusive, that is, it confirms that IRAs are treated as defined contribution plans, but does not serve to exclude non-governmental 457(b) plans. Such plans are subject to the 401(a)(9) rules, and most 457(b) plans are defined contribution plans, as that term is usually understood. The Proposal does not address this question directly, except to confirm that all 457(b) plans are subject to the RMD rules under Code section 457(d), and laying out a set of rules that apply to “defined contribution plans” in the -5 portion of the regulations. In Footnote 1 of the preamble, the Service references Code section 402(c)(8)(B)(iv) and (v), but does not provide further guidance on non-governmental 457(b) plans. In any event, we recommend that the Service confirm in the final regulation that all section 457(b) plans, except those that are defined benefit plans, are treated as defined contribution plans for purposes of the SECURE Act changes described in Code section 401(a)(9)(H), which includes the 10-year rule.

Thanks Luke. Yes, I agree, any allocation would have to be in accordance with the employer's fiduciary duties, so proportional is probably the most appropriate if no other reasonable rule is in the plan document.

Thankfully I've never encountered this, but state law can still set up a conflict even without automatic revocation in the plan document. Although other states' laws may or may not be similar, the one I'm most familiar with: (1) automatically revokes an ex-spouse's retirement plan beneficiary designation upon divorce; (2) deems the ex-spouse to have pre-deceased the participant for purposes of determining the beneficiary; and (3) if federal law preempts those two things from happening, imposes a personal liability on the ex-spouse who receives the payment in favor of the person who would have been the beneficiary under state law. So even if the plan terms do not revoke, and the plan still pays, the ex-spouse following a divorce, someone else may have a claim against the ex-spouse. I guess that's not the plan sponsor's problem, but I would probably suggest the plan sponsor interplead any meaningful account balance to avoid the argument altogether, particularly if they know the ex-spouse will not willingly turn over any proceeds to the person entitled under state law.

Appreciate your thoughts, Brian. I have never seen this become an issue either, but can see it potentially arising in some circumstances and was curious if others had encountered. For example, the employer has built up a funding reserve from lower-than-expected claims and wants to use some of the reserve for other corporate purposes. In that case, following the logic of dividing employee vs. employer contributions, I would think you would need to find a way to differentiate how much the employer could use without causing a PT or violating the exclusive benefit rule. Either way, thanks for your input.

Thanks Brian. I am familiar with that guidance, and I think my real question is at the following step: Once the amount of plan assets is determined, is there any guidance, best practice, or widely used administrative procedure to determine when those plan assets are considered used. For example, in my hypothetical there would be $200 of plan assets and $800 of non-plan (employer) assets. If a claim was made the next day for $300, would it be reasonable for the employer to pay the claim with all $200 of plan assets and $100 of non-plan assets, such that after that claim there is only $700 of non-plan assets? Or would it more customary to continue to treat the entire account as 1/5 plan assets and 4/5 non-plan assets?

My understanding is that employee premiums paid toward self-insured group health coverage are considered ERISA plan assets, but may be held in the employer's general assets due to the trust-requirement relief. However, the other ERISA rules (fiduciary obligations, prohibited transactions, reversions, etc.) continue to apply to the employee premiums. Say the total premium for a month is $1,000, of which the employee pays $200. The employer sets aside $1,000 in an account in its name and EIN and there is no other indicia of ownership or rights by the plan. How do you determine the portion of the account that is plan assets subject to, say, the PT rules? Can you treat the account as using employee contributions first? Would this accounting restart every payroll cycle when new employee premiums are withheld? Would appreciate any insights.

I think this will be the biggest issue moving forward. It's nice to have several more years to update the document, but HR departments, TPAs, recordkeepers, etc. will be reading/interpreting plan documents and SPDs for potentially five years with outdated provisions. There will likely be turnover in at least one of those roles before 2025. Then someone will have to reconstruct - in November 2025, probably - all the operational features/elections the plan used for the past five years. Then again, it's difficult to update the plan document with any certainty until we have final rules on some of these issues.

You'll probably need to check the definitions/thresholds in the specific state statute creating the program, which can be tied to things like the number of employees reported to the state's employment agencies, etc.

My understanding - someone please correct me if I'm wrong - is that the plan can only file a 5500-EZ as a one-participant plan if there are no other participants during the entire plan year. Here, there clearly were over 100 participants at the start of the plan year, or else they would not have needed an audit. Maybe if the final filing listed one participant all year the DOL would not get involved, but that would seem contradictory to the audit requirement for the same plan year. Perhaps there are further facts that would be helpful.

I wouldn't be terribly surprised if this was intentional to curb the use of the automatic 45-day extension. Maybe we'll see more of them around November/December this year. The 45-day correction period is statutory, so the DOL can't simply eliminate it, but it certainly wasn't intended for the purpose at issue here.

