Peter Gulia

For minimum distribution, what date’s ownership counts to determine a 5%-owner? Here’s my not-entirely hypothetical: A partnership is the plan sponsor and only participating employer of an individual-account § 401(a)-(k) retirement plan. The plan year is the calendar year. The partnership’s tax year is the calendar year. Every partner is on the calendar year for one’s tax year. The partnership has no mandatory retirement age, nor even a presumed ordinary retirement age. A working partner will reach age 73 during 2027 (and expects to continue working into her 80s). “For purposes of section 401(a)(9), a 5-percent owner is an employee [including a deemed employee] who is a 5-percent owner (as defined in section 416) with respect to the plan year ending in the calendar year in which the employee attains the applicable age.” 26 C.F.R. § 1.401(a)(9)-2(b)(3)(ii) (emphasis added) https://www.ecfr.gov/current/title-26/part-1/section-1.401(a)(9)-2#p-1.401(a)(9)-2(b)(3)(ii). Does this mean the measurement date is December 31, 2026? Under the partnership agreement, a partner’s capital interest can change any day. For example, a partner might get distributions from capital, or even might withdraw capital. A partner’s profits interest, if measured as a percentage of the partnership’s profit, can change because the partner’s interest is measured by several factors, including (for a relevant year or other period) the partner’s revenue generation to her practice, expenses specifically allocated to her practice, origination credits for having introduced a client to another partner’s practice, and a proportionate share of the partnership’s general overhead allocated to all practices. The plan’s administrator wants to get the measurement date right so it neither fails to meet § 401(a)(9) nor unnecessarily (and improperly) directs an involuntary distribution the plan does not provide. BenefitsLink neighbors, how’s my guess?

That life insurance is paid for from participant contributions rather than from a matching or nonelective contribution does not by itself nonapply the tax law about a “PS 58 cost” attributed for the life insurance death benefit. Internal Revenue Code of 1986 (26 U.S.C.) § 72(m)(3): Life insurance contracts (A) This paragraph shall apply to any life insurance contract— (i) purchased as a part of a plan described in section 403(a), or (ii) purchased by a trust described in section 401(a) which is exempt from tax under section 501(a) if the proceeds of such contract are payable directly or indirectly to a participant in such trust or to a beneficiary of such participant. (B) Any contribution to a plan described in subparagraph (A)(i) or a trust described in subparagraph (A)(ii) which is allowed as a deduction under section 404, and any income of a trust described in subparagraph (A)(ii), which is determined in accordance with regulations prescribed by the Secretary to have been applied to purchase the life insurance protection under a contract described in subparagraph (A), is includible in the gross income of the participant for the taxable year when so applied. (C) In the case of the death of an individual insured under a contract described in subparagraph (A), an amount equal to the cash surrender value of the contract immediately before the death of the insured shall be treated as a payment under such plan or a distribution by such trust, and the excess of the amount payable by reason of the death of the insured over such cash surrender value shall not be includible in gross income under this section and shall be treated as provided in section 101. http://uscode.house.gov/view.xhtml?req=(title:26%20section:72%20edition:prelim)%20OR%20(granuleid:USC-prelim-title26-section72)&f=treesort&edition=prelim&num=0&jumpTo=true 26 C.F.R. § 1.72-16(b) https://www.ecfr.gov/current/title-26/part-1/section-1.72-16#p-1.72-16(b). 26 C.F.R. § 1.402(a)-1(a)(3) https://www.ecfr.gov/current/title-26/part-1/section-1.402(a)-1#p-1.402(a)-1(a)(3). That someone mentioned premium payments sent directly to the insurance company instead of to the recordkeeper suggests possibilities for violations beyond the one you ask about. Consider reevaluating whether, or on what fee, and with what scope of engagement, you want the prospective client. This is not advice to anyone.

“If the Beneficiary does not predecease the Participant, but dies prior to the distribution of the death benefit, the death benefit will be paid to the Beneficiary’s ‘designated beneficiary’ (or if there is no ‘designated Beneficiary,’ to the Beneficiary’s estate).” That’s from a document a big recordkeeper provides its customer. (The copyright notice does not name the text’s author.) While that might be a possible beneficiary-ordering regime (and might appear in many service providers’ forms for plan documents), it’s not the only beneficiary-ordering regime, even before looking for a default beneficiary. A plan’s document might not state a beneficiary’s-beneficiary provision, and might provide something else—for example, exhausting all participant-named beneficiary designations, primary and contingent, before turning to anything else. Unless an adviser already knows what its particular advisee’s plan document provides, consider many BenefitsLink neighbors’ reminder: RTFD—Read The Fabulous Document.

