RMD and death in year one turns 70.5

[Guest Sieve]

Don't mean to stir up the pot (well . . . maybe!!), but what do you make of this:

"Will a plan fail to qualify as a pension plan . . . solely because the plan permits distributions to commence to an employee

on or after

April 1 of the calendar year following the calendar year in which the employee attains age 70-1/2 even though the employee has not retired or attained normal retirement age under the plan as of the date on which such distributions commnece?" (Treas. Reg. Section 1.401(a)(9)-8, Q&A-9. Emphasis added.)

The answer is "No", it won't fail to qualify, but payment has to be on or after that April 1 (whether or not a 5% owner) to fit within the 4 corners of the Q&A.

Notice that this is the normal RBD for someone reaching age 70-1/2. Based on the response to this question, payment prior to the RBD is not permitted, but payment ON the RBD is ok. I know that this is dealing with the issue of the permissibility of an in-service distribution from a pension plan, but I find it strange that payment prior to the RBD but after reaching 70-1/2 is not allowed. So, does this lend any credence to the earlier thought that perhaps payment of MRDs prior to that April 1 are not MRDs unless they literally are paid on April 1--or am I smoking something illegal?

[GMK]

Sieve - I don't see any mention of RMD or RBD in the -8 Q&A-9.

I read it to say that a plan can allow distributions to participants who are over age 70-1/2 and who have not yet retired or who are past or not past the normal retirement age. Non-5% owner participants over 70-1/2 who are still employed are not required to take RMD's and do not have RBD's yet. This Q&A comfirms that the plan can allow them to take a distribution anyway. And it goes on to say that the plan can allow the 5% owners to take distributions of more than the RMD amounts.

Or is there something more subtle that am I missing?

[Guest Sieve]

Well, I think this Q&A stands for the proposition--in fact, it says--that if you want to provide an MRD to a 5% owner who attains age 70-1/2 but has not reached the pension plan's NRD (i.e., an MRD that is required by IRC Section 401(a)(9), by the way) then you cannot pay that MRD until April 1 of the year following the year of attainment of age 70-1/2. You cannot pay it earlier, i.e., before the 5% owner's RBD (based on attaining age 70-1/2).

In the first place, I find that amazinglingly limited--and I would suspect that MPPPs do not operationally do it that way. Secondly, it suggests that a MPPP may allow 5% owner pre-NRD MRDs for the first distribution calendar year only if the distribution is made literally on the RBD itself, and, by insinuation, therefore, that any distribution after age 70-1/2 is not an MRD until that specific April 1. And, doesn't it force the 5% owner to take 2 MRDs in one year?

But I may be stretching. Just wondered what others may think of it.

[GMK]

Sorry, but I don't think Q&A-9 limits RMD's, nor do I think is meant to. It doesn't 'say' MRD or RMD anywhere.

As I read it, Q-9 looks at the situation where RMD's would normally be due, but because the participant is still employed, they aren't due. And it asks if it's OK for a plan to allow distributions to participants even in this case where RMD's are not due. The answer says 'yes' and confirms that even for 5% owners (where RMD's are due), the plan can allow additional distributions on or after the April 1 date. It does not say, for example, that RMD's cannot be made before April 1. It only talks about distribitions made on or after April 1.

Yes, the 5% owner must take RMD's for the year she/he turns 70-1/2 and for each year thereafter, and the first part of the distribution(s) she/he takes is RMD until the RMD amount is satisfied, but none of that is because of Q&A-9. IMO, Q&A-9 simply says that the plan is OK if it allows distributions to participants who are over age 70-1/2 and still employed, without regard to whether or not they are required to take minimum distributions.

[Guest Sieve]

OK, got it. I can accept that this Q&A is aimed at permitting pension plan non-MRDs after that April 1, and that it does not address MRDs per se. Not how I read it at first, but I see it now.

[GMK]

A most interesting thread.

I'll bet BruceC (aka BruceM), the OP, is happy he's dealing with IRA's. (What was the question?...)

You say potato, and I say potato. You say MRD, and I say RMD.

Potato, potato. MRD, RMD. Think we should call the whole thing off?

(or is it just part of the ERISA Mash?)

[jevd]

A most interesting thread.

I'll bet BruceC (aka BruceM), the OP, is happy he's dealing with IRA's. (What was the question?...)

You say potato, and I say potato. You say MRD, and I say RMD.

Potato, potato. MRD, RMD. Think we should call the whole thing off?

(or is it just part of the ERISA Mash?)

