Steelerfan

They'll tell you anything to get rid of you. Was the QDRO signed by a judge? You need a family law lawyer well versed in employee benefits matters. The determination of whether assets are marital property is for a state court to decide, and the plan or employer, as applicable, can be forced to comply with a court order (you generally don't need a QDRO, just a DRO should do). A contempt order might get their attention.

The more I thought about this strange arrangement, I thought the same thing jpod did. The employer created a plan that probably fits the definition of an ERISA pension plan--deferral of compensation to termination of employment, and should have satisfied the minimal ERISA filing requirement--statement to DOL, etc. There was clearly some kind of "promise" to make a payment in the future. But if I were the ER, I would argue there is no enforceable ERISA claim or contract. It's going to be the employer's word against the EEs. [see Gallione v. Flaherty [70 F 3d 724 (2d Cir 1995)] (The court concluded that the participants had no remedy under ERISA's enforcement provisions and then analyzed the top-hat plan under traditional contract theory. Finding no plan provision that vested the participants in their plan rights at any particular time, the court concluded that the employer had the right to terminate the top-hat plan at any time and that the participant's contract claim had no merit because the plan was terminated before the participant retired (the event that gave rise to benefit entitlements under the terms of the plan)).] But see Carr v. First Nationwide Bank [816 F Supp 1476 (ND Cal 1993)], a federal court ruled that participants in a top-hat plan have a federal ERISA common-law right to have their top-hat plan benefits protected. The court held that there exists under ERISA a "common law doctrine of unilateral contract" under which an employer-sponsor of a top-hat plan can unilaterally bind itself (as First Nationwide had) to provide a specific top-hat plan benefit and under which plan participants may sue under ERISA to enforce the employer-sponsor's unilateral promise. There are alot of problems here so why not just sweep the 1993 promise under the rug and figure out how to come as close as possible to giving the employees what they "expected" to get by any means legal and possible to avoid potential ERISA issues? Obviously there was poor tax planning here and the employees will have to deal with any unintended tax consequences--it always comes down to careful communication when an ER messes up--or just gross them up--if they are executives they have a God given right to have their taxes paid for them.

This is a can of worms I am afraid to even talk about!

mjb: My understanding of this example is that it confirms that despite an irrevocable election to defer, if an employee receives a payment in that same year of deferral, for example, because he retires, then the amount is still "deferred compensation" in terms of reporting, etc. I don't see how it creates authority to pay back amounts. Unless I'm missing something big, it looks like the opposite to me--that once you receive payment, it retains it's character as deferred comp, and if you had no right to receive it, then a failure occurs unless an exception applies. It sounds like the RR you mention uses a breach of contract theory to provide tax relief, but will it hold water post 409A? Under your theory, which would be great, this provision creates a large "oops I made a mistake exception" to 409A, which I don't believe it does. jpod: These approaches would be fine, but my concern is that they may not provide relief if the IRS assesses penalties after you've "fixed" the problem and you're a couple years down the road facing an audit. I certainly don't advocate giving up, but I think correction of failures has to be one of the next major 409A projects the IRS needs to take up (after they tell us how to report).

This topic is so interesting I had to respond. I had a tax court judge as a professor who was mentally impaired enough to think that QDROs could apply to NQDC plans. People who are too afraid not to use a QDRO are setting a bad precedent and analytically, it is clear that non qualified plans are not subject to QDROs. Even just the logic of applying a "qualified" rule to a "nonqualified" plan ought to give some hints as to the right answer. That being said, litigation of wrong positions can't always be avoided. BTW ERISA does not and has never interfered with a state's right to decide what is marital property (except with respect to the joint and survivor annuity). The only limit ERISA places on divorce courts is that with plans subject to QDRO rules, you must procedurally use a QDRO to get at the assets themselves. I thought Gburns misunderstanding of the division of marital property was entertaining.

Based on page 9 in the final regs, this question should be raised to the IRS. In light of the general rule (see the first paragraph on page 9) that any early payment would constitute an impermissible acceleration, there would seem to be sufficient doubt as to whether the prior IRS authority would constitute an exception to the 409A failure rules. The example the IRS uses is "if a service provider has made an irrevocable election to defer an amount of his or her salary to a future year, that amount is treated as deferred compensation regardless of whether the service recipient actually pays such amount to the service provider during the year in which the services are performed." I know that provision is probably not intended to cover mistakenly made payments, but it's enough to make me nervous about administrative failures and the consequences of messing up without any formal guidance as to correction methods.

