Luke Bailey

Rick S. I think what they're saying is that (a) the $3,000 was an employer contribution, and (b) they withheld the appropriate taxes when they included it on your wife's W-2, so she is not out of pocket. Including the amount in her income while it was still subject to an SRF (which apparently is what occurred) seems inconsistent with Section 457(f), but (i) the IRS is unlikely to be put out by the fact that they got their money sooner (or, in this case, at all) rather than later, and (ii) assuming that the amount the employer withheld and paid to IRS in fact was at least as much as the increase in your and your wife's tax liability, the argument that she was damaged by the employer's conduct is complicated at best. Good luck.

RosemaryCR raises a good point and demonstrates how important the facts and circumstances analysis will be. I do believe, however, that the facts and circumstances could show in specific cases that the intent was for the employer to pay the fee, so it is reimbursing the plan, with the reimbursement being allocated back to individual accounts in the same amount as the expense was borne, not making a contribution.

I agree with all of these very helpful comments as to (a) what IRS's position is on the issue, and specifically that the conclusion of Rev. Rul. 2007-43 has found its way virtually verbatim into the Service Center compliance check letters that it sends out, and (b) IRS's most likely rationale for the position (i.e., that it's hard to tell whether someone who is classified as a "quit" was really voluntary if the company was going through a RIF). Having said that, and assuming you could determine with reasonable certainty that some employees had quit or were discharged for cause, I don't see the basis for concluding that those folks were "affected" by the partial termination. Note that if I understand you correctly, Pension Panda, you were able to demonstrate to IRS that there was such a large number of for cause terminations that the employer did not have a partial termination. I will probably not be able to do that in the case I have, since as I understand it the employer in quesstion did do a RIF. They 100% vested everyone in the RIF, I believe, so all I want to do is not have to 100% vest retroactively the folks who were not in the RIF, but who quit or were discharged for cause. Thanks to all.

Rev. Rul. 2007-43 seems to me to provide some reasonable clarity for determining when a partial termination has occurred. You divide the number of "employer initiated" active participant terminations during the period in question (e.g., plan year) by the number of active participants you had at the beginning of the plan year, and, depending on how much the resulting percentage is over or under 20%, you determine whether you had a partial termination. 20% or over, you probably did, under 20%, you may not have, and other facts and circumstances can also be brought into the analysis if you're close. All this seems reasonable, to me, and as far as I can tell is rationally related to the case law. But Rev. Rul. 2007-43 also seems to assume, as far as I can tell without any basis in regulations or case law, and possibly, in my opinion, contrary to the plain meaning of the statute (IRC sec. 411(d)(3)), to say that if you determine you had a partial termination, then everyone who terminated without full vesting during the period in question, including voluntary terminations, is deemed to have been "affected" by the partial termination and therefore is required to be fully vested. Below is the relevant paragraph from Rev. Rul. 2077-43: "If a partial termination occurs on account of turnover during an applicable period, all participating employees who had a severance from employment during the period must be fully vested in their accrued benefits, to the extent funded on that date, or in the amounts credited to their accounts." Does anyone besides me think that this may be baseless and lacking common sense? Am I missing something that makes Rev. Rul. 2007-43's conclusion regarding the definition of "affected participant" correct?

I can't cite precedent on this, but I would think that it depends on the documents and arrangement. Generally, if the company meant to pay, and they were paid by plan by accident, you could probably characterize the reimbursement as that, and not a contribution. But generally speaking, if the plan intended to pay, and now the employer is saying, "Gee, I wish I had paid it directly," it's probably a contribution. Depends on what the plan and employer intended at time of payments, in other words.

I think the question boils down to whether the employee really is, now, an independent contractor. Usually, if the individual has the same job, working hours, etc., and the former employer is the only client and the alleged independent contractor has no capital invested in tools, equipment, etc. (and working out of home probably doesn't count), then they are still an employee. But it depends on facts and circumstances and there are no bright lines. Make sure at least that the plan excludes employees who are treated by employer as contractors, even if reclassified by state or feds. So-called "Microsoft" provision after famous case involving Microsoft from late 90's.

Calexbraska, to answer the question in your third paragraph, if the VCP submission is not successful (and as PensionPro points out, it might succeed, depending on many factors), you would not be able to limit the correction to the statute of limitations period (the routine audit period is irrelevant) if you wanted to maintain the plan's qualification. In the words of Frank Sinatra, it's "All, or nothing at all."

