Luke Bailey

Can't you read the reg in such a way that contributions for HCEs that may have outward appearance of SH based on plan wording and the way they are calculated are really not safe harbor contributions? 1.401(k)-3(b) and (c) refer only to making SH contributions for NHCEs. In fact I don't think the regs really ever define "safe harbor contributions," but rather describe contributions that, when made for NHCEs, "satisfy the safe harbor contribution requirement." The rules in regs and 2016-16 that say what you can't change refer to not changing the safe harbor contributions, but if the contributions for HCEs are not really safe harbor, no problem. This essentially says that what C.B. Zeller points out you can do if you're clever enough to draft it that way is, in fact, what you can do anyway/already. Will be interested in any authority otherwise.

Mike, I think I follow your math, but your first example shows up one of the problems with saying that the "period" referred to in 2202(b)(2)(C) is the 3/27/2020-12/31/2020 suspension period. I guess you're right that if that is the way it is to be taken, since the period of the suspension ends 12/31/2020, but the payments skipped during the suspension period don't restart until a year after they were suspended, you would need to return to the original payment amount as of 1/1/2021, then switch to the reamortized amount as of the first payment that was delayed by one year. An extra credit math problem to be sure. Would be much easier if the one-year period being referred to was the period by which the first payment was postponed, which is what I assumed in my original example. In that case, you would merely reamortize once, to take into account the interest for the delay in payment, when the loan restarted exactly one year after the first missed payment, and adding exactly one year to the period that the loan had left on it at the time of suspension. Is this not a possible interpretation?

So RBG, you are saying that when 2202(b)(2)(C) says you can lengthen the original repayment period by "the period described in subparagraph (A)" it is referring to the 3/27/2020 - 12/31/2020 suspension period, and not the 1-year period by which each payment may be delayed? I think the statutory language would support either approach, and the latter would sure be much simpler.

RatherBeGolfing, yes. There is a lot of flexibility here, but requires a lot of sophistication and handholding by/of employer to make happen for participant. Could wear out patience of many employers to provide loan relief at maximum level. But the beauty of the Notice 2005-92 approach is that the employer can choose what it wants to do on the range of the possible (from not doing the extension at all to probing the maximum extent of the relief legislation), and be OK as long as it is consistent and nondiscriminatory. Right. The notice draws a distinction between a "deemed loan," say to a current employee who revoked withholding for loan payments or otherwise has a default while in service, which would not, per IRS, qualify for the favorable KETRA (and likely CARES Act as well) tax treatment, and a "loan offset" distribution, say to an employee who is laid off and qualified for a distribution while a loan is defaulted, which would be treated favorably for tax purposes (at least for KETRA, probably also now for CARES).

OK. I'm pretty sure we're in agreement, but let me take it out a few more steps. If the 3/15 payment is made up, after 3/15, then the 3/30 payment (assuming semimonthly amortization) falls in the period. So a loan that would otherwise default based on a missed payment the date for which fell before the beginning of the COVID-19 period, if the delinquency is not cured, can be brought into the COVID-19 relief by making up (before the end of the cure period) all of the payments that were missed before the beginning of the cure period. Then all the payments after that get the one-year extension. Agreed?

I'm pretty confident you will find no statute or regulation on point, but I have a clear recollection that after the 1987 crash there was a handful of cases, say in the 1989-92 period, maybe some in 9th Circuit, that upheld retroactive valuations in non-daily valed, balance forward plans (which were probably in the majority then). I seem to recall that even amending a plan to add language like that which you are quoting, mariemonroe, after the crash, was upheld. As Larry indicated, courts found the plan's case is sympathetic. I don't recall similar cases after 2008, but that may be because most plans were daily valed by then and also the prior precedent on balance forward was settled.

By the way, if you go the route of saying that only the scheduled payment, not a permissible cure payment, is what has to fall within the CARES Act period, then it's a bright line, right? If the payment was due 3/15/2020, but not made, you would be just as ineligible for the one-year delay as if the missed payment was in fourth quarter of 2019, right? And do you see a distinction with respect to an employee who terminates employment at a time when the loan is current, but with a balance, and the plan's loan policy is to accelerate on termination if the remaining balance is not repaid within a short period, the end of that short period falls during the CARES Act period? If the loan is not repaid, the resulting distribution will be a loan offset that can be treated as a CARES Act distribution (assuming the participant qualifies as COVID-affected), and that treatment would be better in most cases than getting the one-year repayment extension, but assuming the employer wanted to adopt the CARES Act loan relief to the limit, the participant would probably have a choice, right? On the other hand, if the employer decides to only view regular payments under the loan's original amortization schedule as eligible for the CARES Act relief, it would appear to be able to do that if IRS applies same rules as did for KETRA.

