The ILIT That Isn't Quite Right

The trust does not remove the policy from Harlan's estate. A trustee who can borrow against trust property and direct trust investments holds powers over the policy that constitute incidents of ownership under §2042. The proceeds would be included in Harlan's gross estate - the exact result the ILIT was built to avoid, and at $6.8 million of net worth the estate tax exposure on a $2 million policy is severe.

Fixes, in order of urgency:

• Resign as trustee and appoint an independent trustee with no beneficial interest and no relation to Harlan who could be influenced.
• Amend or decant the trust to strip the borrowing and investment-direction powers.
• Accept the three-year rule. Relinquishing incidents of ownership within three years of death pulls the proceeds back under §2035. Because the policy is in its first year, resigning now starts that clock rather than finishes it. The family should understand that if Harlan dies before the three years run, the ILIT fails and the estate pays.
• Stop informal administration. Premiums paid from Harlan's checking account must be documented as gifts to the trust, and the trust must be the applicant, owner, and beneficiary throughout.

A trust that "works on paper" but is run out of the grantor's checkbook is the most common ILIT failure, and the CPA's comment suggests nobody is watching the administration.

The conclusion is right by accident; the reasoning is wrong.

At $19,000 per donee, Harlan's exclusion covers $38,000 across two children. The remaining $36,000 is a taxable gift unless he elects gift splitting with Lena under §2513 - permitted if she consents, both are U.S. citizens or residents, and neither has remarried. Split, the exclusions double to $76,000, which covers the $74,000 premium with no taxable gift. But splitting requires filing Form 709 for the year, and the CPA's "no return is needed" answer would forfeit the election.

Mechanics that must be in place, or the exclusion fails and the gifts are taxable:

• Crummey withdrawal rights in the trust instrument for each beneficiary, exercisable for a reasonable window (typically 30 days).
• Actual notice to each beneficiary each year, with the amount and the deadline. Rights that are never noticed, or that are pre-waived by side agreement, are not present interests.
• Watch the lapse. A lapsing general power of withdrawal beyond the greater of $5,000 or 5% creates a taxable release - the standard 5-and-5 cap belongs in the drafting.
• Consistency: no contributions to the trust before the year of creation, and the trust should be funded with the policy at issue, not later.

No GST issue arises while the beneficiaries are children.

No - the product is wrong for the goal. A survivorship policy pays nothing at Harlan's death. If he dies first, the trust receives nothing, keeps paying premiums for years, and Lena has exactly the problem he was trying to solve. The proceeds arrive at her death and go to his children. Naming her a beneficiary of a second-to-die policy gives her a reminder in the trust document, not money.

There is a second, sharper problem: who funds the policy after Harlan dies? The $74,000 annual premium came from him. If Lena is expected to contribute after his death and is not a donor to the trust, the policy lapses into exactly the moment it was meant to perform.

What actually serves the goal:

• Keep the survivorship policy for the estate tax and liquidity purpose it is genuinely good at, and be honest that this is an estate-planning tool, not a support tool.
• Fund Lena's access to the house separately - a life-estate or right-of-occupancy provision in Harlan's revocable trust, a QPRT, or a single-life policy owned by an ILIT with Lena as beneficiary.
• Budget the premium from a dedicated source and address the post-death funding question in writing today.

Adding Lena to this trust should not be done casually either: it changes the beneficiary structure his attorney drafted and can affect how later generations and any GST exemption are allocated.