Multiple beneficiaries after Required beginning date

[Guest anthony devito]

Here is situation. Account owner of Traditional IRA dies after required beginning date (80 years of age), December 4, 2000. Two sons are joint beneficiaries, 50% each. They would like to separate the IRAs into two IRAs for several reasons including to give flexibility of one to take cash and other to get stretch. Also if the IRAs are able to be separated, do they each use own life expectancy or the life expectancy of the older of the two sons?

[KJohnson]

You can defintely divide it into two IRAs but put you need to keep each IRA in the name of XXX, deceased (father), for the benefit of YYYY (son 1), and XXX deceased (father) for the benefit of ZZZ (son 2).

Whether they can each use their own life expectancy is a closer question. The regs require that you must use the life expectancy of the eldest unless there were "separate accounts" in the father's IRA for each beneficiary. However, in a private letter ruling late last year, the IRS appeared to take a fairly broad view of what could consitutute separate accounts.

You can't rely on a PLR so unless you are confident that there were separate accounts or unless you want to get your own PLR, I would use the life expectancy of the eldest for both.

I don't think the new proposed regulations alter this analysis, but I have not gone back and looked at them on this point.

[Mary Kay Foss]

I agree with the previous response.

Each custodian has their own agreement which must be followed. If the custodian does not allow separate accounts, the life expectancy of the oldest beneficiary must be used. You may have to move to another custodian just to get the split.

[Appleby]

This separation of account is just for reporting and self-direction of investments purposes only. Some custodians need to separate the account to handle the individual tax reporting. According to the old regulations if the participant dies after the required beginning date, EVEN if the accounts were separated, the life expectancy of the oldest beneficiary must be used to figure post death distributions. The PLRs that were issued that allowed the beneficiaries to use their own life expectancy were based on the decedent dying BEFORE his required beginning date. Because the deceased was over the age of 70 1/2, some misinterpreted it to mean that he was past his required beginning date. However, he inherited the IRA from his wife. He took her RMD and moved the assets to his OWN IRA the next year. His required beginning date would have been the April 1 following the year he took the assets as his own, but he died before that April 1 date. Therefore, the IRS treated him as dying before his required beginning date and allowed the beneficiaries of his IRA ( that he inherited from his wife and treated as his own) to use their own individual life expectancies.

In summary, if the participant died after the required beginning date, the life expectancy of the oldest beneficiary must be used to determine life expectancy factors.

The new regulations - issued January 12, addressed the separate IRA issue for the first time,. It states that if the assets are segregated into separate account by December 31 of the year following the year the participant dies, then each beneficiary would be allowed to use their own life expectancy.

The only issue now is that the new regs do not seem to be retroactive, as least that has not been explicitly stated. However, some experts have determined that all IRA owners and beneficiary may start using the new rules. This means that if the beneficiaries separate the assets by December 31,2001, they may be able to use their own life expectancies.

I would like to add also that Kjohn is right about the titling of the inherited IRAs- they should include the name of the deceased, with DEC'D by his name and the name of the beneficiary, with Bene by his name or some othe titling to show who is the deceased and who is the bene- The social security number used for tax reporting should be the beneficiary's

[KJohnson]

I thought that under the old regs 1.401(a)(9)-1 Q&A H-2 the separate account rule also applied to distributions to beneficiaries for death after a participant's required beginning date as long as the accounts were separate as of the required beginning date.

Then in PLRs 2000-52044, 2000-52043, 2000-52042 the IRS seemed to imply that simply naming separate beneficiaries as of the RBD was good enough to satisfy the separate account rule as long as each beneficiary received a proportianate share of gains or losses.

In these PLRs the holder of the IRA was 72 or 73 at the time of her death in 1999 and she had begun receiving required minimum distributions. It does not appear that she had inherited the IRA (or at least not recently) since she named her beneficiares back in 1991 (before her RBD).

[Appleby]

KJohnson,

You are right- in fact we both are.

the difference is- you are addressing an issue where the participant had seperate ( more than one ) IRAs and named a different beneficiary for each IRA. In this case, as each beneficaiy was in fact the oldest and only primary beneficiary of the IRA, as at the participant's required beginning date, then each beneficairy was allowed to use their own life expectancy.

I am addressing an issue where there was only ONE IRA, with two primary beneficiaries. My response was targeted to anthony devito's situation.

