Peter Gulia

You are right. Even more broadly than your reading of ERISA sections 101 and 105, nothing in part 1 or any part of subtitle B of title I of ERISA governs a plan if the plan is not ERISA-governed because the plan is not an employee-benefit plan within the meaning of ERISA § 3. A nonapplication of ERISA’s title I does not, at least not by itself, preclude a plan’s administrator from generating an illustration; but that’s a choice, rather than a response to ERISA § 105’s command.

Even if her investigation has a focus about one employer, the Secretary of Labor has broad powers to examine almost anything about an ERISA-governed employee-benefit plan. With further powers, the Secretary may “require the submission of reports, books, and records, and the filing of data in support of any information required to be filed with the Secretary under [ERISA].” For example, the Labor department may require production of every record behind any entry in the whole pooled-employer plan’s Form 5500 report. ERISA § 504(a)(1), 29 U.S.C. § 1134(a)(1) http://uscode.house.gov/view.xhtml?req=granuleid:USC-prelim-title29-section1134&num=0&edition=prelim. And that power applies even if the Labor department has no reason to suspect even a potential ERISA violation. Compare § 504(a)(1) with § 504(a)(2).

The tax-qualification condition for some nongovernmental and nonchurch § 401(k) or § 403(b) elective-deferral arrangements to include an automatic-contribution arrangement if the elective-deferral arrangement was not established before December 29, 2022 begins with a plan year that begins on or after January 1, 2025. Perhaps in the last few months of 2024 the Internal Revenue Service might announce a nonenforcement delay of Internal Revenue Code § 414A? If PredictIt [https://www.predictit.org/] were to offer a yes-or-no future on whether the IRS announces a nonenforcement by December 31, 2024, how much would you pay for each $1.00 yes future?

About a tax credit regarding startup expenses other than contributions: I have never interpreted any tax credit about retirement plans. Yet, here’s a general clue to consider: Does the text of the Internal Revenue Code section that provides the credit refer to expenses the employer “paid or incurred”? In the Internal Revenue Code, a phrase of that kind might signal alternatives related to whether the taxpayer’s accounting uses “the cash receipts and disbursements method” [§ 446(c)(1)], “an accrual method” [§ 446(c)(2)], or something else, which might include tax accounting conventions the Code provides and combinations of allowed methods [§ 446(c)(3)-(4)]. If the statute’s text doesn’t answer your question, one might favor an interpretation that is logically consistent with the employer’s tax accounting method. Also, a tax credit might allow a taxpayer some choice about which tax year is the first for a number of years of the credit.

Even if a delay might be unstated in a Form 5500 and otherwise not a big deal, a fiduciary that can work to push itself and its recordkeeper to prompt processing of contributions and other payroll payments should try to do so, if it does not detract from higher-order fiduciary responsibilities.

Before sorting out whether a creation of a new pension plan might have a minimum participation, coverage, or nondiscrimination issue: Consider suggesting that the former business owner seek, if he hasn’t already done so, his lawyers’, accountants’, tax advisers’, investment managers’, and financial-planning advisers’ advice about whether creating a pension plan fits his interests and the considered integration of the whole of his planning. What to do after an operating business’s sale of its assets often calls for full-picture advice, involving everyone who might advise or handle useful information.

If there is no Treasury regulation, a tax return might assert the taxpayer’s interpretation of the statute, and may do so with attaching a Form 8275-R. https://www.irs.gov/forms-pubs/about-form-8275-r. If a tax-return position is supported by substantial authority (which may be “a well-reasoned construction of the applicable statutory provision”), one need not attach Form 8275 to avoid an understatement penalty. 26 C.F.R. § 1.6662-4(d)(3)(ii) https://www.ecfr.gov/current/title-26/part-1/section-1.6662-4#p-1.6662-4(d)(3)(ii). If a tax-return position is less confident than substantial authority (which can be less confident than more likely than not [51%]) but has at least a reasonable basis and is disclosed (using Form 8275), this too avoids an understatement penalty. https://www.irs.gov/forms-pubs/about-form-8275. A certified public accountant who obeys AICPA professional-conduct standards does not recommend a tax-return position or prepare or sign a tax return taking a position unless the CPA “has a good-faith belief that the position has at least a realistic possibility of being sustained administratively or judicially on its merits if challenged.” Or, a CPA may prepare or sign a tax return that reflects a position if the CPA finds “there is a reasonable basis for the position and the position is appropriately disclosed.” Recordkeepers, third-party administrators, and other service providers often wish for guidance to interpret recent (and sometimes not-so-recent, or even decades-ago) tax legislation about retirement plans. But an absence of guidance sometimes affords a wider range of interpretations.