My guess is the reference to "personal 415 limit" is an unintentional red herring, the questioner thinking that perhaps the daughter could defer $0 because she had no valid compensation under the plan. But as C. B. Zeller points out, the problem seems to be simply failing to follow the plan document, which in this case can be fixed in EPCRS self-correction as described.

To avoid uncertainty, plan documents should specifically include a slayer provision, either their own or incorporate state law, as msmith points out. If, as often seems to be the case, the plan document does not include a slayer statute provision, then my impression is that, contrary to almost every other area where ERISA preemption has been implicated, the federal courts have by and large found a way somehow to not preempt state slayer statutes. As PensionPro points out, the most recent example of this was in the 7th Circuit. I vaguely recall that there was at least one case years ago that went the other way, maybe at the district court level, but most seem to keep the benefit out of the hands of the wrongdoer. You should probably interplead to get a court to decide.

jkharvey, from time to time I've had reason to think a little about this or at least similar issues. I don't think you are going to be able to find anything in ERISA or the Code specifically on point. In asset sale transactions, the buyer usually wants the seller to terminate the plan, but occasionally the buyer will want the seller's plan and the deal documents will say that the seller will permit assumption of the buyer's plan, and the plan may be amended before the assumption to make clear that this is permissible. Although in that context, allowing the seller's plan to be assumed by buyer and merged into buyer's existing plan seems to have a more obvious business purpose, it is still the case that, as in your circumstances (which, as Bird points out, you haven't yet fully explained), the plan sponsor is essentially permitting another company to become the sponsor, in effect transferring sponsorship. Again, there may be case law where this has been an issue, but I am not aware of any provision of ERISA or the Code that would bear directly on this. IRC sec. 414(l), for example, merely provides the standards for what happens in a plan merger, not when you can do a plan merger. Your risk would be that an employee who wants a distribution based on his/her termination of employment sues to get it, saying that the plan document's rules for distribution on termination had not been followed, or that the buyer commits a fiduciary breach after the transfer of sponsorship/merger and the employees who lose money because of the breach sue saying that the seller's allowing the buyer to take over the plan was a failure to follow plan documents and/or a breach of fiduciary duty.

A few points. First, 415 Limit, take note that RatherBeGolfing and Tom Poje have correctly and importantly corrected me that since, in your case, the excess deferrals are attributable to contributions to a single plan, they can be distributed after April 15, and should be, per EPCRS self-correction. However, the double taxation penalty on the participant will not be avoided. Second, I agree completely with Tom Poje's statement that that if the Roth excess deferrals occurred under a single plan, they need to be segregated into a separate account with tax characteristics unique to excess deferrals attributable to Roth contributions. The tax treatment of that segregated account is unique, however, in that (a) the amount is treated as pre-tax, and thus the original contribution amount and earnings are includible in the participant's gross income when distributed following a distributable event, but (b) they do not qualify for rollover under Treas. reg. 1.402(g)-1(e)(8)(iv). I'm still wondering, in a case where the excess deferral is caused by pre-tax and Roth contributions to the plans of two different employers, and the IRS aggregates the deferral amounts on the participant's w-2's and determines the excess, whether the IRS is going to adjust the participant's 1040 based on the assumption that the excess was attributable all to the pre-tax, all to the Roth, or some combination. The employer's accounting for the Roth contributions would seem to need to take that into account.

The taxable year referred to in Section 402(g) is the taxable year of the participant (invariably, the calendar year), not the plan year, where the plan year is other than the calendar year. I don't believe there is any exception to the April 15 deadline in IRC sec. 402(g)(2)(A)(ii), so the excess deferral would need to stay in the plan until the individual otherwise has a distributable event. When the excess deferral is eventually distributed, then to the extent it is attributable to the pre-tax elective deferrals, it will be taxed again. The second tax can be postponed (but may be greater, because of earnings) if the amount is rolled over. To the extent the excess deferral, when eventually distributed, is attributable to the Roth contributions, it will not be eligible for rollover and will be taxed again, even though contributed as Roth. See Code sec. 402A(d)(3). I am unaware of any guidance on how to determine, in a case such as you have posited, whether the $4,500 is attributable to the Roth or pre-tax deferrals. You may be able to adopt a plan rule that specifies this. Maybe someone else has experience with this issue, but it would seem to me to depend, at least in terms of fairness, on how the IRS Service Center treats this. If when the Service Center gets the corrected W-2 showing total deferrals of $22,500 and Roth of $4,500 the Service Center increases the individual's reported income by only $4,500, they are in essence treating the Roth as the excess. On the other hand, if they include a larger amount in the individual's income, they will be treating all or part of the pre-tax as part of the excess, e.g. if the IRS Service Center treats $9,000 as taxable in making an adjustment to the individual's return, it will be treating the entire $4,500 excess as pre-tax. Maybe this is controllable by the employer in the way reported on the W-2 or W-2(c) and covered in the W-2 instructions, but I think the W-2 instructions avoid the issue. The have an example on page 19 of an excess deferral that is attributable to both Roth and non-Roth, but they assume that the excess is timely distributed.

