RMD and death in year one turns 70.5 Options

[GMK]

Wow. Sometimes you even get what you want (an answer to post #22). Thank you, Sieve and JSimmons.

Although no one appears to agree with JSimmons' conclusion (except maybe Kevin C), IMHO, his logic appears to be bullet-proof with regard to qualified plans. Of course, if I knew for certain, then I wouldn't have to ask the questions.

If a plan faces this situation, I would not decide based on my level of conservatism or the size of the distribution. I would get what I need by having the plan get an opinion in writing from the plan's attorney, as jevd suggested previously and as JSimmons just recommended.

[Guest_named_mjb_*]

GMK --

Don't know if anyone ever answered your post #22 re: different RBDs and whether more than 1 can apply. The answer is No--if the death RBD applies, then the age 70-1/2 or termination of employment RBD no longer applies.

Although stated declaratively, that Sieve's opinion. Mine is different. The situation of the OP meets the explicit requirements of both -3 and -5, nothing in the regulations provides that if -3 applies then -5 cannot. My position is that both apply to the OP situation. Ignore -5 at your own peril--or get a legal opinion (so you might have malpractice recourse if need be).

IRS Pub 590 explicitly informs taxpayers that "If an IRA owner dies after reaching 70 1/2 but before April 1 of the next year no minimum distribution is required because death occurred before the required beginning date".

The above language is consistent with the MRD regs. which definitively state that no MRD is required if the IRA owner dies before April 1:

1.401(a)(9)-2 Q/A-2(a) "The term required beginning date means April 1 of the calendar year following the later of the calender year in which the employee attains age 70 1/2 or retires from employment" and -3 Q/A-1 states "if an employee dies before the employee's required beginning date (and thus before distributions are treated as having begun in accordance with Section 401(a)(9)(A)(ii)) (Ed- i.e., over the employee's lifetime) distribution will be made in accordance with one of the methods described in sections (B)(ii), (ii) or (iv), etc. (Ed-which relate to beneficiary distributions which did not commence by the MRD).

If an MRD is required for the year age 70 1/2 is attained why would the regulation explicity note that death occurs before the distributions are treated as having begun over the employee's lifetime?

Reg 1.408-8 Q/A-1 states that "...the rules of reg. 1.401(a)(9)-1 through 9 for defined contribution plans must be applied except as otherwise provided in this section."

The symmetry in the above regulations and IRC 401(a)(9) demonstrate the obvious: No MRD is required if death occurs before April 1 of the year after the IRA owner or participant attains 70 1/2.

Are you saying if a taxpayer relies on the above language in Pub 590 and the MRD regulations the IRS would impose a 50% excise tax for the failure to take a MRD for the year the IRA owner attained 70 1/2 but dies prior to April 1 of the following year?

[Guest_named_mjb_*]

Wow. Sometimes you even get what you want (an answer to post #22). Thank you, Sieve and JSimmons.

Although no one appears to agree with JSimmons' conclusion (except maybe Kevin C), IMHO, his logic appears to be bullet-proof with regard to qualified plans. Of course, if I knew for certain, then I wouldn't have to ask the questions.

If a plan faces this situation, I would not decide based on my level of conservatism or the size of the distribution. I would get what I need by having the plan get an opinion in writing from the plan's attorney, as jevd suggested previously and as JSimmons just recommended.

Given the language in Reg 1.408-8 there is no basis in the MRD regs for differing between MRD required under an IRA and a qualiied plan.

[J Simmons]

As a result of this thread, I know a couple of sources I can send those who are death beneficiaries of a (former) employee that reached age 70 1/2 but then died before April 1 of the next calendar year for an unequivocal legal opinion on the topic.

[Guest_named_mjb_*]

As a result of this thread, I know a couple of sources I can send those who are death beneficiaries of a (former) employee that reached age 70 1/2 but then died before April 1 of the next calendar year for an unequivocal legal opinion on the topic.

You haven't answered my question:


Given the statements in IRS publications and the MRD regulations 1.401(a)(9)-2 and 3 and reg 1.408-8, would the IRS impose the 50% excise tax if an MRD is not taken where a participant who attains age 70 1/2 dies before April 1 of the following year?