Some signers use for Form 5500 reporting authorizations a signature that is deliberately illegible and unlike the signature the signer uses for her personal banking, credit-card, and other financial matters.

Not only for the circumstances you describe but also for other interests that might be met by segregating owners from employees, many practitioners set up two plans. The first plan has only owners (including spouses of owners), excluding employees. The second plan has only employees, excluding owners and deemed owners. To the extent a test of coverage or nondiscrimination is needed, one looks to the appropriate aggregation of the two plans. In this setup, the first plan is a non-ERISA plan, and the second plan is an ERISA-governed plan. For the second plan, it might happen that no participant chooses for her investment any nonqualifying asset. If so (and if the count of participants with balances remains small), ERISA § 103 would not require an independent qualified public accountant’s audit of the plan’s financial statements. About an ERISA § 412 fidelity-bond insurance for the second plan, if that plan’s assets is less than $10,000, the required coverage is $1,000. (I don’t know what price an insurer seeks for a $1,000 coverage limit, perhaps a minimum premium based more on the records bother than the risk insured.) If, as you conjecture, the one LTPT participant chooses no elective deferral (and gets no employer-provided contribution), the second plan’s Form 5500 report would show one participant, zero participants with a balance, $0 in plan assets, and so on. Yet, the employer/administrator would want to file the reports.

Might this speak to your question? “The account balance is increased by the amount of any contributions or forfeitures allocated to the account balance as of dates in the valuation calendar year after the valuation date. For this purpose, contributions that are allocated to the account balance as of dates in the valuation calendar year after the valuation date, but that are not actually made during the valuation calendar year, are permitted to be excluded.” 26 C.F.R. § 1.401(a)(9)-5/Q&A-3(b) (emphasis added) https://www.ecfr.gov/current/title-26/chapter-I/subchapter-A/part-1/subject-group-ECFR6f8c3724b50e44d/section-1.401(a)(9)-5.

Paul I, thanks again! Yes, I presume most distribution claims are submitted online, not paper. I hope that allows more opportunity to present the idea that a plan might allow more than one kind of distribution on the same facts, and that a participant might choose which kind fits her needs and interests. Electronic claims also allows quicker and cleaner use of information already known and suppression of information that cannot apply in an individual’s circumstances. I think almost not at all about § 402(f) notices, waivers of that notice period, § 3405 withholding explanations, § 3405 elections (to the extent allowed), and other communications tax law requires. No matter how bad the writing in the IRS’s norms, it’s too hard to ask a recordkeeper to customize those elements. And anything I might imagine about ways to streamline claims might see no use. None of my plan clients brings a recordkeeper enough profit that it would change a general method it decided on. Rather, my hope is that some recordkeepers, especially those that program one’s own software or build overlays and interfaces to improve uses of a supplier’s software, might consider designing systems not only for wholly computerized processing of claims for voluntary distributions but also for friendliness to participants in informing choices. Paul I, thank you for indulging me with this thought exercise.

Paul I, thank you for your wonderfully helpful explanation of some disadvantages of the one-form idea. If you’ll indulge me a little more, here’s some follow-up questions: I like the idea of a communication—ideally, no more than one visually attractive page—that explains the kinds of distributions and the conditions for and characteristic of each. Imagine a plan has the resources to work on plain-language writing so this is at an eighth-grade reading level and yet explains everything a participant might want to know to make an informed choice. Here’s the hard part: If that page is not included within each claim form, or at least the normal and hardship claim forms, how do we get anyone to read the one-page explanation? About a difficulty some might encounter if different kinds of claims call for different information, might that outlook be different if for every claim the only evidence a claimant provides is self-certifying the statements presented on the form? About a worry that a claimant might mistakenly fill-in information in each part, what if each part for each kind of distribution has nothing to fill-in? The fill-in for the claimant’s name and taxpayer identification number would be near the beginning of the form. At the end of the form would be the amount requested, a series of checkboxes for the kind of distribution claimed, and the self-certification that the claimant met the conditions for the kind of distribution claimed. In the middle would be the text for each set of self-certifying conditions, one set of which the claimant would adopt with the checkbox. Does that work? Or are there practical reasons it wouldn’t? (Assume none of the claims ever calls for a spouse’s consent.) (Assume the service provider is instructed to pay every good-order claim, deny every not-in-good-order claim, and never bother the plan’s administrator until a claimant challenges a denial.) And BenefitsLink neighbors, any further or different observations?

