Walton GRATs in a Low §7520 Rate Environment: When to Lock In and How to Structure

Why the §7520 rate is the only number that matters

A Grantor Retained Annuity Trust works on a single arithmetic premise: the grantor transfers an asset into the trust, retains a fixed annuity for a term of years, and the remainder passes to the beneficiaries at the end of the term. The IRS values the gift by subtracting the present value of the retained annuity from the value of the transferred asset. If the annuity is sized correctly, the gift is zero.

The discount rate the IRS uses for that present-value calculation is the §7520 rate — 120% of the federal mid-term rate, published monthly by the IRS. Every GRAT created in a given month locks in that month's rate as the hurdle the trust assets need to beat. If the assets grow faster than the §7520 rate, the excess passes to beneficiaries gift-tax-free. If they grow slower, the GRAT simply returns the original asset to the grantor in installments — no harm done.

The math is asymmetric: the downside is zero (you get your stuff back), the upside is the entire appreciation above the hurdle. In a low-rate environment, the hurdle is easy to clear — which is why GRATs become much more popular when rates dip.

The Walton case and the zero-out

Before Walton v. Commissioner, 115 T.C. 589 (2000), the IRS's position was that the GRAT's retained-annuity value had to be calculated as if the annuity terminated on the grantor's death (whichever came first — the term or death). That position made it impossible to zero-out the gift because the contingent termination reduced the annuity's present value.

The Tax Court rejected the IRS's reading. The court held that a GRAT can be structured with a term-certain annuity — meaning the annuity pays for the full term regardless of whether the grantor lives — and that the full term-certain present value can be used in the zero-out calculation. The IRS acquiesced in Notice 2003-72. Treas. Reg. § 25.2702-3 was amended accordingly.

The practical drafting consequence: every modern GRAT includes language that the annuity payments continue to the grantor's estate if the grantor dies during the term. The trustee pays the grantor (or the grantor's estate) on the schedule regardless. This is the Walton structure, and it's what makes the zero-out gift possible.

Structuring decisions: term, annuity, growth assumption

Three structuring choices drive every GRAT's economics:

Term length

• 2-year (rolling). The standard. Short enough that mortality risk is low. Permits a series of rolling GRATs — at the end of each 2-year term, the returned annuity payments are immediately re-GRAT'd. Each new GRAT locks in that month's §7520 rate and the then-current asset value. This is the textbook approach for liquid concentrated equity.
• 5–10 year. Appropriate for assets with a clearer long-tail appreciation thesis — pre-IPO private company stock, real estate held for development, a closely-held business with a known liquidity event timeline. Mortality risk is meaningful; use only when the grantor has solid actuarial expectancy.
• Long-term (15–20 year). Rare. Locks in the current rate for the full period. The mortality risk is substantial and the §7520 rate may move favorably during the period anyway. Generally avoided unless the grantor has unusual longevity expectations and the asset has a known long-tail.

Annuity profile

Treas. Reg. § 25.2702-3 permits the annuity to step up by as much as 120% per year. This is meaningful in a low-rate environment: backloading the annuity means the trust retains the asset (and its growth potential) for longer before paying out, which compounds the appreciation that exits the estate.

A typical structuring choice for a 2-year GRAT funded with $10M: annuity year 1 = $4.5M, annuity year 2 = $5.4M (a 20% step-up). Under most current §7520 rates that approaches a zero-out.

Asset selection

GRATs work best with assets that have asymmetric upside — concentrated positions, pre-event securities, single-stock biotech holdings, founder shares. The strategy is to GRAT volatile assets and hope for outperformance; if the asset underperforms, the GRAT simply returns it.

Conversely: do not GRAT a balanced portfolio. The expected return of a 60/40 portfolio over a 2-year period is barely above the §7520 rate; the GRAT will return all the assets and you'll have generated lawyer fees and accounting work for nothing.

Mortality risk and rolling GRATs

The single material risk of a GRAT is that the grantor dies during the term. If that happens, the entire trust corpus is included in the grantor's estate under IRC § 2036 — completely undoing the planning.

Mitigation is structural:

• Use short terms. Two-year GRATs have a meaningfully lower mortality probability than 10-year GRATs for any given grantor.
• Use a rolling series. A two-year GRAT that succeeds can be immediately re-GRAT'd with the returned annuity payments. If the grantor dies during any one GRAT, only that GRAT's assets are included — prior successful GRATs are already out of the estate.
• Layer life insurance. A term policy on the grantor for the GRAT term, owned by an Irrevocable Life Insurance Trust (ILIT), offsets the estate tax exposure if mortality hits.

GRAT vs SLAT vs IDGT decision matrix

Quick rules of thumb when the client's situation could support more than one:

• Use a GRAT when the asset is volatile, the §7520 rate is low, and the client has solid mortality expectancy. No exemption needs to be used; the gift is zero.
• Use a SLAT when the client wants to use their exemption proactively (e.g., sunset hedge), retain practical access through their spouse, and the asset profile is more stable. See our SLAT drafting post.
• Use an IDGT (sale to grantor trust) when the asset is illiquid and a discount applies (closely-held LLC, real estate), and the client wants a true freeze-and-sell rather than an annuity structure.

A worked example

Client funds a 2-year GRAT in June 2026 with $10M of pre-IPO equity. §7520 rate that month: assume 4.6%. Annuity backloaded with 20% step-up: year 1 = $4.66M, year 2 = $5.59M. The IRS's present-value math:

• PV of $4.66M one year out @ 4.6% = ~$4.46M
• PV of $5.59M two years out @ 4.6% = ~$5.11M
• Total PV of retained annuity ≈ $9.57M
• Gift = $10M − $9.57M = ~$430K, or with rate adjustments, often near zero

If the equity grows 30% per year (plausible for late-stage pre-IPO), the corpus at year 2 is roughly $16.9M. Annuity payouts return $10.25M to the grantor; the remainder — about $6.65M — passes to the beneficiaries gift-tax-free.

If the equity grows 0%, the trust simply returns the original $10M in installments. No harm, no foul.

The bottom line

Two-year rolling GRATs funded with concentrated appreciating assets remain the workhorse strategy whenever the §7520 rate is below the asset's expected return. The Walton structure makes the zero-out gift possible; rolling mitigates mortality; backloaded annuities maximize the appreciation that exits the estate.

Drafting the GRAT is the easy part. Picking the right asset, sizing the annuity for the current rate, and committing to the rolling discipline is where attorney judgment lives.

Related Field Notes

• Drafting a Georgia SLAT in 2026: Reciprocal Trust Traps — for the cases where the client wants spousal access rather than an annuity stream.
• Funding the Georgia Revocable Living Trust — the foundation that sits underneath every estate plan.

This piece is general legal commentary, not legal advice. GRAT economics depend on the §7520 rate as of the month of funding; check the IRS's current monthly publication before quoting numbers to a client.