An intentionally defective grantor trust does the opposite of what its name suggests.

The "defect" is intentional. The structure is rigorous. And for the right asset, the result is millions of dollars of estate-tax savings — appreciation that grows outside your taxable estate while you, the grantor, continue to pay the trust's income tax (which itself becomes a tax-free gift to the beneficiaries).

The catch: an IDGT is a precise tool with a specific job. Founders frequently arrive at IDGT planning conversations expecting it to do something it isn't built for — most often, QSBS exclusion stacking. This guide explains what an IDGT actually is, how the §675(4)(C) substitution power makes it work, the two ways founders fund one, and the single most important warning in the IDGT toolkit: an IDGT is not a QSBS stacking vehicle.

By the end, you'll know whether an IDGT belongs in your pre-exit planning — and what to use instead when it doesn't.

What is an IDGT?

An IDGT is an irrevocable trust that the IRS treats as a grantor trust for income tax purposes but as a completed gift, outside the estate, for transfer tax purposes. The "defect" is the deliberate income-tax-side mismatch. That mismatch is the entire reason the structure exists.

Here's the mechanic in one paragraph. IRC §§671–679 (the grantor trust rules) and IRC §§2036–2038 (the estate inclusion rules) use overlapping but distinct lists of "retained powers." Some powers trigger income-tax inclusion under §671–679 without triggering estate inclusion under §2036–2038. A trust agreement that includes such a power gets the best of both regimes: the grantor pays the trust's income tax (income-tax inclusion), and the assets stay outside the estate (no estate-tax inclusion).

The grantor's annual income-tax payment on trust earnings is settled doctrine: it's not a §2511 gift to the trust beneficiaries. The grantor depletes their estate by paying tax on income they don't actually receive — a quiet but compounding estate reduction.

The §675(4)(C) "swap power"

The most common way to make a trust intentionally defective is to include the power for the grantor to substitute trust property for property of equivalent value, exercised in a non-fiduciary capacity. That single sentence in the trust agreement does three things:

  • Triggers grantor-trust status under IRC §675(4)(C). The grantor is treated as the income-tax owner of all trust property.
  • Does not cause estate inclusion under §§2036–2038. The substitution power isn't on those sections' retained-powers list.
  • Gives the grantor flexibility to swap appreciated assets out of the trust for high-basis assets — a "basis-burn" move that's especially useful late in life when the grantor wants to step up basis at death.
"The power of substitution under IRC Section 675, by its mere inclusion in a trust, extends grantor trust status to the entire trust without any further action by the grantor."
— Practitioner monograph on §675(4)(C) swap powers

Two drafting requirements matter: the power must be exercisable in a non-fiduciary capacity (a trustee acting in fiduciary capacity won't trigger grantor-trust status), and the assets exchanged must be of equivalent value at the time of substitution (independent appraisal is the practitioner default).

The economic effect

Three things happen once an IDGT is funded:

  • The grantor pays income tax on the trust's earnings every year. The trust accumulates wealth tax-free relative to the alternative — a non-grantor trust paying its own tax at compressed trust rates, or the grantor receiving the income directly and paying tax on it.
  • The annual tax payment is itself a tax-free gift to the trust. No §2511 gift-tax consequences. No use of the lifetime gift exemption.
  • All future appreciation accumulates outside the grantor's taxable estate. At the grantor's death, the trust's assets pass to beneficiaries without estate tax.

That's the whole machine.

How an IDGT actually works (the two funding paths)

There are two ways to put assets into an IDGT.

Path 1: Completed gift

The grantor gifts assets directly to the trust. The gift uses lifetime gift-tax exemption — under OBBBA, that exemption is permanently set at $15M per individual ($30M for married couples) starting in 2026, indexed for inflation thereafter.

Upside: simple. No installment-note accounting. No interest payments back and forth. The grantor's estate decreases by the gift amount immediately.

Downside: uses gift exemption. No continued cash flow back to the grantor.

Path 2: Sale to IDGT (installment note)

The more sophisticated funding path:

  1. The grantor seeds the trust with a completed gift — typically about 10% of the asset value.
  2. The grantor then sells the remaining ~90% of the asset to the trust in exchange for an interest-bearing promissory note at the applicable federal rate (AFR).
  3. The trust pays the grantor interest annually (and principal per the note's amortization schedule).