[KJohnson]

In the PLRs dited above there was only one IRA, with three primary beneficiaries of the single IRA. The beneficiares were then allowed after death and post-RBD to divide the IRA three ways with each beneficiary using their own life expectancy. I had always thought that using separate IRAs with separate beneficiaries was the way to go, but these PLRs call into question whether that is really required. Then again, they are only PLRs.

I did go back and look at the proposed regs and it seems like the separate account rule only needs to be met at the end of the year following the year of death in the case of distributions made prior to RBD. However, for death distributions after RBD it appears that the separate account rule still must be satisfied at the RBD. 1.409(a)-8 Q&A 2. I don't think that 1.409(a)-8 Q&A 3 provides any more guidance as to what constitutes a "separate account" beyond what was in the prior proposed regulations.

So, I guess I am back to where I was before. Unless you are confident that that there were "separate accounts" as of the required beginning date or unless you want to get a PLR, I would use the shortest life expectancy even after the IRA is "divided"

[Guest anthony devito]

I'm impressed by the sophistication of the discussion, but I'm also getting a little confused here.

Let me restate the situation and my understanding of the resolution based upon the discussion that has taken place.

Wife dies in 1995. Surviving husband and sole primary beneficiary converts wife's IRA into his own. (I doubt that this fact alters the situation, but I mention it just in case.) There is ONE IRA in the father's name and the two sons are primary beneficiaries. Father dies December 4, 2000 at the age of 80, long after RBD.

Everybody seems to agree that the IRA can be separated into two IRAs--one for each of the two sons--provided this is accomplished before December 31, 2001. The IRAs must be set up as beneficiary IRAs. The way in which the account is titled is important.

The assets were not segregated into separate accounts before death or before the RBD. To be safe,it seems that each of the sons must use the life expectancy of the older son for the purpose of computing the MDR. Using the respective life expectancies for the purpose of computing the MDR is based upon private letter rulings and may be problematic unless heirs wish to obtain their own PLR.

Is that a fair and accurate summary or did I miss something?

Thanks for all the input. I do mostly investment management and IRA distribution requirements is not my area of expertise although I do enjoy it.

[KJohnson]

I think you've got it right (at least as I understand the proposed regulations).

[KJohnson]

Not to complicate things, but here is an excerpt from a long article that Noel C. Ice published on benefitslink today on the new proposed regs. His view is, apparently, that beneficiary designations, without more, do create separate accounts. While I won't speak for him, it seems that from his comments he would not think that you need a PLR.

9.2 Separate Accounts.

Q&A 2 tells us that if benefits under an IRA or defined contribution plan are divided into separate accounts (for which you may always substitute the term “segregated share” in the case of a defined benefit plan), then the MRD rules apply separately with respect to each separate account. So what exactly is a separate account? Q&A 3 tells us that “a separate account in an individual account is a portion of an employee's benefit determined by an acceptable separate accounting including allocating investment gains and losses, and contributions and forfeitures, on a pro rata basis in a reasonable and consistent matter between such portion and any other benefits.”

So, how hard is that? If P designates A, B and C as death beneficiaries, do we have separate accounts without more? Of course we do, unless you can tell me that what B takes out will directly reduce what A and C get. Think about it. Is there any state in the union where if B gets to the bank first and says, give me more than one-third, the bank will give it to him, without recourse? If B takes one-third and A and C do nothing, will C continue to be allocated a share of gains and losses? Of course not. As far as I am concerned, it is very hard not to have a separate account. Perhaps the safest course would be to have the account physically divided after P’s death, but this really should not be necessary.

What if P’s beneficiary designation says give D $100, and divide the rest between A, B and C? I think that this too creates separate accounts, but some would object that it does not, since D would not ordinarily be entitled to investment gains and losses. My own feeling it that this is quibbling (but, to be safe, I would concur in recommending against doing it this way). I think that the phrase “including allocating investment gains and losses” modifies “acceptable separate accounting,” as does “contributions and forfeitures” and that if allocating gains on pecuniary gifts is not an acceptable accounting practice, then one is not required to allocate them. In other words, I don’t think that allocating gains is a separate requirement applicable to pecuniary gifts, any more than a provision for allocating forfeitures is required if there cannot be any.