Mike Preston

Oh, to be complete, you also need to know if there is a plan imposed limit. I'll assume there isn't. But if there were, and if this person violated both 415 and a plan imposed limit, there is an ambiguity in the regulations and it is unclear whether you are forced to use your catchup on the Plan Imposed Limit (PIL) before correcting the 415 violation. So my spreadsheet asks two questions that I think need to be asked before one can be sure that there is a consistency in applying the calculations to a specific plan: 1) Correct 415© before Plan Imposed Limit? 2) Include Plan Imposed Limit violations in ADP? Bet you are sorry you asked!

Your conclusions are correct. But since you are an actuary, your arithmetic is incorrect, so your assumptions as to amount are wrong java script:emoticon('') smilie. My calculator says that $11,000 plus $18,000 is $29,000, not the $28,000 you suggest. It looks like the calendar year 2004 deferrals were conveniently precisely equal to the allowable deferral without catchups ($13000) plus the allowable catchup deferals ($3,000), or $16,000. Clearly, $3,000 of that amount counts as catchup for 2004. You don't mention whether the $5,000 contrubuted in the plan year ending 6/30/04 was subject to an ADP test that failed, thereby using some portion of the 2004 catchup in the 6/30/04 plan year. I'll assume there wasn't any so that the full $3,000 catchup for 2004 was available to the 6/30/2005 plan year overlap (7/1/2004 through 12/31/2004). Still with me? As of 1/1/2005, everything is copacetic. That is, you have used the 2004 catchup to the fullest extent allowable. Now we look at the six month period ending 6/30/2005. We find $18,000 of deferrals, which just so happens to be the limit of both the 2005 deferrals without catchups ($14000) plus the allowable catchup deferrals ($4,000). Sounds wonderful, huh? Oh, you fail your ADP test, huh? You do so counting only those amounts which aren't catchup deferrals under 401(a)(30), right? That amount is: 2004 = $11,000 minus $3,000; or $8,000 2005 = $18,000 minus $4,000; or $14,000 Total of $22,000, not the $24,000 you mention. You suggest, although you don't specifically say, that $9,000 is this participant's limit for the 6/30/2005 ADP test. You suggest that you test $24,000 and that $15,000 must be returned. I suggest that you test $22,000 and, assuming the same limit of $9,000, you return $13,000. Do you agree with my math? Couple more points, that probably are irrelevant. To do a complete test, you need to know that the person's compensation exceeds the deferrals (highly likely with an HCE, but you never know) and that there are no other ER contributions which might casue 415 to be violated if aggregated with the deferrals. Who wrote these rules? <lol>

Essentially, yes. However, for ABT purposes they are tested together (you test the years that end in the same calendar year together).

Could have set up separate plans.

I think I'd ask the question in reverse. What is the citation under 414(l) that calls for aggregating separate defined contribution plans of the employer into one plan for coverage purposes?

Tom was being nice to me because he knows what I meant, rather than what I wrote. As to coverage, the plans can be tested separately or combined. However, if not combined, you still use the total population of the combined entities is used to determine whether either individual plan passes. How you test coverage determines how you run ADP/ACP. If you combine for coverage, you combine for testing. If not, you don't. One catch, though. If you don't combine, the ADP/ACP of any HCE that participates in both plans is determined for both plans based on combined comp/deferral/match/eecont.

Don't jinx us, Tom.

Put it in now. Make sure anybody who terminates is in a group which gets the amount they otherwise would have gotten had the formula not been changed.

Neither? It depends on the definition of a17 compensation in the plan. Certainly, it can never be more than the a17 limit. The only question is what is that limit? Assuming the plan applies annual compensation limitations, you would indeed limit it to $205k. If the plan, however, implement a17 on a pro-rata basis, by month, I think you are stuck with 12 * ($35k + $205k + 83333.33) / (2 + 12 + 5) = $204,210.52. What does your document say?

Ouch!