If (for 2024) a § 401(k), § 403(b), or governmental § 457(b) plan provides many kinds of distributions such a plan may provide without tax-disqualifying the plan, a participant might face a choice of two or more kinds of distributions allowed on the same set of facts. (For now, let’s ignore anything about a § 402A(e) emergency savings account.) Each kind of distribution might invoke advantages and disadvantages all or some of which would not result from another of the kinds that could be taken in the same set of facts. This need for a participant to choose which kind of distribution one claims can occur not only before severance from employment but also after severance. I worry that if a plan’s administration uses a distinct claim form for each kind of distribution, a severed participant might reflexively use the form for a “normal” distribution, and so might miss an opportunity to claim a different kind of distribution with different features. Or a before-severance participant might reflexively use the hardship form, perhaps missing an opportunity to consider another claim that would preserve advantages a hardship distribution lacks. For example, imagine a 401(k) participant in his 30s or 40s who recently severed from employment. He wants to take $5,000 to help meet expenses. He could get what he wants stating no fact beyond his severance from employment. Yet on his facts, he also could get a qualified birth-or-adoption distribution. Taking that § 72(t)(2)(H) distribution would result in a Form 1099-R coded for the IRS not to look for an extra 10% too-early tax. And it would preserve the distributee’s opportunity to repay the amount into a retirement plan. If a plan’s administration uses a distinct claim form for each kind of distribution (and the participant had not read carefully the plan’s communications about the several kinds of distributions allowed), how would he know he could choose a qualified birth-or-adoption distribution? Or imagine a before-severance participant whose need would be met if she claims no more than $1,000 and gets what results after Federal income tax withholding. The participant’s request to the call center says she needs money, but doesn’t mention that $1,000 would be enough. If not carefully scripted and trained, might a call-center worker—facing pressures on his time and attention—reflexively send or point to only the hardship form? How would the participant know that she might want the form for a § 72(t)(2)(I) emergency personal expense distribution? BenefitsLink neighbors, what do you think: Should a plan’s administrator—practically, its recordkeeper or third-party administrator—put all the available kinds of distribution on one claim form? Why or why not? What would be the advantages? What would be the disadvantages or difficulties? If you think its unwise or impractical to put all or many kinds of distribution on one claim form, what methods would you suggest to inform a participant about one’s opportunity to consider different kinds?

Apart from whatever Federal COBRA might or might not provide: A State’s “mini-COBRA” law regulates an insurer and its insurance contract; it does not regulate an ERISA-governed employee-benefit plan, and does not govern an employer or former employer (except insofar as one has an obligation under a regulated group insurance contract). That a buyer acquired all the shares of a seller does not by itself mean that the buyer assumed the seller’s obligations under a group insurance contract. Yet, a “mini-COBRA” continuee might want his or her lawyer’s advice about the insurer’s obligations under its insurance contract and the continuee’s rights regarding the insurer. That a seller and former employer ended its group health plan might not necessarily extinguish all the insurer’s obligations under its group insurance contract with the former employer. Rather, just as retirement-plans people say Read The Fabulous Document, the situation the originating post describes might call for Read The Felicitous Contract (including that contract’s governing State insurance law).

