Peter Gulia

Even presuming consistency rules for all an employer’s employee-benefit plans (of those that seek to meet a tax law condition that refers to Internal Revenue Code of 1986 § 414(q)): The Treasury department’s temporary rule (adopted February 19, 1988, and last amended June 27, 1994) interprets § 414(q) as it was in effect for 1996 and earlier years. Under that rule, § 414(q)’s definition for a highly-compensated employee applies if another Internal Revenue Code section refers to § 414(q). 26 C.F.R. § 1.414(q)-1T https://www.ecfr.gov/current/title-26/chapter-I/subchapter-A/part-1/subject-group-ECFR686e4ad80b3ad70/section-1.414(q)-1T About health, other welfare, and fringe-benefit plans, that § 414(q) temporary rule lists sections 89, 106, 117(d), 125, 129, 132, 274, 423(b), 501(c)(17)-(18), and 505. Section 223—Health Savings Accounts states no reference to § 414(q). Does some coverage or nondiscrimination rule apply intermediately or indirectly?

I have never seen the Labor department question a change from cash-receipts-and-disbursements accounting to accrual accounting (if the Form 5500 report shows an adjustment). Many small plans, especially those with no TPA, begin reporting on cash accounting. Among several reasons, a plan’s administrator might adopt, without editing, the information a recordkeeper furnished. A recordkeeper might lack enough information to form an accrual entry. And even if that information is available to a recordkeeper, many prefer to avoid anything that could suggest even a slight discretion. When a small plan becomes one for which the plan’s administrator must engage an independent qualified public accountant, an IQPA usually persuades the administrator to change to accrual accounting.

I do not know anything that would help you answer your question. But the circumstances you describe suggest another question: Does an actuary assume a pension plan’s current provisions and an absence of change in the provisions (until the plan’s sponsor or administrator instructs different assumptions)? Or does an actuary’s professional conduct require her to make assumptions about a plan’s likely future provisions, even if one’s client has furnished no such instruction or guidance?

When a recordkeeper's customer-service person does not cite an authority, my experience suggests there is none. And even if there might be some Internal Revenue Service notice or announcement (or even a Revenue Ruling), only the Treasury department's rule binds a taxpayer.

Thank you for sharing Chief Judge Hall’s opinion. On your who’s-responsible question, there were plenty of people with plenty of opportunities to avoid unwelcome results. Among those, the multiemployer plan’s sponsor—the joint board of trustees, which decides the plan’s provisions arguably as a nonfiduciary creator, could have much more carefully stated the plan’s provisions. It is unclear (some might say doubtful) whether the plan provisions Judge Hall found were the provisions the plan’s sponsor intended. Further, even if one assumes only Judge Hall’s fact findings (which are incomplete), other possible interpretations of the plan (and of the plan’s application to each of the disputed beneficiary designations) are at least permissible and might be persuasive. The opinion suggests little or nothing about whether the plan’s administrator breached its responsibility because the opinion describes no analysis on such an issue. Further, it seems likely none of the interpleaded claimants presented a fiduciary-breach counterclaim or crossclaim. If there is an appeal, the appeals judges should defer to the trial judge’s findings of fact (unless clearly wrong), and might defer to the trial judge’s interpretations of the governing plan document, the summary plan description, and the plan’s form for making a beneficiary designation. If so, there might be little or nothing left in public law issues on which appeals judges would do a fresh analysis.

If, for a plan’s cash-or-deferred arrangement, the plan does not provide an automatic-contribution arrangement, doesn’t the absence of a participant’s affirmative election to defer mean she elects “cash” compensation (that is, no § 401(k) elective deferral)? See 26 C.F.R. § 1.401(k)-1(a)(3)(ii) (explaining that the absence of an affirmative election has a default consequence), https://www.ecfr.gov/current/title-26/chapter-I/subchapter-A/part-1/subject-group-ECFR6f8c3724b50e44d/section-1.401(k)-1#p-1.401(k)-1(a)(3)(ii). However, the plan’s administrator might evaluate whether an employee received the plan’s governing document, a summary plan description, or some other notice of her opportunity to elect for or against § 401(k) elective deferrals. If an employee is or was a minor, an administrator might evaluate whether notice to the minor’s natural guardian or conservator (a parent) is or was notice to the minor. That a plan’s administrator might have breached a fiduciary responsibility by failing to deliver a summary plan description does not by itself mean the plan or its cash-or-deferred arrangement fails to tax-qualify under Internal Revenue Code § 401(a)-(k).

The discussion has turned from what does or doesn’t belong in a W-2 wage report to whether compensation, for one or more retirement plan purposes, for a year might include something paid after that year. And that might relate to Belgarath’s originating post about a client’s request for a projection illustrated on an assumption that 2022 compensation will include a bonus paid after 2022 ends. C.B. Zeller, of course you’re right that you didn’t suggest anything about how 26 C.F.R. § 1.415(c)-2(e)(2) (if relevant at all) or a plan’s definition or provision based on that rule might apply. I wonder whether a plan’s administrator might interpret “first few weeks” to count a bonus that relates to 2022 services and is paid or delivered as late as by March 15? And you ask a smart question about some practical consequences if an IRS employee might consider such an interpretation as allowing more than the Treasury department intends the tax-law rule to allow.