This post has been edited by mjb: Oct 30 2008, 01:12 PM

[J Simmons]

As a result of this thread, I know a couple of sources I can send those who are death beneficiaries of a (former) employee that reached age 70 1/2 but then died before April 1 of the next calendar year for an unequivocal legal opinion on the topic.

You haven't answered my question:


Given the statements in IRS publications and the MRD regulations 1.401(a)(9)-2 and 3 and reg 1.408-8, would the IRS impose the 50% excise tax if an MRD is not taken where a participant who attains age 70 1/2 dies before April 1 of the following year?

Are you opining the IRS wouldn't in the DC context?

[Guest_named_mjb_*]

As a result of this thread, I know a couple of sources I can send those who are death beneficiaries of a (former) employee that reached age 70 1/2 but then died before April 1 of the next calendar year for an unequivocal legal opinion on the topic.

You haven't answered my question:


Given the statements in IRS publications and the MRD regulations 1.401(a)(9)-2 and 3 and reg 1.408-8, would the IRS impose the 50% excise tax if an MRD is not taken where a participant who attains age 70 1/2 dies before April 1 of the following year?

Are you opining the IRS wouldn't in the DC context?

Have you been reading my posts?

Here is my statement From #32:

"The symmetry in the above regulations and IRC 401(a)(9) demonstrate the obvious: No MRD is required if death occurs before April 1 of the year after the year the IRA owner or participant attains age 70 1/2."

From # 33

Given the language in Reg 1.408-8 there is no basis in the MRD regs for differing between an MRD required under an IRA and a qualified plan.

What part of no dont you understand?

Are you ever going to answer my question?

[J Simmons]

Have you been reading my posts?

Yes, I've read them. And while I respect your inductive reasoning--if the IRS says in a publication (not regulation, proposed regulation, Rev Rul, Rev Proc, GCM, etc) about IRAs that no MRD is required for the calendar year of 70 1/2 if the IRA owner dies after reaching that age but before his RBD on April 1 of the next calendar year, then that's an interpretation that also you also apply to the DC context--I would not base any legal opinion letter issued under my signature on that basis. I'd certainly argue it to the IRS if a client came to me not having taken the MRD for the calendar year the DC plan employee reached age 70 1/2 but then died before his RBD. I'd certainly argue it in a request for PLR or in tax court.

But for the death beneficiary seeking my opinion as to whether the RMD for the calendar year of 70 1/2 should be taken by April 1 of the next year despite the employee's death, I would make sure to advise that there is nothing directly on point from the IRS--as well as your inductive reasoning argument. I would also point out that the IRS could claim on the basis of -5 that the MRD for the year of age 70 1/2 is required, and they'd be seeking a 50% penalty tax. I'd want my client to understand both interpretations, the arguments favoring and disfavoring each, and what the consequence of foregoing the MRD might be (to weigh against the loss of further tax-deferral on what the MRD payout amount would be), so that my client could make an informed decision as to what course of action to take.

What part of no dont you understand?

Is that a trick question? Before I answer, I'd like to know what parts of no are there?

Are you ever going to answer my question?

I think I just did in this post. Will you be answering mine, i.e. Are you opining the IRS wouldn't [assess the 50% penalty] in the DC context? If so and you'd be willing to give death beneficiaries and estates an unequivocal written legal opinion to that effect, I'll know where to send those not satisfied with my drawing out the pros and cons of the two viable interpretations.

This post has been edited by J Simmons: Oct 30 2008, 06:14 PM

[Guest_named_mjb_*]

Have you been reading my posts?

Yes, I've read them. And while I respect your inductive reasoning--if the IRS says in a publication (not regulation, proposed regulation, Rev Rul, Rev Proc, GCM, etc) about IRAs that no MRD is required for the calendar year of 70 1/2 if the IRA owner dies after reaching that age but before his RBD on April 1 of the next calendar year, then that's an interpretation that also you also apply to the DC context--I would not base any legal opinion letter issued under my signature on that basis. I'd certainly argue it to the IRS if a client came to me not having taken the MRD for the calendar year the DC plan employee reached age 70 1/2 but then died before his RBD. I'd certainly argue it in a request for PLR or in tax court.