I'm assuming you mean a 6/30/21 PYE timely filed on 4/15/22? In my experience, the DOL usually sends the 45-day correction notice for a timely filed 5500 with no audit attached. While I understand either IRS or DOL can penalize for an incomplete return, the IRS going directly to penalties does seem like a deviation from prior practice and might rule out one of the "standard" options for a late audit that it sounds like was used here. In that case, DFVC for the whole filing may be safer, particularly if DFVC can avoid IRS penalties as well. We usually end of up with a small handful of these situations each year, and will this year, so would be curious to hear others' input as we get close to the filing deadline.

As Bill notes, I have most often seen the participant's name, an acknowledgement that they entered the plan too early under the regular eligibility rules, and a specific date of entry (but not DOB, DOH, etc.).

A number of our clients reduced the IRA force-out threshold to $0.01 in their Cycle 3 restatement to address this issue. I think it's a good idea.

I agree with Bird on measuring the 12-month period, but wouldn't a 12/31/23 statement sent on 3/31/24 be late under the 45-day period in DOL FAB 2006-3? In either case, if you sent the 12/31/22 statement on 2/10/23, then sent the 12/31/23 statement on 2/11/24, I think you're still fine.

I personally would recommend erring on the side of caution, partially for the reason you point out, Peter. Even if you don't have a prohibited sale of property under 4975(c)(1)(A) or transfer under (c)(1)(D), which you might at least indirectly, I think arguably you run into a problem under (c)(1)(E) prohibiting a fiduciary from dealing with plan assets directly or indirectly in his own interest. The DOL's interpretation is fairly broad: "Whether the proposed transaction would violate sections 4975(c)(1)(D) and (E) of the Code raises questions of a factual nature upon which the Department will not issue an opinion. A violation of section 4975(c)(1)(D) and (E) would occur if the transaction was part of an agreement, arrangement or understanding in which the fiduciary caused plan assets to be used in a manner designed to benefit such fiduciary (or any person which such fiduciary had an interest which would affect the exercise of his best judgment as a fiduciary).... [T]he Department further notes that if an IRA fiduciary causes the IRA to enter into a transaction where, by the terms or nature of that transaction, a conflict of interest between the IRA and the fiduciary (or persons in which the fiduciary has an interest) exists or will arise in the future, that transaction would violate either 4975(c)(1)(D) or (E) of the Code." I guess you could argue the transaction is not intended to "benefit" the soon-to-be spouse as fair value would be paid. And that there is no conflict of interest, but it seems to me that argument would be more difficult when the IRA owner and fiance(e) (a person in whom the IRA owner clearly has "an interest") are on opposite sides of a real estate transaction. The excerpt is from DOL Adv. Op. 2000-10A, which, interestingly, involved an IRA owner attempting to meet the minimum investment threshold to invest with one Bernard L. Madoff Investment Securities. ("You further represent that Mr. Adler believes that Madoff would effectively manage assets for the IRA....")

There have been some prior threads on this, including one I believe I started, and my recollection is there's not one perfectly clear answer. Some take the position that belonging to an overlapping group makes an employer part of one larger group, which in your example would continue to grow. Because A, B, and C are related, and C and E are related, then A, B, C, and E are all one employer. If that's true, because D and E are related, and A, B, C, and E are related, then A, B, C, D, and E are all one employer. I suppose F would be separate in any event if unrelated to every other entity. This is the approach taken in Who's the Employer. Others take the position, maybe more as a practical alternative, that you test each group separately, which would give you five tests. I have heard from several TPAs who test using this approach.

Thanks Luke. I agree that, at least in my experience, these types of payments are often treated as wages from the seller. (Although I have had a buyer who insisted that seller-promised post-closing retention payments, along with employer-side FICA, be transferred from seller to buyer in a purchase price adjustment, then paid to the affected employees by buyer and reported on a W-2 by the buyer, for exactly this reason.) The specific 415 language is interesting, though, in my opinion, and I think is a little broader than the similar provisions under 3401(a). In any event, if the seller is taking the position that the payment is wages from the seller for reporting and withholding purposes, it makes sense to treat the plan compensation accordingly. Assuming it's not a temporary leasing situation, and the seller has actually terminated the employees, would you treat a post-closing retention payment as 415 compensation under the post-severance timing rules? That seems to me a separate issue, i.e., it could still be "compensation," just not paid in time to be recognized by the plan.