Before other steps, check all records available to, and facts known to, the plan’s administrator to look for whether there is a surviving spouse. Look for the participant’s designation of a contingent beneficiary. Then, RTFD—Read The Fabulous Documents. Mainstream plan documents typically set an ordering regime to determine a beneficiary or beneficiaries when no participant-named beneficiary is alive. Don’t imagine or guess what’s provided. These might vary with choices made by an investment provider, service provider, document provider, or even a document user. It might even vary within one provider by different service platforms’ documents or different vintages of documents. After finding, if not the identity of the default beneficiary, at least the relationship that makes one a default beneficiary, RTFD (and any provisions assumed under a remedial-amendment regime) to consider carefully whether a minimum distribution is required. If possibly relevant regarding the plan’s provisions, consider whether the default beneficiary is or might be an eligible designated beneficiary. If the plan does not require a minimum distribution, consider whether an involuntary distribution is mandated, permitted, or precluded. Don’t assume a rollover; not every distribution is eligible for a rollover. If, after carefully following the plan’s default-beneficiary ordering, a default beneficiary is a decedent’s estate, consider whether the plan’s administrator would approve or deny a claim from a claimant who shows not a court’s appointment as a personal representative but rather a small-estate affidavit. This is not advice to anyone. Further, seeing how much effort this might be to dispose a < $1,000 account, should the plan sponsor consider writing more efficiency into the plan’s provisions?

To adjust the § 414(v)(7)(A) amount, its base period is “the calendar quarter beginning July 1, 2023[.]” https://benefitslink.com/boards/topic/80061-is-150000-the-limit-on-2025-fica-wages-before-a-participant-must-make-2026-age-based-catch-up-elective-deferrals-as-roth-contributions/ In July, I estimated that, for 2025 FICA wages to drive how § 414(v)(7) applies for 2026, the $145,000 will become $150,000. John Feldt, how’s my logic and my math?

If the plan is ERISA-governed and the employer/administrator seeks ERISA’s supersedure of a State’s wage-payment law, that likely requires a notice beyond merely having delivered a less-often-than-yearly summary plan description. ERISA § 514(e)(3)(A) requires that the plan’s administrator deliver the notice “within a reasonable period before [each] plan year[.]” 29 U.S.C. § 1144(e)(3)(A) http://uscode.house.gov/view.xhtml?req=(title:29%20section:1144%20edition:prelim)%20OR%20(granuleid:USC-prelim-title29-section1144)&f=treesort&edition=prelim&num=0&jumpTo=true. Likewise, if the plan sponsor or a participating employer wants the arrangement treated as an eligible automatic contribution arrangement, tax law requires a notice before each plan year. I.R.C. (26 U.S.C.) § 414(w)(4) (“before each plan year”); accord 26 C.F.R. § 1.414(w)-1(b)(3)(i) “for a plan year[.]” http://uscode.house.gov/view.xhtml?req=(title:26%20section:414%20edition:prelim)%20OR%20(granuleid:USC-prelim-title26-section414)&f=treesort&edition=prelim&num=0&jumpTo=true; https://www.ecfr.gov/current/title-26/part-1/section-1.414(w)-1#p-1.414(w)-1(b)(3)(i). Although I might suggest delivering a revised summary plan description before every year, many plans’ administrator don’t use that means of communication. This is not advice to anyone.

