Peter Gulia

MoJo, your recordkeeper’s way is much better than many others.

If one translates the communication to what might (depending on the service agreement) be arguably legitimate: The service provider offers its service to support your plan’s provision for a small-balance involuntary distribution only if your plan’s provision sets its condition at the maximum amount ERISA and the Internal Revenue Code permit. Many service agreements include provisions designed to support service changes. If so, the service recipient’s choices might be: 1) Fall in with the recordkeeper’s business preference. 2) Change the plan to omit a provision for a small-balance involuntary distribution. 3) Don’t change the plan, but accept that the recordkeeper will perform its services applying the maximum amount. 4) Find another recordkeeper. 5) Administer the plan without a recordkeeper. Many service recipients will find choices 3-5 impractical. If I were to advise a plan sponsor (one that also serves as its plan’s administrator and trustee) that received this communication, my advice might be grounded, in some (not all) meaningful parts, on (1) whether the service provider has a contract right to treat the service change as accepted if not expressly rejected; (2) the service recipient’s overall working relationship with the service provider; (3) whether the service recipient has bargaining power; and (4) whether it is feasible for the service recipient to find a recordkeeper that would behave better.

Jaded, thank you for sharing the information. About the situation for which the National Labor Relations Board asserted an unfair labor practice: 1) Did the collective-bargaining agreement provide for the union-covered workers to participate in a multiemployer plan or a union plan? 2) Did the collective-bargaining agreement provide, expressly, that the union-covered workers are excluded from the employer’s plans? 3) If neither #1 nor #2, did the collective-bargaining agreement at least recite that retirement benefits had been considered in the bargaining?

EBECatty, if, after you complete your research and analysis, the ESOP-owned corporation and Joe’s corporation are one employer, could the problems that result be solved with qualified separate lines of business? Or is QSLOB treatment unavailable in the circumstances? Or would QSLOB treatment not solve the particular problem Joe seeks to resolve?

MoJo is right that a careful lawyer advises her client not only about whether the facts might or might not support a § 415 restorative payment but also about whether providing restoration, even with no concession, might alert someone about a fiduciary’s arguable breach and a loss or harm beyond the one on which the fiduciary provided restoration. Also, banks and insurers offer many ways to resolve a market-value adjustment (if a plan or its participant has not by other means done something for the adjustment not to apply).

Unless you work for the national firm, consider suggesting that the plan’s administrator lawyer-up to evaluate whether the administrator has good claims against the third-party administrator. Yet, a lawyer so engaged might fair-mindedly consider that the fact that a contracted service was not performed does not necessarily mean the service provider breached its contract. Until the whole story is discovered, it’s at least possible that the service recipient did not meet a condition. That recognized, a lawyer doesn’t concede her client’s weaknesses, and might assert anything that can be stated truthfully. Further, consider that the employer-administrator might want to protect the secrecy of its communications with evidence law’s privileges for lawyer-client communications and attorney work product.

The Internal Revenue Code § 401(a)(4) and § 410(b)(3)(A) exclusions refer to “employees who are included in a unit of employees covered by an agreement [that] the Secretary of Labor finds to be a collective bargaining agreement between employee representatives and one or more employers, if there is evidence that retirement benefits were the subject of good[-]faith bargaining between such employee representatives and such employer or employers[.]” So, before a § 401(k) plan’s administrator decides to run coverage and nondiscrimination testing without the second union’s covered workers, one might read the collective-bargaining agreement. A collective-bargaining agreement that follows labor-relations norms typically has some chapter, article, or section with a heading that uses the word “retirement”. Or if you don’t see something like that, look for a sentence saying what topics the parties considered. It can be okay that a covered worker has no access to a plan with an elective-deferral arrangement if the worker is covered by some retirement benefit. That might be a union, multiemployer, multiple-employer, or single-employer defined-benefit plan. But if a collective-bargaining agreement says nothing about which retirement plan the covered workers participate in and lacks a recital that the parties bargained over retirement benefits, a tax practitioner might not assume the § 401(a)(4) and § 410(b)(3)(A) exclusions until “there is evidence that retirement benefits were the subject of good[-]faith bargaining[.]”

Thanks. My point is that, even for a plan sponsor that uses an IRS-preapproved document, spending a few minutes at the plan-documents stage—to read the default-beneficiary provision, or even without reading to replace that default with the plan sponsor’s preference—can save lots of foolishness later.

