EBECatty

I'm generally aware of the IRS position on "linked" nonqualified plans where a formula benefit under one NQDC plan would offset another NQDC plan, i.e., that they generally view any formula under a 409A-covered plan that can offset or change the amount, time, or form of payments under another 409A-covered plan as impermissible. What about where one plan is (or both plans are) exempt from 409A? For example, an employee has a short-term deferral exempt from 409A that provides a lump sum of 10 times his current base salary of $250,000 if he is still working at age 65. The payment is reduced dollar-for-dollar by the amount of any other vested nonqualified plan benefits payable to the employee. Say the employee is age 64 when the employer gives the employee a fully vested 409A-covered plan that provides 10 annual installments of $225,000 each starting at age 65. The employee would receive a payment of $250,000 upon reaching age 65, then $225,000 a year for 10 years. I don't see this as falling under the IRS's prohibition on "linking" nonqualified deferred compensation plans. For example, Section XI.B of Notice 2010-6, generally addressing the issue, speaks only in terms of a "nonqualified deferred compensation plan" linked to "another" "nonqualified deferred compensation plan." There's only one 409A-covered "nonqualified deferred compensation plan" that is not impacted by another one. Alternatively, say you have the same lump sum agreement but it's offset by the value of any vested 409A-exempt stock options outstanding at age 65. The options can be exercised for 10 years following termination. In that case, neither one is subject to 409A, but the parties may arguably in a sense "defer" the lump sum by allowing the employer to give the employee options before reaching age 65 that the employee can exercise any time over the 10-year period. Apart from the linking, any argument that this is a change to the time/form of payment of the lump-sum benefit? Or an initial deferral of a short-term deferral? I don't see any explicit authority for this, except maybe general anti-abuse. Would appreciate any insight.

We had a similar situation a few weeks ago. We have had several VCP applications pending for over a year now, but submitted a late amender (failure to restate for EGTRRA and PPA) and had the signed compliance letter back in about a week. The still-pending applications are more individualized with non-safe-harbor correction requests. The late amender submission was on a Form 14568-B. I assumed that was the difference.

I think this is probably right, but in my experience the plan being terminated is usually a short-term deferral and therefore not "a plan that provides for the deferral of compensation within the meaning of section 457(f)." Maybe I'm not getting the more aggressive pitches where terminating a plan subject to 457(f) is proposed.

Thank you both. I saw both of those threads, and I guess am a little surprised there's not a clearer answer to what seems like a fairly common and unobjectionable scenario. XTitan, my impression from that thread was primarily that it was fine from a 409A/NQDC side (the 409A preamble seems directly on point), but I've also seen some commentary suggesting that even though it's permissible under 409A, it's still a violation of the contingent benefit rule on the 401(k) side. Luke, I had tried to shoehorn it into the same exception, and agree your example would work, but the only "limit" exceeded in my example is the percentage of compensation that can be matched, which I think is a stretch under the language of the NQDC exception in the contingent benefit rule.

I see a few related threads on the topic, but is there a clear answer on the following: In its 401(k), the employer matches 100% up to 4% of compensation. For people earning above the 401(a)(17), their match is limited by the 401(a)(17) cap at under 4% of their total compensation. So if they earn $400,000 and defer $19,000, their match is capped at $11,200 ($280,000 * 0.4) whereas a match based on 4% of their total compensation would have been $16,000 ($400,000 * 0.4). If the "missed" amount due solely to reaching the 401(a)(17) limit ($4,800 above) is either paid in cash or is contributed (as an employer nonelective contribution) to a nonqualified plan, does that violate the contingent benefit rule?

Agreed, but am more concerned about the 457(f) impact on what seems to be a vesting event followed by a walk right. The concept was presented to me as still intending to defer tax until payment under (or outside of, to be more precise) 457(f).

In what would otherwise be a clear short-term deferral plan under 409A and 457(f), is there any exception from 409A and 457(f) with a vesting/payment trigger based on a voluntary termination (for any reason, not just good reason) at any time within one year following a change in control? The identical form (one lump sum within 30 days) and amount (fixed dollar amount) of payment is available under several other situations (employment until stated date; death; disability; good reason termination; involuntary termination without cause). Unless I'm missing something, the right to payment is no longer subject to a SROF upon a change in control, even if the employee remains employed. The one-year timeframe rules out a short-term deferral. The only thing I can think of is some type of separation pay window program, but I'm not sure it would meet the requirements where the same payment is available in several other circumstances (and, for 409A, exceeds 2x base pay/401(a)(17)).