For reinforcement of Belgarath's answer, notice how the following paragraphs from the IRS's rollover notices imply the ability to borrow from a rollover account in an employer plan, whether the rollover is non-Roth or Roth. First, from the non-Roth version of the notice (emphasis supplied): Where may I roll over the payment? You may roll over the payment to either an IRA (an individual retirement account or individual retirement annuity) or an employer plan (a tax-qualified plan, section 403(b) plan, or governmental section 457(b) plan) that will accept the rollover. The rules of the IRA or employer plan that holds the rollover will determine your investment options, fees, and rights to payment from the IRA or employer plan (for example, no spousal consent rules apply to IRAs and IRAs may not provide loans). Further, the amount rolled over will become subject to the tax rules that apply to the IRA or employer plan. Next, from the Roth version of the Notice (emphasis supplied): Where may I roll over the payment? You may roll over the payment to either a Roth IRA (a Roth individual retirement account or Roth individual retirement annuity) or a designated Roth account in an employer plan (a tax-qualified plan or section 403(b) plan) that will accept the rollover. The rules of the Roth IRA or employer plan that holds the rollover will determine your investment options, fees, and rights to payment from the Roth IRA or employer plan (for example, no spousal consent rules apply to Roth IRAs and Roth IRAs may not provide loans).

Even though the IRS FAQs, or to the best of my recollection any published guidance, do not directly address this issue, if I were confident that they really had committed dishonesty (e.g., that would support termination for cause), I would be tempted to treat them as if they had quit, i.e., not count toward the partial termination 20% threshold. But again, very dependent on "facts and circumstances." The rule ultimately is a moral/fairness one, i.e., did the employee, or the employer, deprive the employee of the opportunity to continue to vest. If the former, do not count the individual toward the threshold, if the latter, do count him/her. In this circumstance it would appear the employee in essence quit. But again, you'd want to be really confident of the basis and fairness of the dishonesty determination.

OK. Thanks, John Feldt. If no 410(d) election, then 4(b)(2) of ERISA kicks you out of ERISA.

Wouldn't ERISA preempt the application of state law?

You vest at NRA, right? Plus the plan may not provide for actuarial increases or even accruals if you retire after your NRA. I'm assuming it's a DB plan.

I thought it was an interesting question. As with many questions, it requires some clarification.

The rules seem to be different depending on whether it's non-Roth employee after-tax or Roth money. The Code, regs, and IRS are clear, I think, that you can't "source" a loan to non-Roth after-tax contributions. Requiring non-Roth after-tax to come out pro=rata across all participant accounts for any type of a distribution (including at least by implication a deemed distribution on loan default) was introduced into the Code in 1981 or 1986, or maybe 1996. I'd have to check. But this was a specific Code change made by Congress for some reason and IRS of course follows this. When Congress enacted and subsequently amended the Roth deferral and in-Plan Roth rollover rules, they specifically said that all Roth contributions must be in a separate "Designated Roth Account" and did not include a rule that distributions from a plan containing both Roth and non-Roth have to be treated as pro-rata, but rather left the Code open to the interpretation that, unlike the situation with employee after-tax contributions, a participant could choose to have distributions come from (or to source all or part of a loan to) his/her Designated Roth Account, if the plan document or loan policy permits this. (Of course many don't address, and the recordkeeper in such a case will typically source distributions and loans pro-rata across the Roth and non-Roth accumulations, just like non-Roth employee after-tax.) If, as the original question implies, the participant's basis was in non-Roth employee after-tax contributions, then I think the defaulted amount should probably be treated as consisting of after- and pre-tax amounts pro rata. On the other hand, if the loan was properly sourced under the plan document and loan form to the participant's Roth account, then you would follow Roth rules. If a distribution from the Roth account would not have been qualified, then a portion would be taxable, a portion would be recovery of basis, just like any other distribution from the participant's designated Roth account.