The word "few" is indefinite. Might a plan's administrator interpret "first few weeks" to allow up to two and a half months or by March 15?

If one considers the rule C.B. Zeller mentions, the plan's administration would read carefully the plan's governing document to discern whether the plan's text supports such a measure of compensation.

Does the plan design austin3515 describes work if the matching contribution is 500% of the up to 6% in elective deferrals? (Assume the plan applies 401(a)(17) and 415(c) limits.) (Assume the resulting matching contributions are less than 25% of the participants' compensation.)

For an ERISA-governed plan, “the documents and instruments governing the plan” (ERISA § 404(a)(1)(D)) provide who may be a participant’s beneficiary and which beneficiary designations are or are not recognized. In my experience, a typical plan does not preclude naming an alien, even a nonresident alien. Different Federal income tax withholding regulations, rates, procedures, and forms apply for a plan’s distribution to a non-US person. Among other points, the payer would require a distributee to certify an Individual Taxpayer Identification Number (ITIN). Those 9-digit numbers are issued by the Internal Revenue Service. https://www.irs.gov/forms-pubs/about-form-w-8 https://www.irs.gov/individuals/individual-taxpayer-identification-number

To tax-qualify under Internal Revenue Code § 401(a) and for other U.S. tax law reasons, the plan’s trustee should be a U.S. person. If the plan were ERISA-governed, a fiduciary would obey ERISA § 404(b)’s command to maintain the indicia of all plan assets within the jurisdiction of U.S. Federal courts. Even for a plan not ERISA-governed, a prudent trustee might maintain in the United States (and, preferably, in the State in which the plan’s trust has its situs) the record of the UK company’s securities the plan’s trust owns. That includes a certificate (if there is one). If there is no established and efficient market for the securities the plan’s trust invests in, consider what valuation expert the plan’s administrator or trustee would engage to set—at least yearly, and whenever needed midyear to allow a transaction—an imaginary fair-market value for the securities of the UK company.

K-t-F, are you asking about whether a retirement plan (or some other employee-benefit plan) may invest?

The Form W-2 Instructions tell a payer to report in box 1: “Total wages, bonuses (including signing bonuses), prizes, and awards paid to employees during the year.” A related instruction states: “Calendar year basis. The entries on Form W-2 must be based on wages paid during the calendar year. . . . . For example, if the employee worked from December 18, 2022, through December 31, 2022, and the wages for that period were paid on January 3, 2023, include those wages on the 2023 Form W-2.” https://www.irs.gov/forms-pubs/about-form-w-2 (emphasis added). A retirement plan’s administrator (not a TPA) might want its lawyer’s advice about whether the administrator must not rely on an employer’s W-2 reports if the administrator knows those reports are incorrect.

My question was completely about my intellectual curiosity (recognizing my lack of experience with corrections). Thank you for the education. And in a fog yesterday, I missed that one wouldn’t assert the second plan as the source of the merged plan’s provision for nonelective contributions without reporting on that second plan.

If by establishing a § 125 cafeteria plan, wages that previously were not excluded from FICA taxes become excluded from wages for those taxes, would the employer’s first-year savings on the employer’s portion of FICA taxes cover the service provider’s $400 fee if those wages are at least $5,229 [$5,229 x 7.65% = $400.0185]?

If one does it as Bill Presson describes, is it feasible for the merger of the second plan with and into the first plan to happen immediately after creation of the second plan AND retroactive to January 1, 2022? If so, could one do all Form 5500 reporting solely on plan 001?

Roycal, thank you for describing your view. WCC, thank you for the helpful information about Empower. While that regime is not ideal (for several reasons), perhaps it lessens a participant’s discomfort about being pushed to slightly favor a beneficiary over others. C.B. Zeller, thank you for your explanation of a business point. And from my experiences with big recordkeepers, I know many business decisions have several layers of complexity about how much customer-friendliness customers will pay for. fmsinc, thank you for your article about strategies and steps a domestic-relations lawyer might consider. And your observations remind me of an astute lawyer’s saying I heard: “Lots of things can happen in court, all of them bad.” Here’s a story about seeing that someone must get an odd cent or dollar (or two, or three): I administer a decedent’s estate. It has four testamentary beneficiaries, with equal shares. So far, every interim distribution has been in four exactly equal shares. But many of the estate’s Form 1041 amounts (passed through into beneficiaries’ K-1 reports) were not neatly divisible by four. For those amounts, the tax-preparation software the certified public accountants use applies a series of randomizer formulas to allocate an odd amount so the sum of the four K-1 reports on an item equals the 1041 amount. Further, the formulas are designed so a beneficiary stuck with an extra dollar or two for an earlier item is not allocated an odd share for a later item, so the sums of K-1 pass-through income items are equal (or as nearly equal as possible) among the beneficiaries. What I’ve liked is that the software does this with no exercise of my discretion. And I could explain to the beneficiaries how none had been unfairly advantaged or disadvantaged.

david rigby, thank you for this helpful information. Perhaps a system that allows a decimal place diminishes the effects of a one-third or one-sixth problem.

