How to caluculate life expectancy of spouse if not sole beneficiary.

[Guest ElizabethH]

If spouse is one of multiple beneficiaries, but whose life expectancy is the shortest for purposes of determining the applicable distribution period following IRA's owner's death, how is spouse's life expectancy calcuated? Is life expectancy determined in year following death and then reduced by one for each year thereafter or is special spousal recalculation applied?

[BPickerCPA]

If the spouse is not the sole bene, then the calculation is just like any other bene.

[Appleby]

My explanations are in accordance with the new RMD rules.

If the spouse is not the sole primary beneficiary, then the non-recalculation method must be used to determine post death distributions.

The life expectancy used is that of the oldest beneficiary, however, each beneficiary may be able to use his or her own life expectancies if the following occurs:

If the spouse is not the sole primary beneficiary, then the non-recalculation method must be used to determine post death distributions.

The life expectancy used is that of the oldest beneficiary, however, each beneficiary may be able to use his or her own life expectancies if the following occurs:

1) If the oldest beneficiary disclaims his/her portion of the inherited assets by December 31 of the year following the year the IRA holder dies, he/she will be treated as not having been a beneficiary for purposes of calculating post death distributions. The next oldest beneficiary's ;life expectancy would be used unless the assets are separated into individual beneficiary accounts by December 31 of the year following the year the IRA holder dies

2) If the oldest beneficiary takes a full distribution of his/her portion of the inherited assets by December 31 of the year following the year the IRA holder dies, the same rules (as above applies)

3) If each beneficiary separates the inherited assets into individual inherited IRAs by December 31 of the year following the year the IRA holder dies, they will each be allowed to use their own life expectancies

[Guest fran krawiec]

Are you saying that even if IRA is split into separate accounts by end of year following death, that spouse's share must be based on his/her S.L.E., nonrecalculated? I believe that if spouse is sole beneficiary of his/her separate share, such share's RMD would be based on spouse's recalculated S.L.E.

[Appleby]

If the spouse is the sole beneficiary and elects not to treat the IRA as his/her own, then the spouse must use his/her single life expectancy on a recalculated basis.

The option to recalculate is only available if the spouse is the sole primary beneficiary.

Seperating the assets into seperate beneficiary accounts does not make the spouse the sole beneficairy, but only allows him/her to use his/her own life expectancy. For IRAs with multiple beneficiaries, even if accounts are seperated by the deadline, the spouse must use the nonrecalculation method.

I hope this makes sense.

[Guest fran krawiec]

According to 1.401(a)(9)-8 A-2(B), the rules of 401(a)(9) apply separately to separate accounts. Thus, if the spouse is the sole beneficiary of this separate account, the 401(a)(9) rules would apply separately to this separate account. I see no basis in the new proposed regs for treating the account separately for the purpose of determining the designated beneficiary, but not separately for the purpose of determining the life expectancy method. Therefore, I believe such spouse's life expectancy would be recalculated.

[Appleby]

fran krawiec,

I agree with you- but I think we are mixing concepts here.

Your explanation is for a participant who sets up separate IRAs and name separate beneficiaries for each IRA. In your example, the spouse would be the sole beneficiary of the IRA in question and must use recalculation.

The discussion I responded to is based on one IRA, with multiple primary beneficiaries, one of which is a spouse. The regs are very clear on this. For an IRA with multiple beneficiaries, the spouse must use the rules that apply to a non-spouse beneficiary

[BPickerCPA]

Appleby,

I have to disagree with you.

If a single IRA is split after death and before 12/31 of the year after the year of the account holder's death, it is considered as if it were separate IRAs.

Therefore the IRA of which the spouse is the sole beneficiary will compute spouse-as-beneficiary distributions using the spouse's recalculated life expectancy.

[Appleby]

Barry and Fran ,

Sorry for the delayed response-

It seems you are right!!!- I concede

Appleby

[BPickerCPA]

Request to concede denied....

Concession indicates a victor and a loser. The purpose here is to determine what the rules are and how to use them to advantage, and avoid the pitfalls.

When we do that through discussion, there are only victors, so NO ONE needs to concede.

Barry

[Appleby]

Thanks my good Sir. As usual, you are right.

[Guest fran krawiec]

Appleby & Barry, thanks for your input. keep up the good work.

[Appleby]

You're welcome-

Check out Notice 2001-23 in this IRB.(2001-10) http://www.ppcnet.com/news/articles/tax/20.../article372.htm

An article - which can be found at http://www.ppcnet.com/news/articles/tax/20.../article372.htm, addressed this Notice as well as the other new RMD rules. One thing I found particularly interesting is the following statement

" Basically, a qualified plan that is not amended will be required to make MRDs for 2001 based on the old rules (i.e., it must follow the terms of plan in making distributions to the plan participants). However, the distributee participant can then recompute his or her MRD using the new rules. The difference between the recomputed MRD under the new rules and the amount actually distributed can be rolled over to an IRA. In short, a plan participant who receives a distribution exceeding his or her MRD for 2001 under the new rules can avoid tax on the excess amount by rolling it over into an IRA within 60 days of the distribution."

[KJohnson]

Barry,

Would your response regarding the timing of separrate accounts be the same if the death ocurred after the required beginning date?

It seems 1.401(a)(9)-8 Q&A 2, the segregation has to be by the employee's required beginning date if separate accounts are going to be used for a death post-RBD.

[Guest fran krawiec]

I believe 1.401(a)(9)-8, Q-2 states that separate accounting for lifetime distributions must be determined at RBD, for death distributions, separate accounting must be established by 12-31 of year following year of death.

[KJohnson]

The sentence regarding segregation into separate accounts by the end of the year following the year of the employee's death refers to 401(a)(9)(B)(ii)or (iii) or (iv).

I didn't see anything on 401(a)(9)(B)(i).

That, combined with the first sentence regarding lifetime distributions, is why I thought that separate accounts must be established by the employee's RBD if you were to use separate accounts for death distributions after the employee's RBD.

[BPickerCPA]

My discussions with people in Washington indicate that separate accounts can be done after death, up to the 12/31 of the following year. The only time that you would need separate accounts during lifetime is if the spouse more than ten years younger is to be the sole beneficiary of one specific account.

There is a problem with the wording in the regs, and my comment letter requested that this be fixed.

[KJohnson]

Thanks for trying to get this clarified with the folks in D.C.,

Did you also raise the issue of distinguishing between a rollover and a spouse treating an IRA as his or her own with regard to when an IRA beneficiary is an estate or trust and the corresponding beneficiary to the estate or trust is the spouse under Q&A 5 of 1.408.8?

If you did, thanks again.

[BPickerCPA]

The situation you described, the estate or trust is the bene of the IRA and the spouse is the bene of the estate or trust, means that someone screwed up.

When I raised the question with IRS in Washington I was told that the rules have not changed, which means that it's a facts and circumstance issue to be decided on a case by case basis.

I didn't see any way to resolve it by regulation, since I didn't really see an open question.

Barry

[KJohnson]

It seems to be one of the things that people screw up the most. Q&A 5 just raised the question in my mind whether the "fix" found in repeated PLRs is still available.