Thanks, Mike. and thanks also for pointing out that my original post double-counted the delayed principal. Certainly needed to fix that!

OK. I see your point. CARES Act references IRC sec. 72(p)(2)(B) and (C). (B) does not seem relevant, but (C) probably slightly supports the argument that is the regular payment that must be missed in the relevant time frame, because being referred to in CARES Act. On the other hand, looking at 72(p)(2)(B) and (C) together, and along with the regs, I think one could easily justify that any payment within the cure period is a "payment due" for purposes of CARES. IRC sec. 72(p) and its regs are basically telling you the payments you need to make to avoid having to treat the loan as a distribution, and that scheme includes the cure period and a payment during the cure period that would be sufficient to avoid deeming of loan. Gee, they're giving $2 trillion plus away, you think they would cut a break to a guy or gal who might have cured his or her loan delinquency, but for getting hit by COVID. At no cost to Treasury. But we will see.

OK, thanks, Larry. I was not that deeply interested in the issue at time of KETRA, so was not aware of that. Perhaps they will do it same way this time. Pretty much everyone who would have a loan or need a distribution in the relevant time period will be negatively affected by COVID, so maybe not worth IRS's bothering about it.

From a policy perspective, someone who can't afford to repay his/her loan because of COVID-19 for a year is probably in the same place as someone who takes a new loan because of COVID-19, so if I were IRS, given that the language literally supports it and that the employer would have to implement and could put reasonable guard rails around it, I would probably permit it. Larry, the Notice that RatherBeGolfing pointed to (2005-92) specifically said that a loan offset distribution was a "distribution" and therefore got the seemingly identical KETRA tax benefit. Larry, I think the fact that the participant has to provide plan administrator with self-certification that the distribution meets the requirements implies a role for the employer to at least receive that certification and keep it on file. Besides, presumably there will be a code for 2020 1099-R's to show how treated and that is where rubber will hit road.

nerd-party-administrator: 1. The statute applies the extension to "any loan" outstanding on or after date of enactment, so basically you could take the $100k and have first payment a year later. 2. See answer to 1. Of course, the participant does need to be COVID-19 affected (as defined in statute) to get the extension.

Patricia, this could be something I have just run across, but I'm pretty sure "FICA Alternative" and "FICA Replacement" are referring to same thing. With very few exceptions, e.g. the student FICA exception, no employer of an employee subject to U.S. taxing jurisdiction is going to avoid FICA for that employee unless the employer and employee are governmental. The small exceptions that do exist, e.g. for student employees meeting certain requirements, do not require a retirement benefit to avoid FICA.

RatherBeGolfing, thanks. I just reviewed Notice 2005-92. So assuming IRS issues similar guidance for CARES and that CARES does not create substantive rights enforceable by the participant beyond the Code (both of which are safe assumptions, I think), the answers to my questions are: The CARES Act loan extension for qualifying participants would only apply if and to the extent that the plan sponsor/administrator chooses to apply it, A "deemed" distribution of a defaulted loan to a participant who is a COVID-19 affected participant, but who is not eligible for a distribution, e.g. because still employed, would not be a CARES Act distribution. A "loan offset" distribution to a participant with a defaulted loan who is entitled to receive a distribution would be a CARES Act distribution eligible for the CARES tax benefits (inclusion over 3-year period, no 10% premature distribution penalty, and rollover permissible through end of 3-year period). I do not see anything in Notice 2005-92 that addresses the issue of whether the payment that would need to be made by a COVID-19 affected participant to make current/avoid default on a loan that was delinquent, but that had not yet defaulted under the plan's cure period, could be extended under the CARES Act. Is that your reading as well? Is it obvious to you that the IRS would think that such a payment is just a loan payment, and so would potentially benefit from the extension if the plan sponsor/administrator chooses to treat it that way?