Bob the Swimmer, thank you for sharing your experience. I have counseled recordkeepers that allow any set of percentages or fractions coming in, including those that don't add up to 100. Instead, the plan's governing documents provide proportional adjustment rules that turn whatever the participant stated into something the computers will process.

fmsinc, to answer your last question, the Eighth Circuit’s recent decision (following an earlier Eighth Circuit decision) does not preempt State-law remedies that do not involve the retirement plan or its fiduciaries. As I explained in another BenefitsLink discussion, decisions for the Third, Fourth, Sixth, Eighth, and Eleventh Circuits interpret ERISA as not precluding a remedy that does not involve the plan or any fiduciary of it, and does not frustrate a surviving spouse’s right. In States’ courts, often the court does not even consider ERISA, especially regarding an already-distributed benefit. (If no plan fiduciary is in the State-law remedies litigation, only the plan’s distributee seeking to keep the benefit has a reason to raise the issue. And even when that person’s lawyer understands the argument, many abandon it.) States’ courts use remedies to move property from the plan’s distributee to whoever is the rightful taker under State law. You are right to fear that a plan’s distributee might dissipate the property paid or delivered to that distributee. That’s why another claimant should pursue remedies promptly and vigorously.

Lou S., thank you for your information about Fidelity BenefitsLink neighbors, do we have clues about Empower, TIAA-CREF, Vanguard, Alight, Voya, Principal, Merrill Lynch/Bank of America, T. Rowe Price, . . . ?

Yes, the participant by her inattention failed to change her beneficiary designation. But observe that the Honeywell 401(k) Plan incurred $$$ in attorneys’ fees and related expenses for two layers of Federal courts’ proceedings, with briefings and arguments on several issues. (The plan incurred these expenses, even if the employer volunteered to advance or outright pay them.) Although the court found the plan’s administrator blameless, it’s unlikely the employer/administrator will recover from the participant’s personal representative or any litigant any portion of the expenses. If the expenses were or are charged against the plan’s assets, participants share, however indirectly, in the expenses.

pmacduff, thank you for your helpful observation. Some participants tolerate the constraint and adapt to it in a way you describe. But at least some are offended that a Procrustean computer system requires a participant to specify a difference (however slight in percentage and amount) that pushes one to express (however subtly) a difference the maker does not intend. That offense-taking can happen even when the participant knows the favored beneficiary will adjust the difference by giving money to the others. And even when there is no discord, many people care about the symbolism. Also, some plans’ sponsors and administrators care about this point for other purposes. One of them is decreasing the number of participants who made no (valid) beneficiary designation. Many participants don’t read carefully the beneficiary-designation form. Some miss or ignore the instruction about whole percentages. Some don’t respond to a notice, or even repeated notices, that an attempted beneficiary designation was not processed. (Without those failures, the court case described above would not have happened.) It’s not enough to observe that people should be more careful; human behavior has weaknesses. (I know I don’t diligently do everything I should.) And some fiduciaries seek to make things easier for participants. An absence of a participant-specified designation burdens the plan’s administration because it calls for interpreting and applying on particular facts (which take time and effort to get) the plan’s default beneficiary designation. What if the plan’s default is to pay the personal representative of the participant’s estate, but no executor or other representative was appointed (often because the individual had no probate assets)? Or what if the plan’s default, after confirming the absence of a surviving spouse, is the participant’s children? How does the retirement plan’s administrator know how many children the participant had? If no one submits a claim, must the administrator search for the children? Or if someone submits a claim and states the participant had two children, how does the administrator confirm that there are no more than two? Dealing with these and other default situations is work. The expenses of applying a plan’s default-beneficiary provision can be significant if an administrator or a service provider administers a plan or plans with hundreds or tens of thousands of participants. I’m hoping to discern a range of experiences from plans that use a range of service providers. Among them, how about: Fidelity, Empower, TIAA-CREF, Vanguard, Alight, Voya, Principal, Merrill Lynch/Bank of America, T. Rowe Price, . . . ? I’m guessing some have figured out ways to manage and adjust a whole-percentages constraint.

Bird, thanks for adding a name to our survey. Does anyone know what Fidelity does? A whole-percentages constraint, and how a service provider and the plan's administrator deal with it, matters because a parent with three children might be offended if she must specify one of them to get 34 rather than 33 percent. In my experience, many parents strive to avoid favoring any child over the others.