blguest

acm's note above about the withdrawal restriction is spot on, and in the context of benefit divisions for QDROs, can wreak havoc on or completely derail property divisions for unsuspecting parties to a divorce, and their lawyers. TIAA calls their flagship product so affected by the moniker "TIAA Traditional Annuities" (TTAs). I routinely write QDROs allocating as much of the assigned benefit as possible from the non-TTAs portion of benefits held by the participant, whenever possible. Disentangling marital property is normally high-priority task for divorcing parties, so most parties are only too glad to structure their QDROs this way. However, there are too many cases in which the participant holds most of their plan investments in TTAs, forcing property divisions of limited assets to be dragged out over many years. One would think plan sponsors have enough collective clout to pressure TIAA into modifying their entrenched rules on TTAs because it causes so much hardship for participants going through life changes.

Hi David, yes, there is nothing mysterious about DC loans not being loans in the traditional sense, nor how they are treated in QDROs. The question, which I likely could have posed in a less oblique way for better clarity, is whether some plans treat non-defaulted loans as "vested", even though you and I would not consider loaned-out money as "vested" because it isn't liquid or available for payment to an alternate payee. The reason I asked Benefits Link neighbors the question is because I have seen some plans' documentation containing specific language stating as much, while others have (unofficially) stated they consider non-defaulted loans to retain their vested character unless defaulted. That latter interpretation seems nonsensical to me, particularly with respect to property division cases. Perhaps others here could enlighten us both.

I remember seeing that language in contexts other than for IRAs, Peter, so your surmise Vanguard may have used it in a 401(k) context is probably correct. They did use it in a 403(b) context as of 2008-2010 as well: https://hr.ucf.edu/wp-content/uploads/sites/17/Vanguard-Application.pdf.

Believe it or not Peter, that's a Vanguard choice their lawyers must have made after partying too hard. See: https://investor.vanguard.com/investor-resources-education/beneficiaries/adding-beneficiaries-to-ira

Based on documents from several different plans, I've operated under the assumption that a participant's non-defaulted loans from a DC plan reduce the vested portion of a total account balance, even though they do not reduce the overall total account balance for other purposes, such as for calculations of assignments to alternate payees (except where the loans would limit the assignable amount), because non-defaulted loans are considered assets of an account, even if they reduce the vested portion. Am I wrong, or do different plans handle this in different ways?

Hi David, I think you are right that you are in the minority on this. Your notes in the files you attached ask the question whether a plan has the right to limit the benefits awarded pursuant to state law, and it appears the focus of your inquiry turns on your reading of 1056(d)(3)(B)(ii)(II). The statute provides “domestic relations order” means any judgment, decree, or order made pursuant to a State domestic relations law, and it does not include a carve-out for estate proceedings. You may be focusing on state domestic relations law as driving the boat, but neglecting to factor the federal nautical chart and industry currents. A central theme in Boggs is that a would-be-QDRO issued in a non-domestic-relations context such as an estate matter does not fall within the definition of 1056(d)(3)(B)(ii)(II), because such an order is not made pursuant to domestic relations, but to devise and descent, even if the expectation originated in an earlier domestic relations matter. This is why most plans will not qualify an order for a would-be alternate payee who predeceases qualification. Also, most states' devise and descent laws have at least some treatment of the kinds of property that may inure to an estate, and expired expectations are not usually among them. Just my 2 cents, fwiw.

I'd maybe try inquiring of Spark or ICI. I think you can email them at glossary@ici.org.

1. Yes, if your QDRO tells them to divide the benefit in that way as of the account segregation date, then all assets in the account as of that date will be affected. 2. It depends on state matrimonial law. Most states have either statutory provisions for this or court rules that specify what a retirement benefit division order like a QDRO can and cannot achieve, and most limit that to carrying out existing court orders, such as carrying out a divorce decree incorporating a property settlement agreement, because the function of a QDRO is to do just that. 3. Because the time to negotiate a different outcome than that memorialized in your MSA and now incorporated into your divorce decree has come and gone.

Second Carol's first sentence, though you may want to send it to plan counsel before responding to it. FWIW, there are two recent appointments on the bench at the USDC Albany, both magistrates, both with prosecutorial backgrounds, if that's where the order came from.