Here’s the reopening of the comment period: https://www.govinfo.gov/content/pkg/FR-2023-02-14/pdf/2023-02545.pdf. That comment period closed April 17, 2023. https://www.reginfo.gov/public/do/eAgendaViewRule?pubId=202310&RIN=1210-AB64 One year would be quick in rulemaking time, especially when other topics consumed attention.

I concur with your observation that, ideally, the government agencies could have done better work on their rulemaking. I believe those who worked on the project sincerely did the best they could, facing constraints on their time and attention and the limits of language. Life is imperfect. We observe together that many clients don’t like spending money for a lawyer to parse text interpretation on what a client imagines ought to be a straightforward question with a simple answer. But let it be the client that decides not to spend money on advice (beyond whatever advice you provide without seeking an incremental fee). And with or without advice, let it be the client that owns the consequences of its choices, whichever choices it makes. Observe that for the underlying question we remark on, there is at least one risk in either direction. Deciding that with-a-balance omits an allocation receivable risks that the administrator doesn’t engage an audit when ERISA required it. Deciding that with-a-balance counts an allocation receivable risks that the administrator spends money—which might be the employer’s money, or some participants’ money—on an audit ERISA didn’t require. Governments often write ambiguous laws, rules, and instructions. An adviser tries to help her client deal with an ambiguity. But that help need not relieve an advisee’s responsibility for choosing.

Another lesson we might take from this: Different service providers’ documents vary on what a base document provides as the default when there is no participant-named beneficiary. That’s so even for service providers that license documents from the same vendor. Even within one service provider, defaults might vary by documents for different business lines. Except for providing a death benefit to a surviving spouse, a provision about a default beneficiary might be an “administrative provision” a user may change without defeating reliance on an IRS-preapproved document’s opinion letter. Before adopting a document (including an amendment or a restatement), a plan sponsor should at least read what default the document would provide. A plan sponsor should consider whether the proposed default fits the sponsor’s interests. If the plan sponsor serves as the plan’s administrator, one might consider whether the proposed default could lead to difficult or inefficient administration. Which default provisions produce which kinds of frustrations and inefficiencies can vary with a particular plan’s other provisions, and with the employer’s workforce and the plan’s participants. While some employers might think a default-beneficiary provision isn’t worthwhile to think about at a plan-documents stage, even one beneficiary situation might cost more time and attention than what the plan sponsor would have used on getting the documents right.

Whatever the instruction might mean, imagine a big recordkeeper might not apply it with an allocation receivable in generating draft Form 5500 reports. Imagine a recordkeeper counts participants with an account balance by looking to what “the system” says was the balance actually credited on December 31. If the participants-with-balances count in a draft Form 5500 report suggests the plan’s administrator need not engage an independent qualified public accountant, often the administrator will accept that assumption. An administrator that assumes its plan was, for a to-be-reported year, a small plan often has not engaged a CPA firm to audit the plan’s financial statements; so no auditor probes any count of participants. Administrators of these plans might not think to ask a lawyer for advice. Not every plan’s administrator has engaged a TPA other than the recordkeeper. So for many plans, an administrator might have no one questioning the recordkeeper’s counts. After considering those possibilities (some might say likelihoods), how many 2023 Form 5500 reports will be filed using a with-a-balance count that omitted participants with only an allocation receivable? Then, imagine the Labor department, lobbied by big businesses, publishes revised instructions to clarify that with-a-balance means a balance actually credited on the applicable date, and need not consider an allocation receivable. If austin3515 suggested a plan’s administrator count as with-a-balance a participant with only an allocation receivable, might your small-business client be displeased with what some might perceive as your overly cautious advice? Even when a client prefers not to be bothered with questions it expects a professional to resolve, is it safer at least to ask one’s client its choice about something that involves spending, as you put it, an extra $20,000?

Before one suggests asking the Labor department to clarify what “with a balance” means, consider the proverbial saying: “Be careful what you wish for, . . . .” Or another practical caution: “Don’t unnecessarily ask a question if you’re not sure you’ll get the answer you’d like.”