Belgarath, it seems to me there are two layers of difficulties: 1. For at least some plans, there are ambiguities about exactly what the 2019 and 2022 statutes set as ERISA title I’s command and the Internal Revenue Code’s tax-qualification condition. With its adviser’s help, the decision-maker needs to decide its interpretations so one can discern the plan’s implied provisions, including those put in operation under a remedial-amendment regime. On this difficulty, an adviser might help by providing her advice about the range of permissible, plausible, and practical interpretations, and the range of probabilities about whether an interpretation later would be found to have been used in good faith with reasonable cause. If a client needs or wants that advice before the IRS has released guidance, an adviser does what she can with the information available when she finishes her advice. 2. For a plan that faces no ambiguity or has (at least temporarily) resolved it, there remains a difficulty that the employer/administrator might fail to administer the plan’s implied provisions. Against that difficulty, there might not be much for an adviser to do beyond explaining the plan’s implied provisions and the whole range of potential consequences of not administering them. Query (and I ask this not knowing the practical world in which the LTPT problems live): Is it feasible for a plan administrator’s summary plan description, summary of material modifications, or similar communication however labeled (sent to might-become-eligible employees) to describe the elective-deferral opportunity and who’s eligible for it, and by doing so shift a responsibility to the employee? Whether an employee meets eligibility under the plan’s general conditions or the LTPT conditions, perhaps ERISA § 102 does not require the plan’s administrator to tell an employee that she met the conditions and has become eligible (if the needed information about the eligibility conditions was previously communicated in an SPD or SMM). (I’m aware there are some nice questions about whether ERISA § 404(a) in some circumstances requires more communication than ERISA’s part 1 requires.) I’m aware some administrators (typically through a service provider’s work) furnish something to remind an employee that she met (or is about to meet) a service condition and has become (or soon will become) eligible to elect for or against elective deferrals. But perhaps ERISA § 102 does not command that courtesy. For a plan with no automatic-contribution arrangement, isn’t it the participant’s responsibility to affirmatively elect elective deferrals? If so, might an employer/administrator be responsible for an omission of elective deferrals only if the employer refused or neglected to process the employee’s submitted wage-reduction agreement?

A small firm might have the needed and desired capabilities and sufficient licensing. Local or not doesn’t matter for doing the audit’s work, which ordinarily uses electronic records, electronic communications, and electronic file-sharing. It might matter for licensing. Under many States’ public-accountancy statutes, the act of expressing an opinion on another person’s financial statements and delivering that report is the act that calls for a public-accountancy license. But in searching otherwise suitable firms, you need not limit a selection to firms with an office in the same State as the retirement plan’s administrator. Why? Some CPAs might maintain plenary licenses with two or more States. For example, a firm in Delaware might have a partner or principal who also maintains a license with another State, perhaps because she took the exams after completing her five school years there and applied while living in that State or applied in the State of her domicile. Further, some States’ laws allow some limited recognition of another State’s licensee. But reciprocity for audit or assurance services might be more limited than for other services. If a plan’s administrator acts as a fiduciary in selecting an independent qualified public accountant, the administrator’s duties of loyalty and prudence might suggest looking to a firm’s capabilities, perhaps especially if the circumstances already involve a known breach that suggests a control weakness.

To follow rocknrolls2’s observation: If an employer/sponsor/administrator uses, through its recordkeeper or third-party administrator, a plan-documents set, does the document-assembly software also generate the participant loan procedure and the form of promissory note? If so, will these documents be logically consistent, one to another? Or are there ways errors result even within one service provider’s set?

Consider also whether the spouses were married in, ever lived in, or otherwise invoked the law of a community-property nation or State.

Consider also the terms and conditions of the plan’s administrator’s service agreement with the recordkeeper. That a plan’s governing documents permit or even provide something does not by itself obligate a nonfiduciary service provider to provide a service to support doing that something.

Thank you!

Without considering any plan-document change (most of which can wait until 2025 or the later remedial-amendment time): For an individual-account retirement plan with a § 401(k) arrangement (with no auto-anything provision, and no need or intent to add any); allowing non-Roth and Roth elective deferrals, including age-based catch-up (with non-Roth not restrained for those who had wages more than $145,000 in the preceding year); immediate eligibility with no age, service, or other condition beyond employment; no matching contribution, and no nonelective contribution; distribution permitted on age 59½ or severance from employment; no involuntary distribution (except as IRC § 401(a)(9) requires); no risk of a coverage, nondiscrimination, or top-heavy failure; and with the calendar year as the employer’s tax year, all participants’ tax year, the plan year, and the IRC § 415 limitation year: Which SECURE 2022 changes must the plan sponsor put in operation starting with 2024? Which SECURE 2022 changes may the plan sponsor put in operation starting with 2024?

If you can post here the broker-dealer’s agreement—redacting names, and without breaking the participant’s, the plan administrator’s, and any trustee’s privacy, we might learn something about instructions the agreement permits the broker-dealer to rely on: from which person, and for which transaction.