But for the death beneficiary seeking my opinion as to whether the RMD for the calendar year of 70 1/2 should be taken by April 1 of the next year despite the employee's death, I would make sure to advise that there is nothing directly on point from the IRS--as well as your inductive reasoning argument. I would also point out that the IRS could claim on the basis of -5 that the MRD for the year of age 70 1/2 is required, and they'd be seeking a 50% penalty tax. I'd want my client to understand both interpretations, the arguments favoring and disfavoring each, and what the consequence of foregoing the MRD might be (to weigh against the loss of further tax-deferral on what the MRD payout amount would be), so that my client could make an informed decision as to what course of action to take.

What part of no dont you understand?

Is that a trick question? Before I answer, I'd like to know what parts of no are there?

Are you ever going to answer my question?

I think I just did in this post. Will you be answering mine, i.e. Are you opining the IRS wouldn't [assess the 50% penalty] in the DC context? If so and you'd be willing to give death beneficiaries and estates an unequivocal written legal opinion to that effect, I'll know where to send those not satisfied with my drawing out the pros and cons of the two viable interpretations.

Your analysis of the application of the -5 reg to require an MRD for the year age 70 1/2 is attained if death occur prior to the following April 1 is flawed because -5 Q/A-1(a) states that "the minimum distribution required to be distributed for a distribution calendar year on or before the employee's required beginning date is always determined under 401(a)(9)(A)(ii) "- which is the employee's life expectancy- while -3 Q/A-1(a) expressly provides for distribution to a beneficiary of a participant or IRA owner who has attained age 70 1/2 "if an employee dies before the Required Beginning Date (and thus, before distributions are treated as having begun in accordance with section 401(a)(9)(a)(ii))". -2 Q/A-1(a) defines the RBD as the April 1 following attainment of age 70 1/2. In other words, -5 Q/A-1(a) requires distribution over a IRA owner or participant's life expectancy under IRC 401(a)(9)(A)(ii) for the year such person attains age 70 1/2 which is not possible for distributions under -3 Q/A-1(a) where by definition the IRA owner/participant dies before payments are treated as having begun to be paid over the employee's life expectancy in accordance with IRC 401(a)(9)(A(ii).

[GMK]

While looking at 1.408-8, especially Q&A-5, and related references, I came across this:

http://www.naepc.org/newstech-0701.web

which shows two (probably rare) cases where the regs do not define what the IRS will actually rule in PLR's.

I have very much appreciated the discussion in this thread, but in the end what matters is whether the IRS answers 'yes' or 'no' to mjb's question.

I understand that PLR's do not apply to us all in general, but are there any PLR's (or other pronouncements) in which the IRS has decided to impose or not impose the excise tax in the case of a qualified plan?

[Kevin C]

Here is my statement From #32:

"The symmetry in the above regulations and IRC 401(a)(9) demonstrate the obvious: No MRD is required if death occurs before April 1 of the year after the year the IRA owner or participant attains age 70 1/2."

MJB,

Your statement leads me to a question. A still employed 5+% owner turns age 70.5 in 2008. The 401(k) plan does not allow in-service distributions, except for required minimum distributions. He takes his 2008 RMD on 12/30/2008 and his 2009 RMD on 1/15/2009, then dies on 3/15/2009. Your statement is that his death means there are no RMD's for 2008 or 2009. So, he was not eligible for either distribution. What should the plan do about the retroactive operational failure caused by his death?

[Sieve]

Since it is impossible to determine until April 1 of the year following the year of this owner's attainment of age 70-1/2 whether death will occur prior to the RBD, thus negating the need to make the MRD, I would suggest that the earlier MRDs do not produce an operational failure since, when those distributions were made, the plan was fully complying with its terms. The only way to guarantee that there is no "failure" would be to withhold payments until that April 1--not a very practical approach.

Besides--I think these are MRDs when they are distributed prior to RBD. It's just that, for purposes of making MRDs on account of death, these MRDs would not have been required to be made if distributions had not already occurred. The regs don't say you can't make these distributions--only that, if not made before death, they need not be made after death to the beneficiary.

This post has been edited by Sieve: Oct 31 2008, 11:44 AM

[GMK]

From 1:408-8:

"Q-5. May an individual's surviving spouse elect to treat such spouse's entire interest as a beneficiary in an individual's IRA upon the death of the individual (or the remaining part of such interest if distribution to the spouse has commenced) as the spouse's own account?