As i stated in a previous post RMD MRD

IT DRM ( Doesn't Really Matter)

[Guest mjb]

Kevin C - I agree with Sieve (post #42) that the plan's operations are acceptable, because at the time it made the RMD's, the RMD's were required. There is no way (that I know of) for the plan to foresee that a participant will die, so the event of the death should not retroactively count against the plan's otherwise proper actions. Of course, someone may know of a case where it has.

mjb - Thank you very much for your comments in this thread. My concern remains with what the IRS would actually do. In the reference linked in post #40, Ms. Choate points out that the IRS issues PLR's that quote the IRS's gospel on beneficiary distributions and then rule exactly the opposite of the gospel they cite.

To rephrase your question in post #32,

Does anyone know of any recorded cases where the IRS has ruled to impose or not to impose a 50% excise tax for the failure to take an RMD for the year a participant in a qualified plan attained age 70 1/2 (or if the participant was still an employee at age 70 1/2, for the year the qualified plan participant retired) but died prior to her/his RBD?

How can the IRS impose the 50% excise tax on DC distributions when reg. 1.401(a)(9)-3 Q/A-1(a) explicitly states that if an employee dies before the employee's required begainning date (4/1 following year age 70 1/2 is attained according to -2 Q/A-2(a) unless the plan permits non 5% owners who continue active employment after age 70 1/2 to defer until the 4/1 following year of retirement) and thus before distributions over the employee's life commenced in accordance with 401(a)(9)(A)(ii), distributions will be made to a beneificiary as described in 409(a)(9)(B)(ii), (iii) or iv).

Further, reg -2 Q/A-6(a) states that distributions are not treated as having begun to the employee in accordance with IRC 409(a)(9)(A)(ii) (i.e., over the employee's lifetime) until the employee's required beginning date (see -2 Q/A-2(a) above) WITHOUT REGARD TO WHETHER PAYMENTS HAVE BEEN MADE BEFORE THAT DATE. "Thus if an employee begins receiving installment payments from a PS plan upon retirement at age 65 in the form of a joint and survivivor annuity, benefits are not treated as having begun in accordance with IRC 409(a)(9)(A)(ii) (i.e. over the employee's lifetime) until the April 1 following the year the employee attains age 70 1/2. The last sentence of subsection (a) of Q/A-6 states "This is the case without regard to whether the plan has distributed the minimum distribution for the first distribution calendar year (as defined in A-1(b) of 1.401(a)(9)-5) before the employee's death."

In both cases, those where the employee dies before April 1 of the year following the year age 70 1/2 is attained without taking the MRD and the case where an employee receives installment payouts for years after retirement but prior to April 1 of the year following the year age 70 1/2 is attained there is no required MRD under the 401(a)(9) regulations based upon the life of the employee, hence no 50% excise tax penalty will be applied.

[GMK]

mjb - I appreciate your thorough analysis. I'll say again that this has been a most interesting and informative thread.

At this point, I am just wondering if anyone knows of a recorded IRS ruling on this matter.

One could ask, "How can the IRS rule against its own regulations in the cases Ms. Choate describes in the link in post #40." But it did. As Ms. Choate writes, "But then, the IRS rules exactly the opposite of what its own regulation provides." And it did this not by ignoring those regulations, but after explicitly citing them in the PLR's.

[Guest mjb]

mjb - I appreciate your thorough analysis. I'll say again that this has been a most interesting and informative thread.

At this point, I am just wondering if anyone knows of a recorded IRS ruling on this matter.

One could ask, "How can the IRS rule against its own regulations in the cases Ms. Choate describes in the link in post #40." But it did. As Ms. Choate writes, "But then, the IRS rules exactly the opposite of what its own regulation provides." And it did this not by ignoring those regulations, but after explicitly citing them in the PLR's.

The problem with answering this thread is that several posters added extraneous information on the MRD regs which does not pertain to the question in the orginal post: whether MRDs are required to be taken if a participant or IRA owner dies prior to 4/1 of the year following the year age 70 1/2 is attained.

There is a substanial difference between the application of the MRD Regs to factual circumstances where death occurs before the required beginning date and the situations described by Choate regarding spousal rollovers where a trust or estate is designated as the IRA beneficiary. Where, under the facts described in the final MRD regulations -2 and -3, death occurs after age 70 1/2 is attained but prior to 4/1 of the following year, an MRD is never required and the IRS cannot enforce any penalty for the failure to take a MRD. There is no exception which allows the 50% excise tax to be imposed under those facts because the regs exempt a death under those circumstances from being subject to an MRD, unlike an IRA owner or participant who on dies or after 4/1 of the year following the year age 70 1/2 is attained where an MRD is required to be taken based upon that person's life expectancy by 12/31.