A recordkeeper that exercises discretion in making a decision about a participant's eligibility for benefits, as described above by EPCRSGuru, is not performing a ministerial act but is unilaterally assuming the mantle of a fiduciary. At the very least, the plan sponsor and administrator should review their services contract with that provider to determine an appropriate response, because no response implies consent or acquiescence.

FormerEsq: Did you miss "duty of care" and "if they operate in a fiduciary capacity within ERISA's ambit"? Exercising discretion by making an overpayment falls afoul of §404(a)(1)'s mandate of the duty of care.

Hi David, is yours the "Admiral" favor of the fund? The variants seem to have different inception dates. The "ballpark" method is one the parties came up with (at the suggestion of their divorce lawyers), and that I advised against, but people will do what they will do. In general, when early statements are unavailable for a marital property division but parties do know a beginning and ending balance, many of their lawyers will want to take that beginning, premarital separate property balance, factor in EE/ER contributions at some agreed static level for each (monthly or annual) period, add investment experience for each of those periods per a fund's published return, to arrive at something resembling PV, then subtract that PV separate property portion from the balance on the day the parties separated, to then arrive at a marital portion to be divided. It's the kludgy method for parties who cannot afford to hire an economist of some sort.

@FORMER ESQ. By "prosecute" I meant, as part of their duty of care, to engage, investigate, take on, or pursue, rather than having a duty to file a criminal complaint. Between ERISA, the IRC, SarBox, and SEC rules, recordkeepers exist in a highly-regulated environment requiring constant vigilance to maintain the viability of their liability insurance. If they operate in a fiduciary capacity within ERISA's ambit and do not follow up properly on rectifying their mistakes having overflow effects on plan sponsors and administrators, they risk regulatory, financial, and potential criminal penalties. EPCRSGuru is right to consult with plan counsel.

Hi Peter, your comments are helpful, as always. I did suggest using the S&P returns for the same reason you cite, that the same plan investment array in earlier periods cannot be known with any certainty without documentation. The participant says they only ever had that one fund, and it does appear by that exact name on the earliest statement they have (2003). While Vanguard should know best what the inception date of any of their funds is, the participant said they had that investment as of 1997, never changed it until the first plan merger in 2005, and stockanalysis.com gives that fund's inception date as Aug 31, 1976 (See https://stockanalysis.com/quote/mutf/VFIAX/). I believe there are/were different flavors of the Vanguard 500 Index Fund, with the "Admiral" flavor being one variant. I like your idea of using a five basis points lower than S&P measure as a yardstick and will suggest that if they can't come to some agreement without all this fuss (which has been ongoing for over a year, sigh.) Thanks much!

VFIAX = the "Vanguard 500 Index Fund Admiral". I have a client for whom I'm calculating a marital portion of an old 401k containing that single investment. That 401k has undergone some mergers and recordkeeper changes such that historical statements for individuals are no longer available. The parties agreed to ballpark the present value of the premarital balance, and to estimate that premarital balance by working backward from a later balance by subtracting annual fund overall returns on that single investment. Whether or not this is the right/smart way to approach it is not the issue, finding the historical fund returns is. (I'm a lawyer rather than a math person, but I can run a spreadsheet with reasonable competence.) The annual returns sought on that VFIAX fund as a whole are for the years 1997-2003. I have been able to locate overall fund returns back to 2003, but not before that, and would be grateful if anyone happens to have those. Suggestions and criticisms are also welcome.

That act smells like a recordkeeper most concerned with their own liability and not getting sued by the plan, as no matter the fault of the participant, that participant could not have proceeded without the recordkeeper's own error. Whether or not the participant now has a large sum of money they shouldn't have, liability for that lies with the recordkeeper. If their remedial measures meet the requirements for correction such that the plan has no correction or distribution liability, then the participant's status with that property would seem to be a personnel problem for the employer rather than a fiduciary challenge for the plan administration. ERISA's fraud provisions provide the employer with ammunition enough for termination proceedings regarding the employee, but the question remains how the recordkeeper's duty under ERISA to prosecute that potential fraud (rather than ignoring it) impacts plan administration, and for that you should definitely consult plan counsel.