If a prospective audit client’s starting point is that the plan’s administrator (or “management” in auditor jargon) did not file required reports, a CPA firm might think it’s difficult to make the engagement profitable. If CPAs find the auditee has weak controls or, worse, missing controls, generally accepted auditing standards (“GAAS”) require an auditor to widen and intensify controls testing and audit procedures. That the plan’s administrator didn’t know ERISA and the Internal Revenue Code required it to file yearly reports suggests at least one weak or missing control. And one wonders about what else was not done. Further, an auditor cannot rely on the recordkeeper’s controls if the auditor knows (or using “appropriate professional skepticism” should consider) that the plan’s administrator did not operate what that report describes as the compensating controls. While in theory a CPA firm might increase its fee to get the extra work paid for (and still get a normal or reasonable return to margin), in the real world it might be impractical to increase the fee that much. TPApril, one way to persuade a CPA firm to take on a difficult engagement is to reach out to a firm that knows and trusts you, and persuade the engagement partner that your new client is committed to doing everything you advise them to do.

The Form 5500 Instructions states: Line 6g. Enter in line 6g(1) the total number of participants included on line 5 (total participants at the beginning of the plan year) who have account balances at the beginning of the plan year. Enter in line 6g(2) the total number of participants included on line 6f (total participants at the end of the plan year) who have account balances at the end of the plan year. For example, for a Code section 401(k) plan, the number entered on line 6g(2) should be the number of participants counted on line 6f who have made a contribution, or for whom a contribution has been made, to the plan for this plan year or any prior plan year. . . . . https://www.dol.gov/sites/dolgov/files/ebsa/employers-and-advisers/plan-administration-and-compliance/reporting-and-filing/form-5500/2023-instructions.pdf I’m unaware of any Labor department guidance that further explains whether an account balance (whether as at an end-of-year, or as of a beginning-of-year) includes an allocation from a contribution receivable (or of a distribution payable). I could interpret the instruction several different ways. Also, some aspects of those interpretations might vary with whether the plan provides or omits participant-directed investment. To show only one partial path: The instruction’s nonrestrictive illustration refers to whether “a contribution has been made[.]” Perhaps made and owing mean different things. As always, it’s the plan administrator’s decision, with whatever advice the administrator considers. But a service provider might make business decisions about what services it offers or provides.

Thank you for your kind help.

For a plan that sticks with Teachers Insurance and Annuity Association of America, do you like TIAA’s Target Date Plus model portfolios? Why or why not?

David Rigby, thank you for telling me about another way actuaries work. Do you have a guess on the relative portions of actuaries who work 85% or more on one or the other of the two classes—db or dc—of retirement plans?

For my academic writing, I hope BenefitsLink neighbors will help me with some guesses or observations about what actuaries work on. How many actuaries work 85% or more of one’s time on defined-benefit pension plans? How many actuaries work 85% or more of one’s time on individual-account plans? For actuaries between those outer ranges, what’s the split between db and dc plans? Of work done for individual-account plans, how much (if any) requires an enrolled actuary’s certificate? Of work done for individual-account plans, how much (if any) requires math skills beyond those possessed by other educated people? And although I’m focusing on actuaries in this initial query, I might ask similar questions about people who hold another license or credential.

Under a Treasury rule, a payment a fiduciary makes (and uniformly applies regarding all similarly situated participants) when the fiduciary faces “a reasonable risk of liability” might be a restorative payment, treated as not an annual addition. 26 C.F.R. § 1.415(c)-1(b)(2)(ii)(C) https://www.ecfr.gov/current/title-26/part-1/section-1.415(c)-1#p-1.415(c)-1(b)(2)(ii)(C). A fiduciary’s breach need not be proven or conceded; it is enough that there is “a reasonable risk of liability[.]” (The rule is wider than the earlier Revenue Ruling. And a rule is more reliable than nonrule guidance.) Even if all related decisions were proper and prudent, selecting a stable-value contract or deciding to make it a designated investment alternative (or continuing either decision) might have been a breach (or might set up facts allowing a complaint plausibly to assert a breach) if the fiduciary then knew—or had it exercised the care, skill, prudence, and diligence then required, would have known—that there was more than a remote possibility that the business would be acquired, or even that an owner might seek to sell the business. (One might presume a prudent fiduciary knows that a careful business acquirer typically requires the target to end its retirement plan before closing.) Or, if the plan’s fiduciary finds there is no “reasonable risk of liability” and that the adjustment is an annual addition, allocations of the adjustment might fit within all or most participants’ annual-additions limit ($69,000 [2024]) and might comport with coverage and nondiscrimination conditions. Either way, be careful if the restoration or adjustment disproportionately favors highly-compensated employees or affects a decision-maker’s individual account.