What Choate was describing are situtations where the MRD regs explicitly do not permit a rollover by a spouse because a trust or estate is the named beneficiary of the IRA instead of the spouse (i.e., the spouse is the beneficiary of the trust or estate, not the IRA). However, the IRS will allow a spousal rollover under a PLR if specific facts are present which would make the spouse the actual IRA beneficiary, e.g., such as when the spouse is both the sole beneficiary and sole trustee of the trust with the power to pay the entire amount to herself. In other words, under the situations described by Choate the spouse must present additional facts to the IRS which are not stated in the regs in order to allow a tax free rollover under a PLR, whereas satisfying the facts stated in regs 1.409(a)(9)-2 and 3 is sufficient to prevent a 50% excise tax penalty if an MRD is not taken.

In conclusion think about this: If the IRS could disregard its own final regulations and impose the 50% excise tax where no MRD is required how can taxpayers ever rely on any regulation issued under the IRC?

[Guest Sieve]

With all due respect, GMK, Natalie Choate is chastising the IRS for not having an official policy in the form of a Rev. Rul. since, on a regular basis for the last 17 years in PLRs, it has ruled FAVORABLY for taxpayers in spite of more restrictive regulations on the spousal IRA rollover issue. The IRS is, in fact, interpreting its regs literally in the PLRs, then turning around and saying that it will overlook those same regs in specific instances and rule in the taxpayer's favor. She is not complaining that there's an unexpected surprise that shocks the taxpayers in these PLRS, just that they should not have to go through the unnecessary expense and trauma of obtaining a PLR under such stressful circumstances when the IRS really ought to be officially memorializing its 17 years of rulings into something official.

Clearly, the government can change its own interpretation of its own regs. After all, regulations are nothing more than the official Treasury interpretation of the statutes, and, although given deference by the courts, regulations aren't accepted religiously by the courts. Personally, if you were a client, I would tell you not to worry about the issue in this thread--but, to the extent you are not as certain as I am about the outcome, I can understand your trepidation.

Personally, I know of no rulings on this issue. Tripoodi, however, would agree that death prior to RBD eliminates the need for any distributions to occur as a result of having reached 70-1/2. In fact, he indicates that distributions for the 1st or 2nd distribution calendar year which occurred before death could be rolled over as MRDs, even though they were not required to be distributed. (Tripodi, Chapter 6(VII)(F)(1) et seq, pp. 6.527-528.)

I know . . . this has been a most interesting and informative thread . . .

[Guest Sieve]

mjb -- Not sure what you mean by the following:

Where, under the facts described in the final MRD regulations -2 and -3, death occurs after age 70 1/2 is attained but prior to 4/1 of the following year, an MRD is never required and the IRS cannot enforce any penalty for the failure to take a MRD. There is no exception which allows the 50% excise tax to be imposed under those facts because the regs exempt a death under those circumstances from being subject to an MRD, unlike an IRA owner or participant who on dies or after 4/1 of the year following the year age 70 1/2 is attained where an MRD is required to be taken based upon that person's life expectancy by 12/31.

A death under the circumstances you suggest still imposes a requirement of MRDs for the beneficiary(ies) following the death--it just eliminates the participant's/owner's RBD and therefore any requirement that an MRD be paid to or on behalf of the particiapnt/owner. And, of course, the reason death after 4/1 mandates an MRD is because at least 1 MRD would have been required by that 4/1, so the death is AFTER RBD. But, then, maybe that's your point.

[GMK]

Thanks to all, especially mjb, Sieve, and JSimmons. I appreciate your patience with me.

I think I see how the rulings discussed in the Choate article may actually not contradict the regulations, and I am satisfied with that. In the trust case, the regs say that it is not enough for the spouse to be the sole beneficiary of the trust. The spouse must also have the unlimited right to withdraw the moneys. In the estate case, the PLR states, "Although not specifically stated in the regulations, a surviving spouse may not ..." So this is not contrary to the regs, but to some rule from somewhere else. In both cases, the PLR's do not contradict the regulations.

I am ready to say thank you and good bye on this, and re-read it, but one more question, please.

Sieve - In your comments from Tripoldi, you mentioned pre-death distributions that could be "rolled over as MRDs, even though they were not required to be distributed." Do you mean the distributions can be rolled over (because no MRD is required), or something else? I think that MRDs (and RMD's) cannot be rolled over.

[Guest Sieve]

Sorry--my editing caused my last sentence to change its meaning entirely. It was supposed to read: "In fact, he indicates that distributions for the 1st or 2nd distribution calendar year which occurred before death could be rolled over except to the extent considered as MRDs, even though they were not required to be distributed."