PNJ, you or someone else will need to educate me on how the concept of a FICA replacement plan is relevant for a nongovernmental plan?

Mike, I just used plug numbers. Did not bother to run an actual example, so the inferred interest rate was not intended. I was just trying to illustrate the concept of how each payment is delayed a year, then when the first new payment is due (a year later) you reamortize and tack a new year on to the end of the original (typically-five-year) period. You're right about the rest of it. Of course the missed principle payments are not really missed, just delayed. It was late last night. I should have proof-read. Editing now.

Patricia Neal Jensen, I think I agree with Peter that there is not going to be a clear answer until IRS provides guidance. Of course the CARES Act merely protects you on plan qualification-type issues (I say "-type") because the statutory language specifically references 403(b) and 457). Does not address FICA replacement, otherwise you would not have asked the question, of course. The 3121(b)(7)(F) regs say the determination of whether the required allocation amount is for a retirement-type benefit is based on facts and circumstances, and contains an example where the requisite amounts are allocated, but because they are available for distribution to participants at any age five years after being allocated, the plan is not a good FICA replacement plan because does not accumulate amounts for retirement. And Congress also did not give COVID-19 folks the option of taking now a one-time lump sum of some portion from SSA now, in exchange for an actuarial reduction of what would otherwise be their eventual SS benefit. So I would be cautious, although this is such a big potential gotcha that IRS should put out guidance on it quickly.

The CARES Act provision for COVID-19 distributions is different from the provision liberalizing the loan rules. The provision for distributions merely describes what a plan can do and not violate the qualification rules, as well as the tax treatment for the participant of Covid-19 distrributions. Clearly, the employer must implement the distribution provisions through an amendment or through a policy change eventually backfilled by an amendment in order for the distribution provisions to apply to a participant. However, the CARES Act loan provision regarding the 1-year payment extension (Section 2002(b)(2) of CARES Act) can be read as a command regarding the administration of a loan. It does not say the equivalent of, "Hey employer, if you want to change your loan policy, you can do the following without having a distribution under 72(p)," but rather it says, "(A) if the due date pursuant to subparagraph (B) or (C) of section 72(p)(2) of such Code for any repayment with respect to such loan occurs during the period beginning on the date of the enactment of this Act and ending on December 31, 2020, such due date shall be delayed for 1 year..." [emphasis supplied]. The problem is that loan payments aren't really due pursuant to the referenced subparagraphs of Section 72(p)(2), but rather pursuant to the loan's legal documentation, which presumably complies with 72(p). So you can interpret this (or not) as meaning that to the extent the loan documents incorporate 72(p), they are overridden as to the one-year extension. Moreover, the CARES Act loan provisions for the 1-year suspension say nothing about amending loan documents or policies, although I guess you could argue that Congress thought plan documents contained loan rules, which obviously is not typically the case, and therefore would need to be amended for the suspension. Another ambiguity, I think, is regarding what payments have a "due date." Suppose a participant who qualifies for the COVID provisions terminated a month or two before 3/27/2020 and under the terms of his or her participant loan, the remaining balance is accelerated and due as a lump sum 3/31/2020, or it will be distributed as an offset? Is the lump sum repayment suspended for a year? If it isn't and a loan offset distribution occurs, is that distribution a COVID-related distribution qualifying for the 3-year averaging, extended rollover, etc.? What if the employee has not terminated, but payment could not be made by withholding for some reason beginning in the fourth quarter of 2019, so that 3/31/2020 was the last day of his/her grace period. (Again, assume the participant does qualify as a COVID-19-affected individual.) If you don't buy my first argument that the requirement to bring the loan current as of 3/31/2020 is suspended for a year, is the resulting "deemed distribution" a COVID-19 distribution?

AMDG, Congress states that any suspended payment is due a year later, and also provides for an extension of the loan period by the length of the suspension, and for reamortization with accrued interest. For example, suppose a qualifying individual had taken out a loan on January 1, 2018 for $10,000, with monthly payments of $200, all of which were timely made through 2/28/2020. The loan was not a home loan,, so due to be paid off 12/31/2022. As of 3/31/2020, the remaining principal amount of the loan was $6,000, say, and a payment of $200 was due 3/31/2020. Assuming the employer was able to act fast enough to suspend the 3/31/2020 $200 payment, and provided the participant with the maximum suspension, you would accumulate interest on all of the missed payments from 3/31/2020 through 2/28/2021, say $600, add that to the principal as of 3/31/2020 of $6,000, and reamortize the resulting $6,600 over the period from 3/31/2021 through 12/31/2023. (Note: I have not taken the numbers in the example from an actual calculation, so the implied interest amounts may be inconsistent and/or unrealistic. Just trying to illustrate how the calculation would be done in concept).