A promise was made in 1993, a promise isn't a transfer of section 83 property. With no documentation other than the policies, isn't there an ERISA written plan issue (or is there an exception for life insurance)? What did the employees promise as consideration in return, any continued employment, etc.? If not, it may not be clear whether there is grandfathering if nothing has been earned and vested, or whether there is a contract, or whether a section 83 transfer would occur (what kind of an arrangement is this?). As far as SOF, assuming there is a contract, from what I can remember, a contract that can be performed w/in one year is an exception, so depending on the meaning of "retirement', the lack of a writing may not be an issue, for example if retirement includes termination of employment for any reason then no writing is generally required, as with contracts of employment at will. Sounds like you need either an army of lawyers or one very smart person

Exactly, but you said it better. From a pure 409A standpoint this seems right to me, but increasing "past" service credit (if that's what is happening) seems to cause other issues involving previously reported w-2 amounts (i.e. what would happen when non taxable deferrals have to be reported code Y?), amounts that were already included in FICA wages and other accounting/actuarial issues. Maybe someone with more accounting or actuarial experience could chime in? It seems to me you would have to separately account for these additional accruals going forward, like in a qualified DB the actuaries have to have to set up separate formulas and track which provisions of the plan applie to each "bucket". If the SERP is a wrap plan and not paying out when participants go over qualified plan limits (is that what's happening?), then changing the design of the plan to a a cash balance plan with a definition of comp tied to the exectuvites salary and or bonus might accomplish the goal

I can't think of any reason why an employer couldn't make these types of amendments to a NQP because they are nonelective. The "deferral" is made prior to the employee receving any LBR to the compensation therefore you satisfy 409A requirements. You could even change the form of payment with respect to the new deferrals, I think.

What's the purpose? If a liquidation is a change in control, then payment should be tied to CIC definition compliant with 409A. Otherwise it looks like you're trying to get around the rules prohibiting funding on insolvency.

If there is a true SRF up until the payment date, the short term deferral rules should take you out of 409A. Otherwise I would think you have a failure. But it's interesting that no employees said anything. If they had refused payment would CR have occured?

I guess I assumed that when you said "my concern deals with the ability to design a plan so that the form of payment, though specified in the plan before amounts are earned, is driven by the PV at the time of commencement", you meant there would be a trigger at that time. If not, then I agree no CR issue. I don't mean to beat a dead horse, but the door would still seem to be open for the IRS (if they end up being really aggressive) to claim that the "down the road" determination of PV (even if assumptions are set pre LBR) is really a second election that is not in compliance with 409A. If that happens all amounts for each participant under all non account balance plans not subject to SRF are in immediate violation. Is that really worth the risk? That's what makes extended SRF a little attractive--409A penalties can't begin until a valid SRF ends. Anyway---good luck!

In a simple situation, you might have a plan that will pay out on the later of attainment of age 65 or separation of service; if 65--annuity, if separation--lump sum. Assuming that passes 409A muster, then in that case, since nothing can happen in b/w the LBR and the determinatin of form of payment (except for continued employment) to change the form of payment, there would presumably be no issue. I wonder if the regs are silent because Treasury assumed it was impermissible or that it would be permissible only if the terms were "objective" and predetermined as you say. Maybe the final regs will speak to this. At any rate I would hire an attorney to work in conjunction with an independent actuary. The "wait and see" approach you discuss is creative, but as a result looks risky for obvious reasons and subject to challenge. In this day and age of "backdating" options, do you think the service will give you the benefit of the doubt as to "objectivity" and "predetermined" assumptions? One other planning tool could be to subject the benefits to SRF until payout (won't this become more of a common planning tool even though executives will no doubt complain?). That way you could pay out benefits in a lump sum (or installments) within 2 1/2 months after year in which the SRF ends, such as separation from service, without regard to 409A and constructive receipt won't be an issue if you plan and operate properly. Then, if you want to do an annuity, it might look better optically to the IRS if the benefits are truly forfeitable when the assumptions are put into place prior to LBR. Remember pre-ERISA-even under qualified plans you could lose your benefit if you didn't stay until retirement age. I'm not sure what you're second question is. If you have a proper subsequent election provision in the plan, you can change the form of payment. Constructive receipt would only be an issue if there is some way an amount is "made available" and then further deferred in violation of 409A.