The concern is not in the 410(b) transition rules, but rather the successor plan rule. With some exceptions, you can't distribute elective deferrals from a terminated plan if any member of the sponsor's controlled group has another 401(k) plan. If the stock purchase closes, the sponsor and the new parent company are in the same controlled group. If the parent has a 401(k) plan, the sponsor cannot terminate its 401(k) after closing and distribute elective deferrals--the plans would have to be maintained separately or merged. This is why plans are typically terminated effective the day before a stock acquisition. If the sponsor terminated the plan before closing, but took longer than 12 months to make distributions, under Rev. Rul. 89-87 the IRS could argue distributions were not made as soon as administratively feasible. Unless you had specific facts to rebut that presumption, the original (pre-closing) plan termination date would no longer be valid and the plan would have a post-closing termination date, bringing you (in my opinion) back to the successor plan issue.

Thank you both. I thought the recent IRS guidance would be helpful as well. Bird, the only context where I've encountered this and I thought it made a difference was a plan termination before a stock purchase. If the plan is not considered "terminated" due to the 12-month distribution rule, my understanding is it could no longer distribute elective deferrals under the successor plan rule and would have to be merged into the buyer's plan. At least that's the general rule regarding stock acquisitions and successor plan rule; not sure the outcome would be any different if the plan termination was post-closing by design or inadvertently as a result of slow distributions.

Interesting, thanks. Agree on the audit--my fingers got out in front of my brain. Out of curiosity, have you ever had a 12+ month situation where the plan was working diligently on distributions and the IRS affirmatively said it was okay (as opposed to it happening without IRS knowledge and no one ever raising it as a problem)?

Would uncashed checks still be considered assets of the plan if they remained uncashed into a new plan year? For example, 401(k) plan terminates on September 15, 2018. All checks are written and other distributions made by September 14, 2019. Someone doesn't cash their check until January 2020. Does the plan need a 2020 5500 and audit? Does the uncashed check mean all assets are not out of the plan within 12 months and the original termination date of September 14, 2018, is no longer valid?

An existing governmental 401(a) plan holds matches as well as allows employer nonelective contributions. Is there any problem with describing multiple different allocation formulas with different eligibility and vesting schedules for match, nonelective, or both? Since it's governmental, we don't need to worry about 410(b), 401(a)(4), or other nondiscrimination testing. Any variations in vesting schedules will be more favorable than the longest permissible vesting schedules and will slightly favor non-HCEs. So, for example, people in category A get a match, but no one else does. People in category B will get a nonelective contribution of X% of compensation, with a six-year graded vesting schedule. People in category C will get a nonelective contribution based on years of service with a two-year cliff vesting schedule. And so on. Appreciate any thoughts.

Say you have the following scenario: Plan sponsor is being acquired mid-year and terminates its 401(k) plan, say, November 1. For administrative/payroll reasons, employees of the acquired company will not participate in the buyer's 401(k) plan until January 1. Plan has a safe harbor matching contribution. The plan sponsor wants to make the participants whole for missed matching contributions during the two-month gap, but profit-sharing allocation provisions would not allow a PS contribution targeting only people who deferred. If the plan sponsor paid the participants a bonus (outside the plan) in the amount of matching contribution they would have received for the rest of the year at their deferral rate in effect on the date of plan termination, does this violate the contingent benefit rule? It arguably does because the cash bonus is contingent on the employee having had a deferral election in place at plan termination. What if the bonus is paid after the plan is already terminated? No one has any further deferral right or election, so it's hard to make something "contingent on" an election that no longer exists. If it does violate the contingent benefit rule, what would be a correction? Appreciate any thoughts.