You're correct that MRDs cannot be rolled over, and that was precisely my point: even though there would be no RBD due to the death pre-April 1, the otherwise MRD (but for the death) is still considered an MRD (if actually distributed before death) and not eligible for rollover. So, the plan which allows no in-service distributions other than MRDs would not be violating the document in the event of a pre-April 1 death, because these pre-April 1 distributions would be considered MRDs to the extent they did not exceed the amount of the MRDs which were due had there been no death. (I haven't gone through the regs to find a cite for that proposition.)

[GMK]

Thanks, again. Now, go and vote ... if you want to (allowed but not required, I understand).

[Guest mjb]

Sorry--my editing caused my last sentence to change its meaning entirely. It was supposed to read: "In fact, he indicates that distributions for the 1st or 2nd distribution calendar year which occurred before death could be rolled over except to the extent considered as MRDs, even though they were not required to be distributed."

You're correct that MRDs cannot be rolled over, and that was precisely my point: even though there would be no RBD due to the death pre-April 1, the otherwise MRD (but for the death) is still considered an MRD (if actually distributed before death) and not eligible for rollover. So, the plan which allows no in-service distributions other than MRDs would not be violating the document in the event of a pre-April 1 death, because these pre-April 1 distributions would be considered MRDs to the extent they did not exceed the amount of the MRDs which were due had there been no death. (I haven't gone through the regs to find a cite for that proposition.)

Your last three posts are too badly garbled to merit a response. Where did I say that MRDs required for the year age 70 1/2 is attained cannot be rolled over if the IRA owner or partricipant dies prior to 4/1 of the year after age 70 1/2?

My reference to MRD not being required was intended to refer to MRDs required under the life expectancy rule applicable to the IRA owner or participant since reg -3 Q/A-1(a) explicitly states that such payments are not due if death occurs before 4/1 of the year following the year age 70 1/2 is attained and under that cited reg payments of the entire interest must be made to the beneficiary under IRC 401(a)(9)(B)(ii),(iii) or (iv) which means that earliest date payment can be required is Dec 31 of the year following the death of the IRA owner or participant.

One Q for you to respond to: why cant distributions not made during the first two distribution calendar years (the year that 70 1/2 is attained and the period from 1/1-3/31 of the following year) on account of death be rolled over since under reg. 1.401(a)(9)-3 Q/A-1(a) the entire interest of an IRA owner or participant who dies before the required beginning date must be distributed under the rules for beneficiaries including a spousal distribution which can be rolled over to an IRA as a payment to a surviving spouse under IRC 401(a)(9)(B)(iv) and reg. 1.402©-2 Q/A-12(a).

[Guest Sieve]

mjb --

I know why my last 3 posts are garbled--because you apparently can't read straight. Can I assume that if my posts do not merit a response, as you suggest, that your posted response has no merit?

1) You're preaching to the choir. Read the posts and you can see that I agree with you on this one--and I may also be the only one left on this thread who is willing to respond to your posts any further.

2) We've heard your arguments over & over--not just from you, but also from other posters (including, through my garbled posts, even me). Some agree with you, others do not. Let it go already.

3) I did not say that potential MRDs NOT distributed prior to a pre-April 1 death cannot be rolled over. What I said was: "even though there would be no RBD due to the death pre-April 1, the otherwise MRD (but for the death) is still considered an MRD (if actually distributed before death) and not eligible for rollover" (emphasis added). Notice that I say the potential MRDs cannot be rolled over IF ACTUALLY DISTRIBUTED BEFORE DEATH. I'm sincerely sorry if my choice of language and sentence structure were not clear enough for you.

By the way, if you care, my primary purpose for raising this issue was to address a concern (with which I disagree) that was expressed earlier in this thread--a thread which has now knitted its own sweater, by the way--that a plan which permits in-service distributions only for MRDs would be violating its terms if an age 70-1/2 owner died before RBD but after taking a distribution, because that distribution would somehow not be an MRD. I believe that such a distribution would, in fact, be considered to be an MRD (see, e.g., Treas. Reg. Section 1.402©-2, Q&A-7(a)). This was a side topic that you ignored, however, but I susect that's beside the point.

4) In response to your final question, your statement is an accurate one. It is not, however, as described above, contrary to anything in my prior posts (garbled or non-garbled).

5) For me, this thread has run its course unless something new is raised. I, for one, am not going to wait around until you have convinced the non-believers that your interpretation is the only correct one (whether or not it actually is). If you ever thought that that's the purpsoe of this Board, then think again.