Paul I, you’re right to mention that Circular 230 rule. Here’s another outlook. Even when 31 C.F.R. § 10.37 applies or someone volunteers to follow it, the rule refers only to written advice. Further, even if a practitioner applies the rule to oral communications, one might distinguish between advice and information. Here’s § 10.37(a)(2)(vi): “The practitioner must—Not, in evaluating a Federal tax matter, take into account the possibility that a tax return will not be audited or that a matter will not be raised on audit.” Some advisers think it’s possible to provide advice or an evaluation about how law applies if all issues and facts are detected and fully pursued AND provide information, especially if one’s client asks, about whether an issue or failure might not be detected, or about whether an executive agency might compromise, or not pursue, enforcement. There is a difference between advising a client to do something, and informing a client about what seems likely or unlikely to happen if one’s client does a thing it describes. Many of us might prefer that persons obey law, even civil tax law. But I don’t always presume that it’s my right as an adviser to cause my advisee to obey law if that’s not the choice the principal would make for itself. I work to enhance my client’s autonomy by adding to its information and decision-making capabilities. (My clients welcome my judgment, but many don’t want to wholly abdicate the principal’s decision-making.) While I might not volunteer information about nondetection and nonenforcement, if a client asks I won’t give a dishonest answer. I might decline to answer. Or if I answer, I might not say any more than I know as fact. Yet, there also can be situations in which providing candid information about nondetection or nonenforcement might help a decision-maker think through legitimate choices with competing values. I’m mindful that some consider that providing information about nondetection or nonenforcement might in context be tantamount to telling one’s client to disobey law. About that, I show my summertime professional-conduct students the exchange of Yale Law Journal articles arguing different outlooks about that point. And other professional-conduct literature about whether an adviser has a “duty to the system.” I don’t see it as my personal responsibility in my work as an adviser to save the government from the government’s choices about law enforcement. (I might have some responsibilities as a citizen; but that’s distinct from my special-purpose role as an adviser.) If an advisee’s behavior is dispiriting, an adviser might consider one’s professional prerogative to withdraw one’s availability.

That these experienced advisers are not in entire concord suggests there are a few interpretations that might meet tax law’s standard for a “substantial authority” position—one that need not be flagged in a tax return. Does a TPA or other consultant write up an explanation of each of those interpretations, to let one’s client choose its risks and opportunities?

As you help the employer think about those and other questions, consider this: If there is a year for which expenses are more than what had been estimated and paid in as the “level-funded” amount, is the employer obligated to pay whatever is needed to meet all claims and other expenses? If the employer bears a bad experience, should it take a good experience? This is not advice to anyone.

Some clients make sincere efforts to administer a retirement plan correctly. Some do right things the right way for no more reason than they feel good about doing so. But some clients ask intelligent questions about how much enforcement an executive agency does. Some intuit that an agency lacks resources even to spot a potential failure. While many lawyers and other advisers don’t volunteer information about nondetection and nonenforcement, when a client asks one must give an honest answer. Even declining to describe how much effort an agency puts on an issue practically reveals the answer—not much, often none. If the tides at EBSA and IRS don’t change soon, the next generation of retirement-plans practitioners will miss experiences that could have been learning opportunities.

From a recent (Aug. 8, 2025, 5:38 PM EDT) Bloomberg Tax article: “Bessent will step in as House Republicans look to cut the [Internal Revenue Service’s] funding by more than 20% to $9.5 billion, an even bigger cut than the White House proposal.”

The Internal Revenue Service has had six people lead the agency this year. Now, Billy Long is out as IRS commissioner, with the Secretary of the Treasury serving as acting commissioner. The deputy commissioner position is vacant. The chief of staff role is vacant. More than half the officer positions are vacant or “(acting)”. What do we imagine about the pace of new examinations in 2025?

When a one-employer group has about 60 distinct plans, is it feasible to use (after data entry) a commercially-developed software product, or must the service provider do the work by custom-crafting?

When Congress provides a remedial-amendment period regarding a statute’s changes, a variation regarding collective bargaining protects an agreement made before the statute’s enactment. The idea is that an employer and the labor union ought not be required to open negotiations until the next time negotiations would regularly open nearing the end of the collective-bargaining agreement’s term. A premise of labor-management relations is that neither side unilaterally changes provisions about retirement benefits; it’s a subject for their bargaining. If the term of a CBA made before the statute’s enactment expired (or would have expired but for an extension or renewal), the employer and the union had an opportunity to discuss what retirement plan provisions they would make (or not) in response to the statute’s changes. This is not advice to anyone.

Paul I and CuseFan, thank you for your gracious help. Following Paul I’s observation that a user of IRS-preapproved documents could vary from them so much that the IRS would find that the user may not rely on the IRS’s opinion letter: Is the consequence only that the IRS would evaluate whether the written plan, including the user’s additions and variations, states provisions that meet § 401(a)’s tax-qualification conditions? Or is there some other consequence that results from having an individually-designed plan?

If an ERISA-governed retirement plan has access to a recordkeeper’s or third-party administrator’s IRS-preapproved plan documents, what plan provisions or other circumstances would lead a plan’s sponsor to state the plan using an individually-designed document? (I have no stake in this because my law practice is not, and won’t become, available to write a plan document for an ERISA-governed retirement plan.)

