JMT44

So, I made the case to the actuaries (communicating through the employer) that the RMD should be calculated using the individual account method. My reasoning relied on the plan document. The section dealing with RMDs was a copy and paste of the LRM language. That language spelled out the calculation for every conceivable scenario except for a post-70.5 retirement lump sum from a DB plan other than an individual account (414(k)) plan. If neither the individual account nor the annuity method was specified in the plan document, then either could be used. In such a case, the method chosen should be the one most favorable for the participant. This was rejected by the actuaries. Their position was that 1.401(a)(9)-6(d), "does not require a choice to be made and memorialized in the plan document, but rather presents the Plan two options to be used when calculating this amount. Further, please see Section 6.10(a)(iii) of the plan document where the regulations, including these two methods, are incorporated by reference. The Pension Plan’s process around RMD calculations is to use the annuity basis and it does so on a consistent basis when advising employees of this issue." I responded with two requests. First, I pointed out that policies could be revised for good and valid reasons, and a change to a method that better served participants was clearly a good and valid reason. I requested that change. Second, I stated my belief that the election to use the annuity method, even if not in the plan document, would nevertheless have been formally adopted and documented in some manner by the plan. Such a document would have to be available for examination, so I requested a copy or the chance to examine the document at the office of the employer. The employer's response was that the request for a change in policy would be forwarded to the appropriate parties, but that I should not expect a reply for at least several weeks. One day later, the employer responded that the actuaries were recalculating the RMD using the individual account method! No explanation was given!! I can only guess at the reason for this abrupt change. Perhaps someone immediately recognized that such a policy change should be implemented, and decided that it was not necessary to make us wait until the Administrative Committee was able to meet and formally approve the change. Perhaps, since no mention was made of the request to examine documentation regarding this policy, the policy had never been formally adopted and documented. Or maybe it was clear that the participant's spouse had decades of pension legal and technical experience, actually knew a thing or two, and (as a retiree) had lots of time and would keep the employer and the actuaries occupied with this issue for the foreseeable future. If that was the case, maybe recalculating was considered the best way to make the problem go away. The bottom line ... I don't know why the change was made, but I'm pleased with the outcome.

Perhaps the key word here is should. As you suggest, if the plan screws up and understates the RMD, the penalty is payable by the participant. That's because the true RMD is not whatever the plan happens to calculate, but the amount that meets the requirements of the law. Similarly, if the plan calculates an RMD that greatly exceeds what the law requires, the true RMD is still that which satisfies the law. The amount that must be distributed (in my opinion) is equal to the lesser of the individual account and the annuity calculation. As long as the distribution is at least equal to the lesser amount, it satisfies the law. That is why I have an issue with the RMD amount being used by the actuaries servicing the plan. It is not the true RMD, but an amount that exceeds the true RMD by 85%. But unless participants (or participants' spouses) have significant pension technical skills, they will (not surprisingly) rely on the reported figure. Those who would like to minimize their distributions will end up with more withholding and higher taxes. That may be better than a 50% excise tax, but it's still a disservice to the participant.

This is definitely one possibility for sure. However, let me play devil's advocate for a moment. Instead of withdrawing the RMD amount, let's say a participant takes a total distribution. If the participant decides within the 60 day window to roll those some of all the distribution to an IRA (making up any shortfall due to income tax withholding out-of-pocket), would he/she turn to the plan trustees to determine the amount not eligible for rollover? Or, would he/she be the one responsible for the RMD calculation? Once the funds are no longer plan assets, would he/she not be the one to choose the calculation method (annuity or individual account)? I'm not sure that there are definitive answers to these questions. I don't know if these specific questions have ever been addressed. The 401(a)(9) regulations were finalized after I retired, so if there are definitive answers, I would have no way of knowing. However, based on years of experience, I understand that these kinfd of questions that can linger for years with no firm guidance.