Here’s an anecdotal observation about an effect of tax law’s top-heavy condition: Some plan creations might be lost because a business owner is unwilling to obligate her business to a safe-harbor design, and lacks tax-law advice from a smart person like those in this discussion. I remember a service provider that rejected prospective customers a salesperson had sold because the provider feared that a plan—if not reformed into a safe-harbor design—could become top-heavy, and that the customer would blame the service provider. (“Why didn’t anybody tell me . . . !!”) The business executives decided that no set of explanations and warnings—no matter how clearly, conspicuously, loudly, and onerously stated—would deter frustrated customers from blaming the service provider. The service provider operationalized this fear by setting a minimum number of eligible employees, below which any but a safe-harbor plan was rejected. Even with a skilled and motivated sales force, most of the rejected prospects were unwilling to adopt a safe-harbor design. Many refused even to consider it. I describe one illustration, but my experiences with many recordkeepers and third-party administrators reveals the business problem as common. For many reasons (including some the GAO report mentions), it’s impossible to know how many plan creations are lost because of the top-heavy condition. But is the number something more than zero? I don’t here mention my views about minimum-participation, coverage, nondiscrimination, and top-heavy positive laws or tax-law conditions. And I don’t mention my views about designing taxes or tax expenditures.

So, consider seeking a court’s order that names as the alternate payee the former spouse (using only that person’s name), and recites that the order relates to the former spouse’s marital property rights. Your client will want your advice about an executor's, administrator's, or other personal representative's powers to negotiate such a payment.

The tax-law condition about a required beginning date is a no-later-than condition. Meeting § 401(a)(9) does not preclude an earlier required beginning date. A plan’s sponsor may set a younger age if it does not compel an involuntary distribution before the participant’s normal retirement age, and comports with other tax-qualification conditions. Tax law recognizes a plan’s administration contrary to its current documents if that administration later becomes legitimated by a remedial amendment. But what if a plan’s administrator doesn’t yet know what an anticipated amendment will be? And doesn’t know even what the plan’s sponsor intends for a later amendment? (I recognize that getting this information often is about an employer talking within itself, at least for many single-employer plans. But to see an awkwardness that results from tax law’s remedial-amendment tolerance, let’s follow ERISA’s distinction between a plan’s sponsor and the plan’s administrator.) Imagine a plan for which the current documents set a required beginning date in relation to age 70½. Imagine that the plan’s sponsor has not communicated to the plan’s administrator that the sponsor intends to amend the provision to refer to some other age. In those circumstances, the plan’s administrator might interpret 70½ to mean 72, 73, or 75 (as fits the particular participant), but also might interpret 70½ to mean 70½. Some service providers have blithely assumed every plan’s sponsor intends to amend a plan to provide for a required beginning date’s age the latest age § 401(a)(9) permits. Perhaps that’s almost universally so. But shouldn’t a service provider ask the question? Even if it’s the ubiquitous “we assume you want x, unless your written instruction tells us otherwise.” Or even, “we provide our service assuming your plan provides x. If your plan provides something else, you must administer your plan without relying on our service.”

I express no view about who is or isn’t a participant for Form 5500 reporting. The line 5 instructions include this: “Participant” for purpose of lines 5a–5c(2) means any individual who is included in one of the categories below. 1. Active participants (for example, any individuals who are currently in employment covered by the plan and who are earning or retaining credited service under the plan) including: . . . . , and Any nonvested individuals who are earning or retaining credited service under the plan. . . . . https://www.dol.gov/sites/dolgov/files/ebsa/employers-and-advisers/plan-administration-and-compliance/reporting-and-filing/form-5500/2023-instructions.pdf

If a plan accepts an individual’s rollover contribution and credits the amount to the individual’s account, she is a participant, at least within ERISA § 3(7)’s meaning. That an individual has not met the plan’s conditions for sharing in a nonelective contribution or matching contribution, or even for eligibility to elect an elective-deferral contribution, does not mean that the individual is not a participant. A textualist, but acontextual, reading of the line 5 instructions might support a different finding for Form 5500 reporting. But caution suggests counting an individual who has an account balance, even if she has not entered the plan for anything other than the rollover contribution.