The statute states the 180-day period "beginning on the date of enactment...," so 3/27 would be day 1, and 9/22 day 180. If it said 180 days "after" or "from" date of enactment, then I think it would be 9/23, because the convention (e.g. for financial instruments as built into the HP-12C and other financial calculators) to resolve any arguable ambiguity in those alternative phrasings is to exclude the first day of the period and include the last, so then you would be counting 3/24 as day 1 and 9/23 would be day 180. I guess given the strength of that convention the IRS could go with 9/23, but I agree with everyone else that plans should wait and see if IRS goes that we before assuming it will.

Apologize for tardy response. Catching up with BenefitsLink posts, Mark. Does your question go to the economic stress this puts on the employer as sponsor of an "unfunded" plan? The questions I have gotten from clients about 409A in the time of Covid-19 have all been related to contribution cessations and in-service distributions for participants who are still employed. Bottom line on pretty much all 409A issues right now, I think, is that neither Congress nor IRS has had the time or maybe interest to extend any special relief for 409A issues based on the Covid-19 emergency, even though many of the same issues as are applicable in the qualified plan area apply to 409A plans as well; the regulations are probably far more rigid than makes sense given the circumstances. But the individuals affected are a smaller, less distressed (on the whole) group, so no relief yet.

shERPA, good point. Right now there are not even any proposed regulations under 414(m)(5), however, so the IRS would need to argue only on the basis of the Code. Unlike 414(o), which is not effective without final regulations (414(o) was a grant of authority to IRS to write legislative regs), 414(m)(5) is in effect as a Code section, so IRS can enforce it. In a situation involving typical management services (e.g., a "sole proprietor" managing a business's employees as clearly its primary business activity), IRS probably could enforce the statute without much difficulty. In other situations, e.g. professional or other services provided to clients not involving management of employees of the business, where the "sole proprietor" has not a scintilla of ownership in the larger business, so that 414(m)(2)(A) does not apply (assuming professional services), or where there is an argument about what is the sole proprietor's "primary business," it might be harder for IRS to make its case. As you suggest, if there are multiple clients, 414(m)(5) would not apply, whereas it leased owner would, to the portion of the plan attributable to revenue from the "client" business. There is also just the fact that the IRS scrapped, and never returned to, the 414(m)(5) regs project, but was not willing to abandon the "leased owner" rule under 414(o), so I have thought that that said something about its enforcement priorities. Maybe some on this board have had experience with IRS applying 414(m)(5) in an exam. I have not.

I think this is just a reporting issue. The portion of what was thought to be a hardship earlier was actually a distribution of excess contributions. Make sure the two distributions are properly reported/coded on the 1099-R(s) for 2019. File a corrected 1099-R if necessary.

I think that, assuming (big assumption) that the CEO really is an independent contractor, the rule that would cause his/her plan for his/her sole proprietorship to not pass discrimination, assuming the CEO owned, including by attribution, 5% or more or the S Corp, would be only the "leased owner" rule of proposed reg. sec. 1.414(o)-1(b). This rule was proposed in 1987 and, if it were ever finalized in its current/original form, would have a retroactive effective date that goes back almost 40 years. It was the only one of the proposed 414(o) regs that the IRS did not withdraw 20 or so years ago, when it withdrew the rest of the proposed 414(m) and (o) regs, presumably because in the IRS's view the potential for abuse in this particular situation was so blatant. Obviously, in most (all?) of these situations, the independent contractor manager-owners could likely be recharacterized as employees, which would be the more direct path to disqualification, but the IRC ic vs. employee test is such a swamp, I think, that the IRS did not want to fight the battle on that piece of ground.

Degrand, maybe, and a good point. Probably could put it both places to be sure, but I'm glad you agree with basic notion that one could, if he or she wished, expand voting rights to include sales of stock. Thanks.