Thanks Peter, I have an earlier edition of that, and also Shulman's handbook, both of which are helpful but neither of which provide much insight on this issue. I also haven't had any luck with searching similar cases on Fastcase or even Google scholar. I think that of PA's who want to insist on particular benefit division formats, most of which are reasonable, they do not get much pushback precisely because they are reasonable, and if one makes the point that under some circumstance or another that accommodation needs to made, most are reasonably accommodating. It is only in this case among the thousands of others I have come across that the PA's position is unsupportable. As none of our neighbors here have identified any authority that might support the PA's position either, I think what happens next is, if the plan's counsel wants to back the PA's decision, a judge will qualify the order, which will then be served, prequalified, on the plan. The plan will then either do the calculation or continue to refuse, prompting a formal claim for benefits, which the PA may also refuse, at which point the remaining option is federal court. My experience with federal courts is that they tend to read the federal statutes strictly, though of course past experience doesn't guarantee future results. A strict reading of § 206(d)(3)(C)'s subsection (ii), with its multiple "or"s, and a lack of published legal interpretation supporting the PA's refusal, could be helpful, but hopefully it won't get that far if the PA wants to avoid litigation. Sigh. Thank you all for your input and insights, and if you think of anything else, I'll be grateful to hear it, even if critical.

You would be right @Artie M that amending a QDRO would be easier, assuming the numbers are available to an alternate payee with which to do the math. The problem arises when the denominator is not available to an alternate payee. The numbers for a benefit in pay status are always available to a PA though, in this case one who refuses to cite any authority for rejecting a coverture fraction, making writing a QDRO that carries out a garden-variety property division impossible for such an alternate payee. Do you know of any authority permitting an ERISA plan administrator to reject DB orders using such a fraction, without citing any justification other than they just won't do it?

@Peter Gulia I have met those judges too Peter. The question remains though, can you or any neighbors here think of any authority that permits an ERISA plan administrator to reject DB orders using a coverture fraction, without providing any justification whatever, other than they just won't do it? If anyone can think of such authority, I am open to learning what it is. I agree that less expensive and more effective means are usually preferable (particularly in the current political climate), and certainly there are never any guarantees in litigation. However, it seems to me that forced means are not limited to subverting parties' rights, a plan administrator's 'meh' undermining many states' domestic relations schemes without any authority is a problem much larger than one QDRO. @fmsinc Thanks for relaying your experience David. Like you, I push back, and in this case already requested they escalate the matter to plan counsel. We'll see what that individual has to say.

@david rigby, I always review a plan's procedures. In this case, while I requested the plans' procedures, the PA did not provide them and instead provided the plans' model order for a benefit in pay status (identical for both plans). The plans' model does not explicitly prohibit fractions, it instead has fields for amount or percentage. Please help me understand under what authority do you (and QDROphile and Peter) see an ERISA plan being able to circumvent § 1056(d)(C)(ii) by refusing to do the math required by the statute's explicit grant of "manner in which such amount or percentage is to be determined", as long as that manner is clear? Yes to @Peter Gulia, these are definitely ERISA plans and their 5500s are in order. Are you thinking here that § 1056(d)(C)'s "clearly" negates its own subsection (ii)'s explicit grant of methodology when that methodology is clearly defined, and executed in the present orders, as a matter of longstanding industry practice?

Hi Effen, yes I am sure. They stated they will not qualify an order using a fraction for any benefit in pay status, they will only consider hard numbers (straight percentage or dollar figures). They had no issue with the architecture of the fraction or its presentment, only that it was in use at all, which is what has me here canvassing for others' thoughts. As I read § 1056(d)(C)(ii), and of course the plan document, their demand has no legitimacy, but as it is always possible that those here with experience on the PA side of things may have ideas I haven't considered, I'm all ears. The weird part is that these are not small or new or specialized plans. That the sponsor is a quasi-governmental entity, albeit the plans being purely ERISA creatures, may have something to do with it, but the PA must still administer the plans properly.