Does the plan’s administrator have a written claims procedure? 29 C.F.R. § 2560.503-1(b) https://www.ecfr.gov/current/title-29/part-2560/section-2560.503-1#p-2560.503-1(b). If so, what does that procedure provide about whether a claim for a principal-residence loan requires evidence beyond the participant’s statements on the claim form? If the written procedure grants the administrator discretion about whether to require or excuse supporting evidence, what has the administrator done regarding similarly situated claimants? If the claims procedure calls for evidence beyond the claimant’s statement that the participant loan is used to buy the participant’s principal residence, consider these points from the Treasury department’s rule: “The tracing rules established under section 163(h)(3)(B) apply in determining whether a loan is treated as for the acquisition of a principal residence in order to qualify as a principal residence plan loan.” 26 C.F.R. § 1.72(p)-1/Q&A-7 https://www.ecfr.gov/current/title-26/chapter-I/subchapter-A/part-1/subject-group-ECFR807fc2326e73cb3/section-1.72(p)-1. I.R.C. (26 U.S.C.) § 163(h)(3)(B)(i): “The term ‘acquisition indebtedness’ means any indebtedness which— (I) is incurred in acquiring, constructing, or substantially improving any qualified residence of the taxpayer, and (II) is secured by such residence. Such term also includes any indebtedness secured by such residence resulting from the refinancing of indebtedness meeting the requirements of the preceding sentence (or this sentence); but only to the extent the amount of the indebtedness resulting from such refinancing does not exceed the amount of the refinanced indebtedness.” http://uscode.house.gov/view.xhtml?req=(title:26%20section:163%20edition:prelim)%20OR%20(granuleid:USC-prelim-title26-section163)&f=treesort&edition=prelim&num=0&jumpTo=true. And back to the § 72(p) rule: “[A] loan from a qualified employer plan used to repay a loan from a third party will qualify as a principal residence plan loan if the plan loan qualifies as a principal residence plan loan without regard to the loan from the third party. (b) Example. The following example illustrates the rules in paragraph (a) of this Q&A–8 and is based upon the assumptions described in the introductory text of this section: Example. (i) On July 1, 2003, a participant requests a $50,000 plan loan to be repaid in level monthly installments over 15 years. On August 1, 2003, the participant acquires a principal residence and pays a portion of the purchase price with a $50,000 bank loan. On September 1, 2003, the plan loans $50,000 to the participant, which the participant uses to pay the bank loan. (ii) Because the plan loan satisfies the requirements to qualify as a principal residence plan loan (taking into account the tracing rules of section 163(h)(3)(B)), the plan loan qualifies for the exception in section 72(p)(2)(B)(ii). 26 C.F.R. § 1.72(p)-1/Q&A-8 https://www.ecfr.gov/current/title-26/chapter-I/subchapter-A/part-1/subject-group-ECFR807fc2326e73cb3/section-1.72(p)-1.

CKocher, I don't here express a view about whether an employer/administrator selects between Form 5500-EZ and Form 5500-SF based on facts during the plan year, as of the end of the plan year, or when the administrator files its report. Several interpretations are plausible, and I ask what our BenefitsLink neighbors think.

About whether and when a plan might be no longer a one-participant plan as Form 5500 instructions use that term, an employer/administrator might consider not only when a marriage ended but also whether a former spouse who had been a community-property or equitable owner during the marriage became a title-holding partner. For example, if a divorce’s settlement agreement provides a former spouse an interest in a partnership or an interest in a limited-liability company treated as a partnership, that interest could make the former spouse a partner. See 29 C.F.R. § 2510.3-3(c)(2) https://www.ecfr.gov/current/title-29/part-2510/section-2510.3-3#p-2510.3-3(c)(2). BenefitsLink mavens, if a former spouse gets no partnership interest, how would you advise a plan’s administrator to report this situation: The divorce decree is made after the end of the to-be-reported-on plan year but became legally effective months before the plan’s administrator completes its Form 5500 report on that plan year. Would you: Report as a one-participant plan (because those were the facts on the last day of the plan year)? Report as an ERISA-governed plan (because the Form 5500 instructions defining a one-participant plan speak in present tense—“covers”—and when the administrator makes its report the plan covers a participant who is neither a partner nor a partner’s spouse)?