I think your intuition, Mr. Bagwell, that a portion should have been taxable is correct and that (if I understand your correction correctly ) they seem to be confusing the amount that is distributable with the source of the distribution. See Treas. reg. 1.402A1, Q&A-8.

I doubt there is going to be any clear answer to this question. I don't deal much with church plans, but I think it's the identical issue as you have for governmental plans. Certainly for an employee that has already reached the plan's pre-amendment NRA, and so is "vested" under the pre-ERISA rules, you would have the argument that you are impermissibly "unvesting" the employee if you raise his or her retirement age. But for folks who have not yet reached NRA, and so are not vested, arguably you can do whatever you want. That's not a comfortable answer based on common notions of fairness. A careful reading of the plan may indicate that notwithstanding that the law would not require vesting until NRA, the plan actually vested benefits at an earlier age, and so moving the retirement age for vested accrued benefits was, again, a violation of the vesting rules. You can also argue that moving the retirement age as to accrued benefits would violate the pre-ERISA "written plan" requirement of 1.401-1, but that's a tough argument since by its nature an amendment changes the written plan and retroactive negative changes are what 411(d)(6) (from which the plan was exempted) was designed to stop. I think if you really want to know the answer you're going to have to do a lot of research into pre-ERISA rulings and cases.

I'm not sure I understand the facts. A owns the management company and owns B. What do you mean, Calexbraxa, when you say the management company is going to be "replaced?" Sold? I'm not sure I see the transaction. Anyway, I think NQDC spinoffs are fairly common, and certainly we have done them. Since NQDC plans are just promises to pay, what you are doing when you "spin off" part of an NQDC plan to another company, e.g. to the acquirer of a division of the company that has the NQDC plan, is transferring the obligation to pay to the sponsor of the new spun off plan. The biggest issue, I think, is whether the employees covered by the plan can object to having the obligation to pay them transferred to a new company. This depends in part on what the plan says, and certainly if the source of the obligation was executive deferrals and the transferee entity was weaker financially than the transferor, you could have issues, although you're sort of in no-man's land, since ERISA preempts state law in the area, and there is no substantial body of rules under ERISA for NQDC plans. But from the standpoint of the employees' tax issues, they are just getting a new unfunded, unsecured promise to pay money in the future, from someone else, to replace their old unfunded, unsecured promise to pay money in the future. That should not be taxable in and of itself, although you could stray into a minefield if you attempt to give employees' choices. Obviously, the transferee is not going to take on the obligation without getting compensated for doing that. Usually these transactions occur in an M&A context, so the purchase price of the subsidiary or division may be reduced to reflect the liability being taken on by the acquirer. If assets are transferred, e.g. a rabbi trust is divided, or life insurance policies are transferred, there are tax issues you have to work through. Actually, even if assets are not transferred there may be tax issues for the employer, depending on the transaction.

Maybe this is more like an agency arrangement than a loan? The employer receives the money that is going to go to the plan participant from the plan's trust, deposits the funds into its own general account, and pays the same amount of money to the participant on behalf of the plan. That practice seems to raise a silly tax issue (i.e., could an IRS agent having a really bad day allege that the participant did not receive the funds from the plan), but I'm not sure that would be a PT even without PTE 80-26. However, I wonder if the DOL could have concerns with this practice with respect to the requirement to hold plan assets in trust. Suppose it's not a nanosecond, but a day or two, and during that day or two the employer goes belly up? The participant does not receive the check, at least not for awhile as things are sorted out in the bankruptcy, and the plan definitely does not have the money. If this is common practice, I guess it is possible that the financial institutions that do this have thought it through, but I do find it a little puzzling. I would be interested to see whether the practice is the subject of any self-serving protective provisions in the financial institution documents. In the situation where the employer pays first, using its check, and then is reimbursed from the trust, you would have a clear PTE 80-26 loan, and there would also be no risk that the funds would be lost in a bankruptcy.

Jamin' JROD, it's not clear to me from the question who is going to own the 2%. If you tell me that, I might be able to respond to your question.

ESOP Guy, you are right about the direction of the loan in PTE 80-26. Rereading the question, I think you are probably also right that at least several of us misread it, but now I'm confused. (I guess I was confused before, but now am confused differently.) Maybe Belgarath can go through the sequence of events with more detail. The question says that the plan may not have a checking account, but then seems to have the plan writing a check. Also, the time period between each step could be important.