Thanks for your thoughts QDROphile. Courts frequently divide DB benefits by application of a coverture fraction, aka the time rule application to the marital portion of a benefit (X is assigned 50% of the marital portion of Y's benefit). Most folks here are aware that the fraction defines a marital portion of a benefit, the numerator representing a length of a marital period and the denominator representing the entirety of a benefit. A QDRO defines these elements precisely so that a plan can calculate the portion allocated to an alternate payee. The plan admin could not cite to any plan document provision exempting them from the application of 29 USC § 1056(d)(C)(ii) (manner in which such amount or percentage is to be determined), and I'm not seeing how the application it of creates any kind of problem for an ERISA DB plan. Of course, some plans may not want to do the math, even though that stance costs litigants (both plan beneficiaries) more money, but that does not give them an out from the statute for qualifying an order. You see it differently though, it seems, and I would appreciate your further thoughts on why. The time rule / coverture fraction exists because parties to a QDRO always know the numerator, but almost never know the denominator. For a benefit in pay status, the denominator is part of the plan's records, and may be known by the participant, but alternate payees and courts almost never know it, and plans are not forthcoming with the data in the absence of participant cooperation. As participants are frequently not around, or have died, that information remains out of an alternate payee's reach, making doing the math impossible for all but the plan.

Anyone ever come across an ERISA DB plan administrator who refuses to qualify QDROs employing a time rule (coverture fraction) division for benefits in pay status? I am dealing with one, first one ever in 30 years of practice. All comments welcome.

A situation that can become an infinite loop when a claim is in the wings but is not ripe to be made. For example, a QDRO-in-waiting that has not yet been submitted for qualification because a sponsor/participant's estate (which has already submitted letters testamentary to the TPA and stands in the shoes of the deceased sponsor/participant), cannot get the TPA to provide a current account statement. I have a client with that very issue right now (you may recall I'm a QDRO lawyer). Trustee sponsor/participant of very small plan dies after the court enters a property settlement, scant paper records in the decedent's estate, no copy of an executed beneficiary form; estate counsel pretty much ERISA-clueless. Sponsor company has a DC plan TPA'd by one firm, and a cash balance plan administered by another TPA. The cash balance TPA won't pony up a current account statement to the estate administrator/PR, so neither the estate nor the alternate payee for that plan can ascertain what exactly is there that is divisible between the estate and the alternate payee. Then, instead of providing a current account statement and their QDRO procedures document, the TPA decides, unbidden, to retain its own counsel to write a QDRO for the alternate payee (!), which, shocker, does not allocate the full components of the benefit, though nothing in the plan document prevents full allocation. (Of course, I would not allow my client to sign such an abomination.) Additionally, the cash balance TPA's benefit statement from several years ago (the only statement the estate has), is labeled for the sponsor's DC plan (the one administered by a different TPA), includes a single line item for the cash balance plan without identifying that plan as distinct from the entire rest of the statement. This is not a small-estate matter and there is likely 500k+ in the participant's hypothetical account. In 30 years of practice, I have never seen a TPA screw up this badly. I'm counting the misrepresentations, fiduciary breaches, and prohibited transactions, and wondering when they'll stop shooting themselves in the foot before I sue their pants off, as they're not listening to reason. Thank the stars original poster Santo Gold has the wits to ask their learned colleagues here for their thoughts when unsure.

I'll rob from Paul here to pay Peter. Paul makes an important point in emphasizing negative consent, which from a plan participant's perspective looks very different from Peter's perspective of advising a plan sponsor. In evaluating a course of action for my own clients, I always put on opposing counsel's hat to determine what their responsive legal argument might be before I make my own argument. For example, if a plan participant is going sue a plan and its fiduciaries, that participant's counsel will home in on actions taken that are arguably not in the best interests of participants, which could be shown by introducing the kinds of evidence you list in your third post, Peter. In particular, negative consent-basis actions would appear to a litigant to be a ripe target for close scrutiny, especially if an end result has demonstrably detrimental effect on participants. So yes, from a legal perspective, I would think a prudent plan fiduciary has a responsibility to consider those possible outcomes before instituting a practice that has many known unknowns and the result of which have the potential for damaging participants. If a plan participant sues, be prepared in discovery to show plan fiduciaries considered potential outcomes. Even with lots of plan policy communications to participants, jurists will consider whether the average plan participant is likely to understand them.