One of the most fundamental rules of minimum distribution planning is to spread out the receipt of retirement funds–and thus the taxation of them–for as long as possible. Final regulations under Section 401(a)(9) simplified the ways of achieving this objective. But a private letter ruling issued recently defied the logic of the regulations when it comes to determining the life expectancy of a trust beneficiary.

The ruling involved a grandmother who had set up a trust to provide support for her two grandchildren, and had named the trust as beneficiary of her IRAs. In the event of her death, the trust would support each of the two grandchildren (ages 5 and 6) until they reached age 30, at which time the trust would terminate. The trust included a provision that in the remote event that both children died before age 30 without leaving any children of their own, the proceeds were to go to an elderly great-uncle, who was 67.

The grandchildren were ages 5 and 6 when their grandmother died. The advisors involved in making the payout arrangements requested a letter ruling as to the life expectancy requirements for the IRA distributions–not surprising since three different sets of regulations have affected this area in the past two years.

When a trust is named as beneficiary of an IRA, the general rule is that as long as four basic requirements are met, the life expectancy of the individual beneficiary of a trust can be used as the measuring life for RMD purposes. The grandmothers trust met all of these requirements. (Generally, these four requirements are: (1) the trust must be valid under state law, (2) the trust must be irrevocable or become irrevocable at death, (3) the beneficiaries of the trust must be identifiable from the trust instrument, and (4) certain documentation must be provided to the IRA trustee as to the identity of the trust beneficiaries.)

The rule for trusts like this one is that if there is more than one beneficiary, the life expectancy of the oldest beneficiary must be used as the measuring life for calculating distributions. The regulations add that even a beneficiary whose entitlement to the benefit is contingent will be considered for this purpose. However, they add that a successor beneficiary (one who has a beneficial interest only because the person could become the successor to the interest of a beneficiary after his or her death) will not be considered for purposes of determining the measuring life.

In an ill-considered twist of logic, the Service determined that the 67-year-old great-uncle was indeed a contingent beneficiary in this situation, not a successor beneficiary. As a result, payouts had to be based on his much-shorter life expectancy rather than that of the two children. This dramatically shortened the degree of tax deferral that should otherwise have been possible.

Although this ruling was issued on the heels of the new 2002 regulations, it was apparently based on the 2001 regulations. However, the principles on which it is based have not substantially changed from the 1987 or 2001 regulations; the regulations have always said that when there are multiple beneficiaries, the life expectancy of the oldest is used. But to view this great-uncles interest as anything other than a remote successor interest is illogical and inconsistent with the apparent intent of the drafters of the regulations.

Probably the best way to avoid an outcome like this is to avoid naming the older contingent beneficiary at all in the trust instrument, and to make any needed provisions for an older beneficiary through a separate account or other trust planning tools. (The number of this ruling is Letter Ruling 200228025.)

April K. Caudill, J.D., CLU, ChFC, is managing editor of Tax Facts and ASRS, publications of the National Underwriter Company.


Reproduced from National Underwriter Life & Health/Financial Services Edition, September 16, 2002. Copyright 2002 by The National Underwriter Company in the serial publication. All rights reserved.Copyright in this article as an independent work may be held by the author.