72(t) with multiple IRA accounts
A financial planner I know contacted me yesterday with this question. He just acquired a new client and in reviewing her file he noticed that she is 2 years into a 72(t) distribution. The problem is that she has 2 IRA accounts and when the previous planner figured out the 72(t) he apparently calculated the amount based on the value in the 2 accounts combined but the actual distributions are only coming out of one of them. One of the IRAs is a standard, brokerage-account based account holding mutual funds, and the other is an IRA annuity. The planner did not tell me which one is making the distributions.
Questions – is this a problem, and if so, what would be the recommended fix?
Margaret A. Stallworthy, CPA
[email protected]
Permalink Submitted by Alan Spross on Fri, 2009-12-04 16:49
It is not a problem if done correctly. Multiple IRA accounts can be aggregated for 72t distributions in the same manner as they can be aggregated for RMDs. When an annuity IRA is included in the initial 72t balance, it is fairly commonplace to set up the distributions from the other IRA account. There will be only one 1099R, and it will probably be coded as an early distribution, so a 5329 will have to be added to the tax return to claim the 72t exception.
Since both accounts are part of the 72t plan, there also is no problem if later on the distributions come from the annuity or part from each account. I might be concerned that the client did not reveal the 72t plan themself and the planner had to discover it in the documentation provided. The planner should establish a file holding the original 72t calculations and assumptions (interest rate, copies of statements showing account balances, age and individual vrs joint life expectancies). If there was an error, depending on the nature of it, some things can be corrected prior to year end.
Permalink Submitted by Margaret Stallworthy on Fri, 2009-12-04 16:53
I didn’t mean to imply that the client didn’t tell the planner about it – she did. Guess I didn’t phrase that very well.
But thank you for the information – I will forward it to him and everybody will breathe a little easier.
Margaret Stallworthy
Permalink Submitted by Margaret Stallworthy on Mon, 2009-12-07 16:36
The financial planner who posed this question to me in the first place says that the back-room people at Raymond James are insistent that this is not correct and there is a big problem.
It would not be the first time they were wrong about something. But the only way to convince them is going to be to point them to specifice IRS code or regs to prove it to them.
Alan – would you mind very much pointing me to the citations so I can send him back with something concrete to show them? Thanks a bunch.
Margaret A. Stallworthy, CPA
Permalink Submitted by Alan Spross on Tue, 2009-12-08 02:52
Unfortuneately, the tax code is limited to only a brief paragraph on substantially equal payments. The vast majority of regulation interpretations are contained in IRS Notice 89-25 and Revenue Ruling 2002-62. Beyond that there are dozens of private letter rulings interpreting what is permissible and what busts a 72t plan. I have never seen the IRS challenge the aggregated use of IRA accounts in what has become known unofficially as the “SEPP universe”.
Perhaps the closest indication is the following copied from Notice 89-25 indicating that 401(a) 9 account value determination is what applies for 72t plans. There are regulations that RMDs can be aggregated and I doubt if James is asking for those. However, in the final analysis the validity of a 72t plan is between the IRS and the taxpayer, not the IRA custodian. Most custodians now limit their support for these plans and code all distributions as early, forcing the taxpayer to claim the exception on Form 5329. So, while the following may not convince the James people, they really do not have to be convinced and the taxpayer would just continue to take the correct distribution and file the 5329. Here is the paragraph from 89-25:
>>>>>>>>>>>>>>>>>>>>
Payments shall be treated as satisfying section 72(t)(2)(A)(iv) if the annual payment is determined using a method that would be acceptable for purposes of calculating the minimum distribution required under section 401(a)(9). For this purpose, the payment may be determined based on the life expectancy of the employee or the joint life and last survivor expectancy of the employee and beneficiary.
>>>>>>>>>>>>>>>>>>>
Permalink Submitted by Margaret Stallworthy on Tue, 2009-12-08 03:51
Wow – thanks for the detailed answer.
I also found IRC Sect. 408(d)(2)(A) and (B) which clearly state that for purposes of distributions under Section 72 all IRAs are considered to be one account and all distributions during the year are considered to be one distribution. That seems to be helpful to the question of whether it is OK to take the distributions from only one of the two accounts.
The financial planner called back. He found somebody more knowledgeable at Raymond James (imagine that) who agreed that given that this is a client coming into the firm with the 72(t) amount already calculated, they will not insist that it is wrong to do it the way it was done. The story is that it is Raymond James’ policy not to allow aggregation of IRAs for the purpose of figuring the 72(t) amount whenever one or more of the accounts is not held directly at Raymond James. Their reason is that it is “too difficult” to determine the proper valuation. So their problem has nothing to do with IRS code or regulations, just with their own self-preservation.
They are covering themselves by saying they will not rely on a valuation provided by another firm. But given that the valuation has already been done, they have no objection to continuing the distributions from only the brokerage IRA.
Something tells me that the fact that they would not acquire the client if they persisted in objecting to the arrangement had a little bit to do with it.
I work with a lot of financial planners with a lot of different firms. Most of the time they are OK. Some days they REALLY irritate me. This is one of those days. Oh well – I’ll chock it up to a learning experience. I now know something about 72(t) distributions that I didn’t know before.
Best regards,
Margaret Stallworthy, CPA