Andy, in your case, I think the prior distribution should be brought forward, at a minimum, at the plan interest rate. At least for purposes of determining the plan benefit that is payable. I could see an argument for bringing forward the prior distribution at current 417e rates to determine the 417e benefit payable, but I understand that two separate rates being used in this way can lead to strange results. Until the 415 regs are effective, I think the best course of action is to do something reasonable. Following the draconian proposed regulations would not, in my opinion, necessarily be reasonable unless a discussion with the plan sponsor and potentially the plan sponsor's attorney takes place.

With respect to the original poster's calculation, the 415 at 68.5, based on 5%, 94GAR, is 18568.69. The lump sum PFEA max, based on 94GAR/5.5% is then 18417.71 * 124.00110 = 2283816. I think when the 417(e) rate drops below the plan rate, you are stuck with the plan rate when bringing forward a prior distribution. Certainly when trying to determine the benefit payable under the plan, anyway. That would be $900,000 * 1.06 ^ 6 = 1276667, resulting in a dimunition of the plan benefit payable of 1276667 / 114.61168 = 11139.07. So, the plan benefit must be reduced by 11139.07. Let's assume that the plan benefit is still very, very high, such that the only concern is the 415 limit. There is a decision to be made: in the valuation are we assuming a lump sum distribution or an annuity payout? That is a critical determination. Let's assume lump sum. The maximum lump sum then would be the maximum lump sum of 2283816 reduced by some reasonable interpretation of what the lump sum offset should be based on the prior $900000 distribution. Prior to the new proposed regs coming out, I would have argued for consistency (rather than "the worst of all worlds", which is what the proposed regs seem to do) and therefore would have reduced the PFEA lump sum of 2302538 by $900,000 * 1.055 ^ 6 = 1240959, resulting in a lump sum to fund for of 1042857. Notice that I don't explicitly calculate the annuity benefit payable. That means i back into that number, depending on what the computer is using for an APR. To be consistent, I would probably argue for using 94GAR/5.5% and therefore argue for an annuity at age 68.5 of $8410. The max accrued benefit at 1/1/05 is relevant to the Current Liability determination. At 1/1/05 the participant is 66.5, with a 415$ limit of 15838.28 and an offset for previously distributed amounts of $900,000 * 1.06 ^ 4 / 122.38076 = 9284.38. The actuarial equivalent of this amount at age 68.5 is the accrued benefit to be used in the valuation for determination of current liability. Again, assuming that the actual plan benefit is so high as to not invoke anything other than pure 415 limits. However, I have always had a bit of difficulty with current liability that exceeds projected liability. Such is the nature of our conflicting sets of assumptions. I think I've reconciled myself to this type of anomaly. There are certainly other reasonable ways to determine current liability, such as a pro-rata approach, or an approach that limits it to no more than the projected benefit valued at current liability assumptions. I'll make a separate post for Andy's hypothetical.

Ooops, posted to the wrong thread. Sorry. All content deleted and moved to correct thread.

mbozek.. Interesting. I certainly concur that if the result upon death is the revocation of previously issued orders, be they separation agreements or QDROs in anticipation of divorce finalization, there is a distinct disconnect. I can't provide a citation, but out here in California, the QDRO in the absence of divorce is almost common. Well, not really common, but not so rare as to be unheard of, either. Thanks for the dialog. I agree that if the spouse has given up rights via an anticipatory QDRO, they should also be willing to execute a beneficiary designation agreeing to somebody else as beneficiary. I would suppose that a properly drafted separation agreement might direct the spouse to do so at the behest of the participant.

I disagree. I think they apply. If it is a corporation, it is W2. If it is W2, it is KNOWN as of the point in time when it is paid.

mbozek, you keep talking about a divorce action being terminated. In the case of a legal separation, there is sometimes no intent to ever have a divorce. I think the regulation is pretty clear: in the case of a legal separation, the plan document can, if it chooses, decide that spousal consent is no longer necessary. Further, even if the document does not invoke that special consideration, if the legal separation document satisfies the definition of a QDRO it would continue to be valid upon death. Of course, the OP should determine what is really going on in the plan by referencing the plan document and then getting confirmation from the plan's lawyer.