Keep in mind that DB SERPs are usually nonelective so the time and form of payment must be set prior to a legally binding right attaching, not necessarily before they are "earned" but sometimes there may be no practical difference. Anyway, IMO I can't see any problem with violating the anti acceleration rule as long as it is clear that the plan will pay out on whatever time is selected that conforms to 409A (e.g. separation from service). The only potential problem I see the IRS challenging is with the "default" being an annuity there would be a "further deferral" when a lump sum was potentially "available", thus violating either the 409A subsequent election rules or the constructive receipt rules (refer to the Veit I and Veit II tax court cases, and the Martin case). Thus the IRS would argue current taxation of the full lump sum. With the payment form being set in advance of earning, it doesn't seem possible that the IRS could win any legal arguments unless the determination of the PV lump sum is somehow within the control of the employee--i.e. they control the actuarial assumptions used to decide PV, then manipulate the data to defer tax in installments. In order to be safe, I would change the default to lump sum and have the annuity kick in if the PV is over 25K, with a very careful independent evaluation of the determination of PV. In a post-409A world this design is questionable, but IMO this design would have even looked and smelled pretty bad in a pre-409A world (see the cases cited above). Historically the IRS hates extensions of payments even more than accelerations. Practically speaking, since 409A leaves the general rules of constructive receipt, etc. in place, the door is left open for the IRS to examine cases like this where there is potential for abuse. Because of that potential the safest thing to do would probably be to build in a second election option allowing for a determination of PV and a change in form of payment 12 months prior to the initial payment date. Of course then the P has to wait five years before receving a payment.

Your 409A analysis looks correct to me-unless you can make the argument that no legally binding right to the bonus attaches until they are eligible to elect the first 20% payout (you need a legal opinion on that), then in that case there has been no initial election until their election to take the 20%, which would occur simultaniously with the LBR. If that is the case, you don't have to do it 12 months in advance. But as a practical matter, if I were an executive, I wouldn't want to make a deferral under this arrangement. It's not an attractive design because the exec can lose an amount he will no doubt consider to have been earned and you'll end up with a plan that just pays out in 5 year installments. As a planning matter, I would not bother with any additional deferral feature under the arrangement you describe unless they make all their elections prior to when the legally binding right to the initial bonus attaches (or prior to the service period to be safer) and initially elect to receive payment, e.g., after 5 years of employment, but again noone will want to futher defer if they can lose it. If you want a golden handcuff, why not just do a service-based restricted stock unit plan (payable in the form of cash) that forfeits if you leave before the end of the 5-year period, and pays out immediately at the end of the 5-year forfeiture period to satisfy the short-term deferral provision of 409A? Then you can mess around with additional deferral features, made either at the time the legally binding right to the bonus attaches or in accordance with the subsequent election rules 12 months prior to the end of the SRF.

It seem as though something is missing in the facts. The"straight" bonus portion would need to be deferred prior to the start of the service period and would normally be elective, but if they can subsequently elect to convert their bonuses to phantom stock units with an additional 20% "kicker", then the 20% could be an additional amount that they would only get if they stay and make a valid subsequent election (the 12 and 5 rule as I call it for short hand). In that case it seems you could subject the additional 20% to an SRF. Another way to accomplish this goal without the need for a subsequent election would be to make the entire portion of the bonus mandatorily deferred, essentially a nonelective phantom stock unit (i.e, no choice to take cash, but must defer, then give additional 20% and subject the whole amount to SRF). You will have to decide what to do about phantom dividends if you want to be thorough about it.