"A-5. (a) The surviving spouse of an individual may elect...to treat the spouse's entire interest...as the spouse's own IRA. ... If the surviving spouse makes the election, the required minimum distribution for the calendar year of the
election and each subsequent calendar year is determined under section 401(a)(9)(A) with the spouse as IRA owner and not section 401(a)(9)(B) with the surviving spouse as the deceased IRA owner's beneficiary. However, if the election is made in the calendar year containing the IRA owner's death, the spouse is not required to take a required minimum distribution as the IRA owner for that calendar year. Instead, the spouse is required to take a required minimum distribution for that year, determined with respect to the deceased IRA owner under the rules of A-4(a) of §1.401(a)(9)-5, to the extent such a distribution was not made to the IRA owner before death."

The question Q-5 explicitly applies both before distributions begin (entire interest) and after distributions have commenced.

The last 2 sentences of the A-5. (a) answer (starting at "However, if ...") seem to say that if the surviving spouse makes the election in the same year that the IRA owner died, the spouse is required to take an RMD for that year of the owner's death, determined with respect to the deceased IRA owner, less previous applicable distributions to the owner before death.

How is this reconciled with the Publ. 590 statement about no RMD required? (I must be missing something, which would be no surprise.)

Does this also apply to qualified plans?

Sorry if I seem like a pest. I don't mean to be. Just trying to learn something.

[Sieve]

This reg stands for the proposition that, if the spouse must take an MRD for the deceased IRA owner in the year in which the spouse elected to rollover into the spouse's IRA, then the spouse takes as beneficiary rather than as owner of the IRA. But it only applies IF the spouse must take an MRD for the IRA owner--and, we all know ( ) that the spouse need NOT take an MRD for the owner if the owner dies prior to the owner's RBD.

The reg you cite--and, specifically, the sentences you point to--cross reference reg. -5, Q&A-4(a), which deals with the requirement to take a distribution with respect to the year of death IF THE DECEASED ALREADY HAS REACHED RBD AND HAD ALREADY COMMENCED DISTRIBUTIONS, and is used to determine the distribution period for calculating the payout during the distribution calendar year.

For example, "[i]f an employee dies on or after the required beginning date, the distribution period . . . is determined as if the employee had lived throughout that year." (Reg. -5, Q&A-4(a).) In other words, IRA owner starts receving benefits, dies after the RBD at age 73, but would have turned 74 before the year ended--then the MRD for that year must be paid, and is based on IRA owner's age 74 (even though IRA owner died before reaching age 74). And,the reg you cite indicates that that MRD is paid to the spouse AS A BENEFICIARY of the IRA.

It does not change the general rule that if an employee or IRA owner dies BEFORE RBD--and remember, there is only one RBD--no MRD is required to be paid.

(And, no, the reg you cite does not apply to qualified plans since you cannot take as the owner of a qualified plan if you are the beneficiary--whereas you can take as the owner of an IRA if you, as spouse, rollover the IRA and treat it as your own. You always take from a qualified plan as the beneficiary, while there might be an issue as to whether you take from an IRA as beneficiary or owner (which the reg you quote deals with).)

In my opinion, of course.

This post has been edited by Sieve: Oct 31 2008, 04:17 PM

[Kevin C]

Since it is impossible to determine until April 1 of the year following the year of this owner's attainment of age 70-1/2 whether death will occur prior to the RBD, thus negating the need to make the MRD...

That was my point. The majority interpretation creates a situation where it is impossible to follow a fairly common set of plan provisions. Either the IRS messed up, or there is something wrong with the majority interpretation.

I don't think your solution of considering them required if paid before death works either. RMD's relate to a "distribution calendar year". Unless I am missing something, part of the majority interpretation is that the first distributon calendar year in this situation is determined either under the life expectancy rule or the 5 year rule. That would make the first distribution calendar year no earlier than the calendar year following the year of death. That means the distributions while alive can't be RMD's because they occur in a year prior to the first distribution calendar year.

As interesting as a discussion of the regulations can be, the only opinion that really matters is the IRS's. Since there doesn't seem to be any official guidance clearly addressing this situation, does anyone know if this has been discussed in an IRS Q&A session? If not, maybe someone could get this on the list?