Couldn't the separation agreement be a QDRO?

Good plan.

Your plan's beneficiary designation form will tell you if it is ok or not. Some plans require that a spouse sign the form, witnessed by a notary, if somebody other than the spouse is to be designated. Does your form require that? Let's assume it does. Is your question whether this participant is to be considered "not married" for purposes of the form? Well, IANAL, but I have always thought that a legal separation was not sufficient to scissor a spouse from benefits otherwise called for under the plan, unless the separation agreement is recognized by the courts and calls for the spouse to give up all rights under the plan. Let's see what the lawyers have to say.

I think I'd go for it. If it is too late to issue a statement now that conforms to the timeframes indicated, then what else are you going to do? Besides, it sounds like as long as there isn't any forfeiture on death that the statement does, in fact, comply. Have you asked anybody to review the quarterlies to see if they might comply with 6057(e)?

I don't think anybody here will be able to offer an opinion based on the facts that you have presented. If the firm you are working with says your chances are pretty good, then it is highly likely that your chances are pretty good. Do you know whether your ex remarried? Do you know who your ex named as beneficiary under the plan? Do you have a marital settlement agreement that specifies what you are entitled to? Was your QDRO stiplulated or contested? If stipulated, who represented your ex? Do you have a copy of the SPD for the plan? Do you have a copy of the actual plan (not just the SPD)? was your ex in pay status when he died? If so, do you know what benefit was being paid? If so, do you know who he named as his beneficiary with respect to his benefit election? As you can see there are a number of questions that need answering and some research documents that are needed before anyone can give you an opinion. But, since I'm not a lawyer, you can ignore everything I say anyway!

Did it hurt?

mbozek, I'm not convinced that 510 enters into this at all. Before one can be accused of an ERISA 510 violation one must establish that there exists a "right to which he is entitled under the provisions of an employee benefit plan." I think the order of things is pretty well established. The participant's benefit is restricted. The QDRO can't order the plan to pay a benefit that is not payable. Therefore there does not exist any right for the beneficiary to argue exists under ERISA 510. I don't have time to do a research project on this, as I've just dropped back in because I was asked to provide the cites that Tom posted earlier on this thread. But there are a bunch of cases that hold along these lines, aren't there? I'm thinking of cases where a participant is entitled to a benefit predicated on a contingency, and the alternate payee comes in and asks that since that contingency doesn't apply to them, they should be able to receive the benefit. The courts have held, pretty universally IIRC, that the contingency applicable to the participant is a contingency applicable to the spouse. An example is the early retirement subsidy that an active participant is entitled to take should the participant actually retire. No retirement? No subsidy. And the QDRO can't be interpreted in light of the contingency being satisfied. In this case, we have an even more difficult contingency to waive. In this case the contingency is predicated upon the operation of the plan (investment performance) and actions of the plan sponsor (contributions) rather than action of the participant (actually retiring). But since I'm here, let's parse 414p3 a bit. (A) is what I think you are referring to. You are reading it to apply to the general forms of benefit available to other beneficiaries. I am reading it to say that any specific option available to the participant must be available to the beneficiary. [We know there is an exception for J&S benefits. That is, a plan does not need to provide the beneficiary with an option to take the benefit in the form of a J&S, even if a J&S benefit is available under the plan to the participant. I forget where that exception is made, but it makes perfect sense to me.] I think I'd need a Supreme Court case to convince me that the proper interpretation of (A) is that which you posit. Until then if it isn't payable to the participant it can never get to the point of being protected by ERISA 510 because it never becomes a right of the beneficiary to receive a form of benefit not otherwise payable to the participant.

Yes. Yes (if there are keys in both). No.

It is now 5. ;-)