I'm interested to hear others' opinions as well as I see this issue frequently. I'm not aware of any specific guidance, but personally I think if you have a plan subject to 409A/457(f) you're going to have some difficulty getting around the substitution and/or exchange rules under both. The circumstances may occasionally get you past the presumption of a substitution, but I think in most cases it would be difficult unless the two events (forfeiture of NQDC and implementation of split dollar) are truly unrelated in time and amount. I see rescission most frequently in short-term deferral agreements that are exempt from both 409A and 457(f). In that case, you don't invoke either set of substitution/exchange rules. I have heard some of the opinion that you need to comply with the 409A rules for making an initial deferral election on a payment that would otherwise be a short-term deferral (1.409A-2(a)(4)) or deferring a forfeitable right (1.409A-2(a)(5)). I think there's a good argument that neither of those rules apply when you are forfeiting one right and obtaining a new, different right. For what it's worth, this is from the BNA Portfolio on Insurance-Related Compensation: Exchanging Nonqualified Deferred Compensation Plan Benefits for the Proceeds of Life Insurance Under a Split-Dollar Plan (Benefit Swapping)-- Before the enactment of §409A, nonqualified deferred compensation benefit swapping was popular. Since the enactment of §409A, the swapping of a benefit from a plan covered by §409A would likely constitute an impermissible distribution as a “substitution” and a violation of §409A. The technique is also probably not available to a nonqualified deferred compensation plan grandfathered under §409A. Such a swap probably constitutes a material modification that causes loss of the grandfathering and brings it under §409A, hence a substitution and a violation that cause immediate taxation. However, for a nonqualified deferred compensation plan that is exempt from §409A, such as one that qualifies for the short-term deferral exemption, the technique may still be viable, although its availability is not clear. In the present environment, this technique should be considered aggressive. ... The IRS had not approved (or disapproved) these transactions in a revenue ruling or letter ruling before the enactment of §409A, but many were being executed (as reflected in proxy statements). The authors and many other commentators believed these pre-§409A transactions did not create taxable income for the executive on the exchange under either the economic benefit or constructive receipt theories, discussed in V.B.2.c., above. Similarly, the other possible theory under which the transaction could be taxed, the assignment of income theory, also appeared not to apply, because the participant would be releasing, not assigning, the benefit and was forgoing a benefit taxable only in the future for a benefit that was currently taxable. ... In the authors’ opinions, giving up a benefit would not be taxable, and taxation of the new benefit should be determined under the rules applicable to that benefit if the transaction does not constitute a §409A “substitution of benefits.” There is no direct authority for these transactions, especially after enactment of §409A, so the technique must be considered aggressive. Practice Tip: The enactment of §409A with its prohibition against “substitution of benefits” has greatly reduced the number of supplemental plans (SERPs) that might now qualify for the swap technique in the first instance. Only nonqualified deferred compensation plans not subject to §409A, primarily supplemental plans that qualify for the short-term deferral exemption (“vest and pay” supplemental plans), might qualify. As noted, even then, the swap technique should be considered an aggressive benefit planning technique.

This may work for self-insured, but in my experience insured policies usually also have an "actively at work" requirement that mandates the employee to be actually working a minimum number of hours per week, and not on paid leave with no duties. I see it (and other variations) happen all the time. I've had clients undergo audits and, thankfully for them, the only bad result has been to take the person off the policy at the end of the month. In theory, though, the insurer can and will deny claims if the amount at issue is large enough. Usually I see the requests come from situations that are otherwise sympathetic but would trigger large claims (e.g., Bob's wife is undergoing cancer treatment and she they want to stay on for 24 months instead of 18 months because of X...can't we do something?).

Earl, if the plan excludes bonuses from comp, and only non-HCEs get bonuses, it disproportionately harms non-HCEs. For what it's worth, I've always been under the impression of 1-2% rule of thumb. I think 5-7% is pretty aggressive.

True legal mergers are rare in private company transactions, but if A is dissolving/liquidating and B is taking over any remaining assets, employees, patients, etc. (even if not paying cash for them) I think at a minimum you have an asset acquisition or "similar transaction."

I got an acknowledgement letter in June for a submission made in January.

jpod, with respect to the OP, yes I think that much is settled. The thread came back up again under slightly different circumstances today.

kmhaab, I'm comfortable with the "true" short-term deferral that includes vesting in 2019 and actual payment before March 15, 2020. I think that's clearly exempt from both 409A/457(f) and can be taxed in the year of payment (provided the employee doesn't have the ability to choose). Luke, I'm glad it's not just me. Part of what still mystifies me is that, if we're assuming the payment is not subject to 409A (and meets the short-term deferral exemption, whether needed or not), then it's also a short-term deferral exempt from 457(f) under the proposed regs. Exemption from 457(f) would kick the payments back into the normal 61/451 timing rules, but in that case you wouldn't have taxed the payment under 457(f) upon vesting in 2019 (because it's not made until 2025), and therefore wouldn't be exempt from 409A as a short-term deferral because no "payment" has been made via 457(f) taxation, which would then take away the 457(f) short-term deferral, which would put you back under 457(f), which would require upon vesting taxation in 2019, which would exempt you from 409A, etc., etc. ad nauseam.