I was using NY as an example, because that is the state that we were talking about when this issue was raised. Other than federal tax deductions for travel expenses (IRC 162(a)(2)), NY is the only state I know of that uses this harsh (and maybe ultimately unconstitutional) rule to tax employees who are physically working out of state (i.e. telecommuters). (A parallel issue exists for purposes of travel expense deductions--an employee is "deemed to live" near his principal place of employent even if his residence is located in another state). I never meant to imply that this was a general rule for state income or other taxes. There are many different permutations of facts, and we may not be disagreeing. As a preliminary matter, for both tax and non-tax legal issues, a physical presence is not the only event that can cause a sufficient nexus (due process) to create a relationship between an entity and a governmental authority (e.g. you can be sued in a state you've never been to if you own property in that state). Also, other factors may or may not create a sufficient nexus as b/w different types of taxes. Since we are talking about income taxation, most states do generally require a presence in their state in order to tax nonresidents who earn money there, as in your example of ballplayers who play a game out state. But the "convenience of the employer test" is used by NY to subject the money earned by nonresident employees of NY-based employers who telecommute to taxation in NY for work performed outside of NY unless the services "of necessity, as distinguished from convenience, obligate the employee to out of state duties in the service of his employer." Sec 631©; reg 132.18(a); Zelinsky v NYS Tax App Trib, 801 NE2d 840 (2003). In other words, the money earned while working at home cannot be allocated to the home state (such as CT) unless the work has to performed at home, and the employee is, in essence, treated as though they live in NY--it's a fiction so that NY can get the tax dollars. There is virtually no way for a telecommuter to meet this test in 99.9% of cases. NY is just peaved that people don't want to live there and they're protecting (or rather IMO extending) their tax base. Theoretically, the new york state dep't of taxation could try to use the convenience of the employer rule to tax a nonresident telecommuter who has never even set foot in NY, but the department does not try do that. Under the regulations a telecommuter who sets foot in NY for only one day could suject all of his income for the year to taxation in NY. This is a very harsh rule, and as you point out it creates impossible administrative burdens of nightmare proportions, not to mention potential double taxation. The only possible way to avoid double taxation would be for the telecommuter to never set foot in NY all year or win a federal court case. I'm not sure how independent contractors fit into this, but my main concern is the obligation of the ER to withhold, the effect on the EE is a secondary concern, unless I'm the EE.

The issue isn't how they get the work to the NY employer, it's that the state wants the income unless the work cannot conceivably be performed in NY. The "convenience of the employer" rule is used to "pretend" that the out of state employee lives in NY and therefore is deemed to be performing the work in NY. It isn't just the state tax Dep't--the NY courts are adhering to this view and hold that it is U.S. constitutional. You may disagree, but you will lose your tax case. Hopefully federal legislation in the works will solve this.

In general your point is a valid one. New York State statute sec. 674 requires an employer to file a NY state return of tax withheld on Form WT-1. If the employer doen't file, how will the taxing authorities find out? A taddle tale, I guess? Isn't the same true where an employee who works all his life for cash under the table and, obviously, the employer never withheld taxes and the employee never filed a federal return? How does the IRS catch these people? I don't know. There is an element of the honor system to obey the law and if you fail to do so and get caught you should be prepared to pay the costs. This is all about the practicality and burdens of the executive branch of governent administering statutory law.

For example in NY, an "employer" is defined as "any person or organization . . . having an office or doing business in New York, whether or not a paying agency is kept in New York. (emphasis added)" (Reg. 171.2). Therefore, in accordance with HarryO's conclusion, virtually any employee of an out of state employer doing his "employer's business" in NY is subject to state taxation for $ earned while standing on NY soil. As a practical matter, in NY, there would be no withholding obligation on the ER as long as work days are less than 15 days, but the EE could still be subject to income tax for one day spent in NY if he earns wages greater than his personal exemption amount. (reg 171.6(b)(4)). As pointed out above, the general rule would seem to be that every state with a personal income tax will want to tax any employee who "works" there as a nonresident, however most payroll systems cannot currently handle such a burden. With the proliferation of long-distance telecommuting, this issue will no doubt be more on the radar screen, as a state by state analysis would have to be done-tracking where the employee is every single day of the year. Some states, like PA and NJ have reciprocal agreements so that if an employer withholds, reports and pays income tax to the state of residence, they are ok in the work state. In addition, the issue with nonqualified deferred comp may hinge on whether the state follows the federal rules for income taxation or tries to tax the $ when earned. This was an issue in PA prior to a recent change in the law that now puts PA in line with the federal income taxation rules on NQDC (income taxation on distribution). Prior to this change, PA argued that personal IT should be paid when the $ is earned. This would have caused problems for individuals whose ERs did not withhold PA income tax and then moved to Fla and took a distribution. I would hope that now PA could not or would not try to tax NQDC plan distributions to a non-resident of PA even if they earned the $ in PA, since they were not taxable by law when they were earned. This may be an area where Congress will have to intervene, like the NY commuter tax. I don't understand the second post in this thread as it jumps from income taxation for FICA taxation without any transition?

