Spousal Lifetime Access Trusts (SLATs)

A Spousal Lifetime Access Trust (SLAT) is an irrevocable trust created by one spouse (the grantor) for the benefit of the other (the beneficiary spouse) and their descendants. By utilizing your lifetime gift tax exemption ($15 million in 2026), you remove assets—and their future appreciation—from your taxable estate while maintaining indirect access to the capital through your spouse’s distributions.

The structural mechanics and advantages of a SLAT include:

However, SLATs carry four major planning risks:

Divorce Risk

If you divorce, your indirect access to the trust assets terminates immediately. Furthermore, unless the trust agreement contains a “floating spouse” clause—which defines the beneficiary spouse as the person to whom you are married at the time of distribution—your ex-spouse will continue to benefit from your trust.

Premature Death

If your spouse dies before you, your indirect access to the trust assets ends. You can mitigate this risk by purchasing a second-to-die life insurance policy or a policy on the beneficiary spouse’s life.

The Reciprocal Trust Doctrine

If you and your spouse establish substantially identical SLATs for each other, the IRS uncrosses them and both sets of assets return to both gross estates. This is the trap that destroys the most SLAT plans, and it is treated in detail below.

Community Property Funding Trap

In community property states like California, SLATs must be funded solely with the grantor’s separate property. If you fund a SLAT using community property, the beneficiary spouse is deemed to have made a gift of their 50% share to the trust, which pulls that 50% back into their gross estate under IRC §2036 upon death, destroying the tax exclusion.

The Reciprocal Trust Doctrine

Every couple that hears about SLATs immediately proposes the obvious thing: each spouse creates one for the other, both exemptions get used, and everybody keeps indirect access to everything. Do that symmetrically and you have built nothing. Under United States v. Estate of Grace, 395 U.S. 316 (1969), the government treats each spouse as having created the trust that names them as beneficiary, which makes it a transfer with a retained life interest and pulls the whole thing back under IRC §2036.

The test is objective, which is the part people misread. Grace deliberately discarded the older subjective standard that asked whether the parties intended to avoid estate tax. Two prongs, both factual: the trusts are interrelated, and their arrangement leaves the grantors in approximately the same economic position as if each had created a trust naming himself as life beneficiary. Nothing turns on motive. “We each had independent reasons” is not a defense, and neither is the fact that your attorney never discussed the doctrine—the examiner is comparing two documents, not attempting to read minds.

Understand what losing costs, because it is worse than it first sounds. The assets come back into your gross estate at date-of-death value, so every dollar of appreciation you spent a decade moving comes back with them. The trusts stay irrevocable. You do not get the property back, you do not get the flexibility back, and you have paid for drafting, trustee fees, and annual returns for the privilege. You keep every cost of the structure and lose its only benefit.

What does not save you. Practitioners routinely paper over reciprocity with differences that are real on the page and worth nothing economically. Treat all of these as cosmetic:

None of these change where the economic benefit lands, which is the only question Grace asks.

What actually works. The differences have to change who gets what, or who controls it. The strongest designs make the two trusts genuinely asymmetric:

Different beneficiary classes.

The cleanest break. One trust benefits your spouse and descendants; the other benefits descendants only, with the spouse excluded entirely. If only one spouse is ever a beneficiary, the trusts cannot leave both in the same position.

Different powers of appointment.

Grant the beneficiary spouse of one trust a lifetime limited power of appointment and give the other spouse none, or only a testamentary power. This is the difference that carried the day in Estate of Levy, T.C. Memo. 1983-453, and it remains the most-cited single distinguishing feature.

Different distribution standards.

One trust limited to HEMS with an independent trustee; the other fully discretionary, or mandatory-income, or with a five-and-five withdrawal right.

Different time horizons.

One terminates and vests at a stated age; the other continues as a perpetual dynasty trust with allocated GST exemption (section “Dynasty Trusts”).

Materially different size.

Not $7.5 million and $7.4 million—different by enough that the economic positions plainly diverge.

Rank the designs, then pick the highest one you can live with.

1.
One SLAT, not two. One spouse funds it; the other keeps their exemption and relies on portability at death (section “Portability and the DSUE”). No reciprocity is possible, and you still move the appreciation on half the exemption. For most couples this is the right answer, and it is the one nobody proposes because it feels like leaving money on the table.
2.
Two different structures. One spouse creates a SLAT; the other does a sale to an IDGT (section “Intentionally Defective Grantor Trusts (IDGTs)”), a GRAT, or a descendants-only dynasty trust. Structures that are not the same kind of thing cannot be uncrossed into each other.
3.
Two genuinely asymmetric SLATs, using at least two differences from the list above and separated by more than a year. This is the aggressive version. It is defensible, and it is litigated.

Build the file now, because the audit is in forty years. The doctrine surfaces on examination of the second estate, when both spouses are dead and an examiner reads two trust instruments side by side. Nobody will be available to explain the plan. Have counsel write a contemporaneous memo, at drafting, setting out the specific economic differences and the non-tax reasons for each, and keep it with the trust originals. Note also that the doctrine is not about marriage—it reaches any interrelated cross-trusts, including those between parent and child or between siblings, and it applies under IRC §2038 as readily as IRC §2036.