Circling back here, I've looked at the proposed 409A/457(f) regs in more detail, and am even more unclear now, particularly on the short-term deferral aspect. I must be either misreading or misunderstanding some aspect. The proposed 457(f) regs say that a short-term deferral under 1.409A-1(b)(4) is also a short-term deferral under 457(f). (Set aside for now the differences in the definition of SROF between 409A and 457(f).) At the same time, IRS issued proposed 409A regulations saying that for all purposes of 409A, including whether a "payment" is made for purposes of being a short-term deferral, income inclusion under 457(f) counts as a payment. In the situation discussed in the posts above, you have a "payment" (both actual and income inclusion) within the short-term deferral period. This seems to easily satisfy both the 409A and 457(f) short-term deferral exemptions. How about a situation where payment is made after vesting, say employee is given a right during 2019 to a $100,000 payment on January 1, 2025, that is never subject to a SROF. My understanding has always been--and continues to be--that vesting results in taxation under 457(f) in 2019. The PV of the future payment is income in 2019, then the actual payment in 2025 is taxed under section 72. But the proposed regs seem circular. If you have a vested right in 2019 to a payment that will not be made before March 15, 2020, it's taxed under 457(f) in 2019. But if you include the vested amount in the employee's income in 2019, it's taxed under 457(f) before March 15, 2020. If the amount is taxed under 457(f) within the 409A short-term deferral period (i.e., before March 15, 2020), it's also a short-term deferral under 457(f). If it's a short-term deferral it's not subject to 457(f). Footnote 2 of the 409A proposed regs seems to say that this would be a short-term deferral for 409A (except earnings accruing after vesting) but is not perfectly clear whether it's also a short-term deferral under 457(f). The first sentence suggests yes, but the ultimate outcome suggests no. Maybe it's a short-term deferral under 409A but not 457(f), i.e., the 409A definition incorporated in the proposed 457(f) regs does not circle back to create a 457(f) short-term deferral? What am I missing?

For what it's worth, at the most recent ABA May Meeting, the IRS took the position that an increase in benefits for all participants in the plan means an increase in benefits for every participant in the plan. The example was "the plan provides a fixed 10% profit-sharing allocation to all participants, but the plan sponsor inadvertently only made an 5% allocation to every participant (not just one or some). You can self-correct by adding the 5% difference for every participant." Their position also was that a failure relating to a plan feature that does not affect every participant (e.g., an across-the-board problem with loans or initial eligibility) does not allow for self-correction under the new rule even if you are attempting to correct for every participant affected by the feature (if that includes less than every participant in the plan).

Wouldn't it also be a short-term deferral exempt from 457(f) as well? I had understood exemption as a short-term deferral under the proposed 457(f) regulations to result in taxation in the year the payment is actually made (i.e., you avoid the "early" taxation rules under 457(f) and the payment is taxed when actually or constructively received). So if it vests in 2019, but is exempt from 457(f), it's taxable in 2020 if paid in 2020. No?

"Very confusing" is the most charitable way I would describe it right now.... I think it's a little clearer from S Corp 2's perspective because you drop the circularity of S Corp 1's ownership. S Corp 2 owns 50% of LLC directly. Under 1563 and 414(c), that's enough to put S Corp 2 into the stock exclusion rules. Both those sections exclude stock owned (whether directly or through attribution) by an employee of LLC. Nothing I see limits the attribution from running up S Corp 1's side and back down to Individual 1's spouse (the LLC employee). Maybe looking at it from another angle would help (for me, at least). Say S Corp 2 really did want an employee of LLC to own the other 50% directly. That would clearly exclude the stock and make S Corp 2 and LLC a parent-sub group. If not for attribution, the LLC employee could just have her spouse own the S Corp 2 stock in his wholly owned S Corp (or LLC or directly). From that perspective, I guess it makes more sense to aggregate S Corp 2 and LLC than it does to aggregate S Corp 1 and LLC. Still seems like an odd outcome.

Luke, would the outcome not still be controlled by the stock exclusion rules of 1563(c)(2)? If I'm reading through correctly (which by all means is not certain), I think you still fall into the stock exclusions. Section 1563(c)(2)(A), laying out the stock exclusion rules, starts with "For purposes of subsection (a)(1)..." Under 1563(c)(2)(A), the parent corporation already directly owns 50% of the subsidiary. The attribution rules in (d)(1) and (e)(4) only apply in that context to determine whether the parent owns at least 50% of the subsidiary to take you into the stock exclusion rules to begin with. Because parent directly owns 50% of subsidiary, the stock exclusion rules apply without any attribution. Under 1563(c)(2)(A)(iii), stock in the subsidiary "owned (within the meaning of subsection (d)(2)) by an employee" of the subsidiary is excluded if subject to restrictions. Section 1563(d)(2), generally dealing with brother-sister controlled groups, says stock is owned by a person if it's (A) owned directly by the person, or (B) deemed owned under "subsection (e)." Section 1563(e) contains all the attribution rules, including spousal attribution in (e)(5).