409A has a special rule for bonuses that are performance-based. If the incentive compensation is paid under a performance-based plan that covers at least a 12-month measurement period, the deferral election can be made up to 6 months before the end of the measurement period. This would mean, for example, that an employee could elect to defer his or her 2006 annual bonus (payable in 2007) up until June 30, 2006. However, in order to qualify, the bonus plan must be based on objective performance factors put into place no later than the first quarter of the measurement period. Regulations are anticipated to borrow from, but not be as strict as, the Code Section 162(m) performance measurement provisions. A guaranteed bonus is not eligible for the six-month rule, and is treated in the same manner as base compensation

HarryO: That's true, if a restatement were necessary, I definitely wouldn't push it. In fact, despite what I think the law is (or should be interpreted), I might recommend repricing options just to make a potential problem go away. However, option agreements are contracts, as well as SEC regulated, and a tender offer may be required to reprice--which can be expensive and time consuming. All that trouble before anyone knows what the heck the law really requires! BTW--Lynn was falling all over the place, how could it not have been interference? Besides, Jackie should have caught that TD pass!

You kind of helped proved my point. Your statement quoted from the notice-- "these regulations provide that a non-discounted stock option, that has no other feature for the deferral of compensation, generally is not covered by section 409A" Announcement 2007-18 contains similar language on page 3. ". . . a stock option granted with an exercise price that can never be less than the fair market value of the underlying stock on the date of grant, and that does not include any additional deferral feature, generally is not subject to 409A (emphasis added). Pursuant to the legislative history, at the money options should be exempt from 409A, you won't get an argument from me. But the addition of the word "generally" in IRS guidance (noticeably absent from language in the leg hist) leaves the door open for them to try and have their cake and eat it to. There is virtually no tax-related guidance on when a NQSO is officially "granted" for puposes of taxation. 409A does not answer this question. There are many stock option plans out there that by their terms are prohibited from granting discounted options, yet because of imprecise grant procedures, arguments can be made from ten different angles on whether or not the result is a "discounted" option for tax purposes. The IRS will use this uncertainty to argue that when a NQSO fails to comply with some rule that is unrelated to 409A that 409A is somehow implicated and everyone should fix their options or pay the excise tax. Bunk!

A third option would be to terminate the plan in accordance with the provisions of the prop regs that allow for discretionary termination of a plan within 30 days before or 12 months after a CIC. The rub with this option is that the reg appears to require that all other "similar" plans be terminated as well. If anyone knows how this rule applies, please let me know! It could mean having to terminate all account balance plans (if that's what the directors plan is) that are sponsored by the employer including plans the employees participate in. In my mind this would never be a viable option and I would prefer to regard this part of the prop regs as a big misprint (an idiotic one at that). Based on the discussion as I follow it, the only viable option would be (2) in Scott's post. But HarryO is on the money and the directors get to decide what to do (despite the fact that the acquiring Co. has a vested interest in the decision). And considering all the 409A uncertainties, is cashing out worth the risk if the plan does not provide for it? This thread opens for discussion a lot of really important issues for the future. It seems to me that option (1) is not viable as this would certainly have to be considered an acceleration, otherwise what's the point of 409A--every employer could amend there plan for an earlier payout, just like a haircut provision. I think the likely scenario is that an amendment to the plan to change the payout timing that is nonelective (that is at the sole discretion of the employer and not employee choice) that occurs during or after the transition period would be treated as a subsequent election subject to the 12 month prior and 5 year payout rule (sorry for the lack of specific language). Also, as regards retroactive amendments affecting "old" money, we are not talking about qualified plans with protected benefits here. Every employer should carefully draft their contracts and plans allowing for discretion to change provisions. Where employee payout elections are concerned, my hats are off to companies that allow such election changes going forward and they have to keep track of which election applies to which money.

Although I don't disagree with any of the above responses, I would point out that when an employer elects to pay the FICA taxes on vested amounts in a DB SERP on an annual basis prior to retirement, the amount taxed will inevitably be "non ascertainable" and therefore a "true up" will almost necessarily be required when the employee terminates. Such is the nature actuarial assumptions. Of course, the answer to the origninal question depends entirely upon whether or not the SERP is an "unfunded" top hat plan.