Am I right in guessing that most plans facing a meaningful risk that the plan might be top-heavy seldom engage a lawyer? And seldom engage an independent qualified public accountant? Leaving only a TPA to consider whether a top-heavy measure is counted correctly? If my surmise is right, are there variations for kinds of businesses? For example, if the employer has few nonkey employees, a disproportion of key employees, and is a small professional-services partnership, does it engage help beyond the TPA?

I express no view about what is or isn’t a related rollover. The CFR section cited above includes this: “[T.D. 7997, 49 FR 50646, Dec. 31, 1984, as amended by T.D. 8357, 56 FR 40550, Aug. 15, 1991; T.D. 9319, 72 FR 16929, Apr. 5, 2007; T.D. 9849, 84 FR 9234, Mar. 14, 2019]”. According to that note of the Federal Register history, the rule was published just months after the August 23 enactment of the Retirement Equity Act of 1984, which added ERISA commands and tax law recognizing a qualified domestic relations order. The 2019 amendment was in a Trump I cleanup of obsolete rules. Could the Treasury’s interpreters have thought about how QDROs relate to top-heavy measures when considering the 1991 amendment or the 2007 amendment? What, if anything, should an interpreter infer from what wasn’t revised? Is there a logic for suggesting that the amount moved from the originating participant’s account to the alternate payee/other participant’s account ought to count for top-heavy measures in one of those individual accounts (I don’t know which), but not both?

Remember how often BenefitsLink neighbors incant RTFD. Consider all documents governing the plan and its trusts. What does each document say about what ends a trusteeship? What does each document say about how a named fiduciary appoints a successor trustee, or an additional trustee? Was all that done? Consider all agreements and other documents with each investment provider. And with each service provider. What does each document say about an obligation to give a party, and perhaps third persons, notice of a change in a trusteeship? Was all that done? If careful readings don’t resolve all questions, consider ERISA (if ERISA governs the plan). If there is a question unanswered by the documents and the statute’s text, consider the Federal common law of ERISA. If the Federal common law of ERISA applies or is relevant to help interpret the statute, consider the American Law Institute’s Restatement of Trusts as an expression of common law. Evaluate carefully whether the recent appointment was of a successor trustee, or of an additional trustee. Consider whether the successor or additional trustee should request that the deceased trustee's personal representative submit an accounting for whatever the decedent did since his most recent accounting. Consider whether the successor or additional trustee should request that the deceased trustee’s personal representative deliver all trust records the personal representative possesses or could possess. Consider whether the successor or additional trustee should request that the deceased trustee’s personal representative confirm in writing that the representation has not done anything to administer the trust. Even if all of what was done was correct under the plan’s and its trust’s governing documents and applicable law, consider redocumenting the changes in the trusteeships. Why? An investment or service provider’s employee might lack discretion to allow anything beyond what a checklist tells the employee to do. This is not advice to anyone.

Paul I, thank you for thoughtful information. Your observations about payroll operations, and about coordinations among a plan’s participating employers, each’s paymaster, each paymaster’s service provider or software supplier, the retirement plan’s administrator, and its service providers are helpful. For a big-enough plan (or for a good relationship), some services with a recordkeeper are negotiable. For actual administration in 2026 and later years, my clients will feed, each year, the recordkeeper’s § 414(v)(7) indicator. But for communications before that indicator is usable, I welcome RatherBeGolfing’s idea if an employer prefers to pay someone else to do a sort about which workers get a communication suggesting one consider changing her deferral elections before December ends.

Here’s the Treasury’s rule: T-32 Q. How are rollovers and plan-to-plan transfers treated in testing whether a plan is top-heavy? A. The rules for handling rollovers and transfers depend upon whether they are unrelated (both initiated by the employee and made from a plan maintained by one employer to a plan maintained by another employer) or related (a rollover or transfer either not initiated by the employee or made to a plan maintained by the same employer). Generally, a rollover or transfer made incident to a merger or consolidation of two or more plans or the division of a single plan into two or more plans will not be treated as being initiated by the employee. The fact that the employer initiated the distribution does not mean that the rollover was not initiated by the employee. For purposes of determining whether two employers are to be treated as the same employer, all employers aggregated under section 414(b), (c) or (m) are treated as the same employer. In the case of unrelated rollovers and transfers, (1) the plan making the distribution or transfer is to count the distribution as a distribution under section 416(g)(3), and (2) the plan accepting the rollover or transfer is not to consider the rollover or transfer as part of the accrued benefit if such rollover or transfer was accepted after December 31, 1983, but is to consider it as part of the accrued benefit if such rollover or transfer was accepted prior to January 1, 1984. In the case of related rollovers and transfers, the plan making the distribution or transfer is not to count the distribution or transfer under section 416(g)(3) and the plan accepting the rollover or transfer counts the rollover or transfer in the present value of the accrued benefits. Rules for related rollovers and transfers do not depend on whether the rollover or transfer was accepted prior to January 1, 1984. 26 C.F.R. § 1.416-1 https://www.ecfr.gov/current/title-26/section-1.416-1. BenefitsLink mavens, interpret away!