Calavera, thank you for your post. You are correct (and so are your assumptions). The $4115 (plus 56 cents) is the RMD amount using the individual account method. I'm trying to make contact with the actuaries in hopes of getting their support, but so far I'm being told that they don't deal with the public. (Time for an actuary joke. How can you tell if an actuary is an extrovert? The actuary looks at your shoes when talking to you!) Sorry about that. I've been retired from the pension field for well over a decade, but apparently that's not long enough to be immune to flashbacks. While the documentation already received clearly states that the RMD is $7,599, I'm hoping to convince the actuaries to use the lesser amount as the 2018 distribution as you suggest. I don't know how this will turn out, but I might have better luck if I quit with the actuary jokes!

Thanks for posting the correct question number. Actually, I'm not surprised by the IRS answer to the question. Since the taxpayer is liable for the 50% penalty for failing to take the RMD, the IRS would understandably hold the taxpayer liable for a failure to take at least the minimum. I can't image the IRS sending a notice to the custodian or administrator looking for the unpaid taxes. The taxpayer, on the other hand, could conceivably go after the custodian or administrator if he/she reasonably relied on that person's calculation. Also, this would be a textbook example of a case which qualifies for a waiver of the penalty. It may not the kind of experience one would enjoy, but the taxpayer should emerge relatively unscathed (financially, that is, not emotionally).

Larry, Thanks for your post. With regard to Question 2, my understanding comes from IRS Required Minimum Distributions FAQs. See Question 2 here: https://www.irs.gov/retirement-plans/retirement-plans-faqs-regarding-required-minimum-distributions. With regard to the 1099R produced by the plan, it is true that a 1099R is used to report taxable distributions. However, that does not preclude a person from rolling the funds during the 60 day rollover period. A direct rollover or trustee to trustee transfer is never required. When filing the tax return, a taxpayer rolling part or all of a eligible rollover distribution would report the total amount of the distribution on line 16a of the 1040 (the 1099R amount), then indicate the net taxable amount on line 16b. An eligible rollover distribution does not include any amount that represent a required minimum distribution, which gets us back to the question as hand. Does 1.401(a)(9)- 6(d)(1) apply in the case of the recipient, or would a plan decision not to recognize 6(d)(1) mean it is no longer available to the employee? (To me, this sounds like, "Yes, the IRC says this is an acceptable way to compute the RMD amount, but only if your employer says it's okay with them.")

I've been retired for more than a decade, so my legal and technical skills are pretty rusty. But suddenly, it seems, I have the need (opportunity?) to put them to use once again. My spouse just retired after working beyond age 70 1/2 and has not been subject to RMDs until now. The plan is to take a lump sum of $105,358 from employer's defined benefit plan. The payment election form states that the RMD is $7,599, which is not eligible for rollover. This represents the correct amount under under Section 1.401(a)(9)-6(d)(2) (the annuity method). Using the 6(d)(1) (individual account method) the RMD would be $4,115. The plan document does not state which method the plan will use (although there may be some ancillary document I have not seen). If memory serves me right, the annuity method (which is almost always far less favorable for the participant) is very uncommon. I am trying to determine if the plan sponsor has formally adopted this less favorable annuity method. From a plan perspective, I can think of no reason why a plan sponsor would knowingly disadvantage a participant. In the meantime, I have two questions. 1. Section 1.401(a)(9)-6(d) is written in a way that suggests the individual account method may be the default if no method has been formally adopted. I say this because subsection 1 says the RMD "is determined by treating the single sum distribution as a distribution from an individual account plan." Subsection 2, on the other hand says the RMD "is permitted to be determined by expressing the employee's benefit as an annuity." If no method has been formally elected by the plan sponsor, does the individual account method automatically apply? 2. The responsibility for determining the RMD ultimately belongs to the payee, not the plan sponsor or the administrator. So, if the administrator distributes the larger amount to the participant, could the participant rollover the excess of the annuity method over the individual account method as a 60 day rollover. In other words, does the choice of methods ultimately belong to the participant who is the one responsible for satisfying 401(a)(9) in the end? Thanks for sharing your thoughts!