Substantially Equal Payments to avoid 10% penalty

[MarZDoates]

Client is age 58 and wants to take $2,000 per month out of one of his IRAs beginning in 2002. In order to avoid the 10% penalty he will have to take distributions that are part of a "series of substantially equal periodic payments" based on his life expectancy.....AND has to continue to receive these payments for at least five years, right? Okay....

Question 1: Can we use old life expectancy tables or must we use new tables?

Question 2: Trying to determine which of his IRAs to take distribution from. His values change daily...so what date do I use to determine the value?

Example: $24,000 x 25.9 (old table) = $621,000 is how much he would need in his IRA to get him the $2,000 per month. Do I use values af of 12/31/01 or "current" value?

Any input is GREATLY appreciated.

[Belgarath]

MarZ - re your first question - I had the same question myself. It's interesting, because someone at the IRS, (maybe Marjorie Hoffman?) had stated that she didn't believe the new uniform tables from the Jan 2001 tables could be used for a 72(t) distribution, because they were for minimum distributions only.

However, the preamble to the new 401(a)(9) regs makes it very clear that the new tables can be used. The question is whether they MUST be used, or if you start with the old tables, must you change to the new in 2003? It matters, because you can get a larger distribution under the old tables. And I don't feel like any clear guidance exists at this point. My gut feeling would be that since in 2002 you can use old regs, 2001 regs, or 2002 regs, that you can use the old table. And you wouldn't automatically have to update. But to be very safe, you would update and use new table in 2003.

I'd be interested in hearing from anybody with contacts at the IRS as to what they think on this.

[Appleby]

Belgarath

According to the Final RMD regulations, he may use the new tables beginning this year. Next year, the new tables are mandatory.

Even if he was already using the old tables, he is permitted to transition to the new tables. According to the final RMD regulation, this will not be considered a modification of the 72(t) payments.

MarZDoates

The value used depends on whether or not you are basing his calculations on a fiscal or calendar year.

For example, if he is starting now (June) you may assume that his year starts June ( fiscal year) and use his market value of the IRA as of month end May. That’s one option.

Another option is to use the market value of the account as of 12/31/2002. This option is used if the distributions will be taken on a calendar year basis. According to PLRs issued by the IRS, someone who starts their 72(t) in mid-year, and will take payments on a calendar year basis, has two options for the first year.

1. Play catch-up and take the amounts that would have been distributed for (in your case) January to May, has he started in January - in other words, the full annual amount is taken for this year and subsequent years

Or

2. Divide the annual amount by 12, and take only the amounts due for June to December. The annual amount for the first year will be less than that for other years, but the IRS indicated that this would be a permissible exception. The full annual amount must be taken for subsequent years.

Should you determine the amount you need to have in the IRA to meet the annual payments, you may transfer such amount to a separate IRA and take the payments from this IRA.. The amount transferred to the IRA would be your market value.

Bear in mind also that the life expectancy method is not the only method of choice.

He may also use the amortization or annuitization method. These two methods give larger annual calcualted amounts, than the life expectancy method.

[Belgarath]

Appleby - thanks for the response. Just to clarify - are you saying that if you start with the old tables this year, that you MUST switch to the new tables in 2003? The new regs say you are permitted to change, but I'm not sure they say you HAVE to, although that may well be what was intended. Thanks. (It's an awfully nitpicky little thing, but may be a hassle for folks who have been doing it on their own, if they HAVE to change, and don't know it!)

[Appleby]

Good point,

The exact wording is

“These new tables also may be used to determine an employee's (or IRA owner's) life expectancy, or the joint life and last survivor expectancy of an employee (or IRA owner) and designated beneficiary, for purposes of calculating the amount of substantially equal periodic payments under section 72(t)(2)(A)(iv)…”

This would suggest that it is optional.

[MarZDoates]

Kewl Beans. Thanks for your help!