I would be interested to hear everyone's thoughts on whether a post-closing retention payment made by the seller, but based on continued post-closing service with the buyer, would be plan compensation (assuming the seller's plan has not been terminated by the retention payment date). Under 1.415(c)-2(b)(1), the default 415 definition requires the compensation to be paid "for personal services actually rendered in the course of employment with the employer maintaining the plan." Would a payment contingent on personal services actually rendered to a different, unrelated employer meet this standard? The alternative definitions based on W-2 and 3401(a) don't include this exact phrase, but I think it would at least be implied. Also, under the timing rules, I'm not sure this would fit into any of the permissible post-severance categories (and I think there would be, for plan purposes, a severance of employment at the closing). Would appreciate any thoughts.

MWeddell, I'd be curious to hear their position for (1) writing this into the buyer's plan document, and (2) not treating it as some form of auto-enrollment. I can't imagine under what circumstances it would be considered an affirmative election in the buyer's plan. At the risk of piling on, I've also looked into this in the M&A context and could not find any support for simply carrying over the employee's deferral election from the seller's to the buyer's plan.

IRS Form 4506. Expect to wait several months.

I believe the timing is correct. Notice 2022-33 did not extend the deadline for a non-governmental 457(b) plan, so it should still be governed by the SECURE Act's statutory deadline. I think there is a (perceived) lack of clarity on whether the SECURE Act's 10-year RMD rule applies to non-governmental 457(b) plans, but I am fairly certain the age 72 RMD rules apply. If the non-governmental 457(b) plan document includes those rules in the plan document, I would think it would need to be amended for SECURE. The remaining SECURE provisions, and all of the CARES Act provisions, do not apply to non-governmental 457(b) plans.

Thank you both for confirming my thinking. CuseFan, neither company is a service organization, so we're clear on that side as well.

Would appreciate a quick sanity check here: Mother owns 70% of Corp A. Mother has an adult daughter, who is married to son-in-law. Son-in-law owns the other 30% of Corp A. Son-in-law also owns 100% of Corp B. (Assume no other attribution aside from family, no excluded interests, no ASG, no management group, not a community property state.) Under 1563, I believe that mother will not be deemed to own any stock in Corp B. Daughter would be deemed to own all 100% of son-in-law's stock in Corp B, but because daughter is an adult, and mother does not own >50% of Corp B, mother is not deemed to own daughter's (attributed) stock in Corp B. Also, the prohibition on double-attribution among family members would cut off deemed ownership of those shares with daughter (i.e., once they are attributed from son-in-law to daughter by spousal attribution, they cannot be attributed again to mother, even if mother did own >50% of Corp B). Likewise, as an adult, daughter would not be deemed to own mother's shares in Corp A because daughter does not own >50% of Corp A's shares. Because mother owns >50% of Corp A, she would be deemed to own any shares in Corp A owned by daughter directly (zero); however, the double-attribution rule would prevent attribution from son-in-law to wife to mother, so mother would not be deemed to own son-in-law's 30% in Corp A. I think I'm left with: Corp A: 70% mother; 30% son-in-law; same 30% daughter through spousal attribution Corp B: 100% son-in-law; same 100% daughter through spousal attribution; 0% mother This leaves only 30% common ownership, which clearly avoids a controlled group unless I am missing something.

Fair point, Luke. I had assumed the question was asking whether a plain salary deferral could be made subject to a SROF for purposes of 409A. I suppose deferred salary could, in the broader sense, actually be forfeited (in that the participant loses the right to receive payment altogether) but the SROF would not be recognized under 409A (or 457(f) or, I would think, 3121 by way of 83).

The following is from 1.409A-1(d)(1). Note that salary deferrals can be made subject to vesting in certain circumstances where the ultimate payment will be materially greater than the salary deferred, e.g., an employee can defer base salary and receive an employer match on the deferral (25% minimum is a good rule of thumb), all of which can be made subject to a SROF. But just a straight salary deferral generally cannot be made subject to a SROF. Except as provided with respect to certain transaction-based compensation under §1.409A-3(i)(5)(iv), the addition of any risk of forfeiture after the legally binding right to the compensation arises, or any extension of a period during which compensation is subject to a risk of forfeiture, is disregarded for purposes of determining whether such compensation is subject to a substantial risk of forfeiture. An amount will not be considered subject to a substantial risk of forfeiture beyond the date or time at which the recipient otherwise could have elected to receive the amount of compensation, unless the present value of the amount subject to a substantial risk of forfeiture (disregarding, in determining the present value, the risk of forfeiture) is materially greater than the present value of the amount the recipient otherwise could have elected to receive absent such risk of forfeiture. For this purpose, compensation that the service provider would receive for continuing to perform services regardless of whether the service provider elected to receive the amount that is subject to a substantial risk of forfeiture is not taken into account in determining whether the present value of the right to the amount subject to a substantial risk of forfeiture is materially greater than the amount the recipient otherwise could have elected to receive absent such risk of forfeiture. For example, a salary deferral generally may not be made subject to a substantial risk of forfeiture.