RatherBeGolfing, thank you; that's helpful information about another way.

For anyone who or that seeks to rely on ERISA § 408(b)(17) to exempt a § 406(a)(1)(A), § 406(a)(1)(B), or § 406(a)(1)(D) (not –(C) nor –(E)) party-in-interest transaction, it’s essential to get a lawyer’s written advice. Even when a directed trustee’s responsibility is as sharply limited as lawyers can imagine, consider that a trustee decides whether the directing fiduciary’s direction is a “proper direction[] . . . and which [is] not contrary to [ERISA][.]” ERISA § 403(a)(1). Interpreting that paragraph and ERISA § 406(a)(1), some might say a directed trustee ought not to follow a direction to do something one “knows or should know”—exercising “the care, skill, prudence, and diligence” ERISA § 404(a)(1)(B) requires—would result in a nonexempt prohibited transaction. If a transaction’s counterparty is the trustee’s affiliate, consider whether it might be self-interested for the trustee to be a judge of whether § 408(b)(17)’s conditions are met. Even if the directing fiduciary affirms that it finds the transaction meets § 408(b)(17)’s conditions, a directed trustee might not abdicate its own responsibility to decide whether the direction is “proper” and “not contrary to” ERISA. If that’s a concern, there might be ways to rearrange trusteeships so assets involved in a to-be-exempted transaction are not under any trusteeship of any interested person. This is not advice to anyone.

WCC, thank you for your reminder that recordkeepers are building an indicator that’s particular to § 414(v)(7) treatment. (None of the recordkeepers I’ve talked with has yet completed its build of that indicator, even to receive a preliminary data feed.) My query isn’t about applying the provision for 2026 and later years; it’s about sorting, now, for which participants get September 2025 and follow-up communications informing one about choices that might be required for next year’s deferrals. (And about a default election to be applied when the without-catch-up limit is exhausted, which could be as soon as January for some § 414(v)(7)-affected employees.) The employer seeks a sort for participants who might be § 414(v)(7)-affected, and hopes the recordkeeper could do that work without any effort from the employer. For example, a sort for participants with regular compensation > $120,000 and an indicator showing not self-employed. I see that their hope is too much. The employer will need to sort and drive the communications.

Of the many tax-qualification conditions and other rules a recordkeeper’s system hopes to help a retirement plan’s administrator apply, a few might turn on whether a participant is an employee or is a self-employed individual (who might be treated as a deemed employee). Among these, a § 414(v)(7) restriction against non-Roth catch-up deferrals does not apply to a participant who is a self-employed individual (who has no FICA wages). I’ve seen in recordkeepers’ systems Yes-or-No indicators for whether a participant is: union-represented, treated as an insider for trading in employer securities, an officer of the employer, or a super-officer. Does a recordkeeper have an indicator for whether a participant is an employee or is a self-employed individual? (I recognize a use of this depends on the census information furnished to the recordkeeper.) A sort for participants who might be § 414(v)(7)-affected might look for those with compensation that suggests that FICA wages for a relevant year might exceed $145,000/$150,000. But without a further sort, that might result in “false positives” by including deemed employees who have compensation but no FICA wages. (Imagine a professional-services business in which hundreds of workers are partners.) Could an employee-or-self indicator sort out this out? If a recordkeeper lacks such an indicator but has a field for ownership percentage, might one use that as a way to classify a self-employed individual? For example, if a worker’s ownership percentage is less than 1% (so it doesn’t trigger other rules) and perhaps as little as 0.0001, could that classify the participant as one who can’t be § 414(v)(7)-affected? (I’m mindful that an employer has the facts, and could control the plan administrator’s communications. But I seek to learn about what communications a recordkeeper can do without the employer/administrator’s effort.)