Why this works: because the trust is a grantor trust for income tax purposes, the IRS treats the grantor and the trust as the same taxpayer for income-tax purposes. The sale is a non-event under Rev. Rul. 85-13. There's no §1001 gain recognition. For QSBS, there's no holding-period restart.

The economic effect is an estate freeze. The asset's value is locked at the sale-date number plus the AFR yield. All upside above the AFR accumulates inside the trust, gift-tax-free. The grantor's estate captures only the residual note value plus any unspent interest.

Worked example

A founder owns $20M of pre-IPO equity expected to triple to $60M over three years. The founder seeds an IDGT with $2M (a 10% completed gift, using $2M of lifetime gift exemption) and sells the remaining $18M to the trust in exchange for a 9-year AFR-rate installment note.

Three years later, the equity has tripled. The trust holds equity worth ~$60M ($20M × 3). The trust still owes the grantor approximately $18M on the note (ignoring AFR amortization for simplicity). Net trust value: ~$42M.

Versus holding the asset directly in the grantor's estate: about $40M of appreciation has moved outside the estate, at the cost of using only $2M of lifetime gift exemption. At a 40% top federal estate-tax rate, that's roughly $16M of estate tax avoided.

The QSBS question — and the most important warning in this guide

Here is the single most important sentence in this article: an IDGT is not a QSBS stacking vehicle.

Founders frequently arrive at IDGT conversations expecting one. They aren't.

The mechanic, in plain English: §1202's exclusion is per-taxpayer. An IDGT is, by definition, a grantor trust for income tax purposes — the IRS treats the grantor as owning the trust's assets for income-tax purposes, including any QSBS the trust holds. So the grantor and the IDGT share one §1202 exclusion, not two. There is no separate-taxpayer treatment to multiply.

"Certain assets, like QSBS stock under IRC Section 1202, may not be good candidates [for an IDGT] because grantor trust status disregards the additional gain exclusion that might otherwise be available to the trust."
— Practitioner literature, summarized in Heckerling QSBS materials

If your goal is to multiply the §1202 exclusion across taxpayers — to claim more than one $15M cap on a single exit — the right vehicle is a non-grantor trust, not an IDGT.

When IDGTs and QSBS do mix

Two scenarios are the exception, and they're both about estate-tax planning, not income-tax stacking:

  • Estate freeze on QSBS appreciation above the §1202 cap. A founder with $80M of expected QSBS gain and a $15M exclusion still has $65M of post-cap gain that's exposed to estate tax if the founder dies before liquidating. An IDGT can carry the post-cap appreciation outside the estate. The QSBS exclusion mechanics don't change; the estate impact does.
  • Sale-to-IDGT pre-cap. Selling QSBS to an IDGT in exchange for an installment note is an "ignored transfer" between grantor and grantor trust under Rev. Rul. 85-13. QSBS character and holding period are preserved. The structure freezes value for estate-tax purposes — but it does not add a new §1202 exclusion.

What to use instead for QSBS stacking

If stacking is the goal, three vehicles fit:

  • A non-grantor trust for separate-taxpayer status — see our non-grantor trust guide
  • A SLANT if the founder's spouse should retain access — see our SLANT guide
  • The IDGT does not belong in this category

For the practical mechanics of stacking, see our QSBS trust stacking guide.

When founders should use an IDGT

An IDGT fits three founder scenarios well:

  • Estate freeze on a high-growth, non-QSBS asset. Investment portfolios, real estate, secondary-market equity, family business interests with no §1202 status. Lock in current value, push appreciation outside the estate.
  • Pre-IPO founder with QSBS gain expected to exceed the cap by a large margin. The first $15M is excluded under §1202 regardless. Above that, the appreciation needs an estate plan. An IDGT can carry that post-cap layer for transfer-tax separation. Whether IDGT or NGT is the right choice for the post-cap layer depends on liquidity goals, beneficiary structure, and stack-vs-freeze priority.
  • Multi-generational wealth transfer where the grantor still wants income. The sale-to-IDGT structure produces a fixed-yield instrument (the note) back to the grantor while the underlying asset grows outside the estate.

When an IDGT is the wrong tool

  • The goal is QSBS stacking. Use a non-grantor trust.
  • The grantor doesn't want the cash-flow hit of paying the trust's income tax. A non-grantor trust may be the right structure.
  • The grantor wants the trust to make distributions back to themselves. The IDGT's grantor-trust treatment doesn't help here, and the retained beneficial interest creates estate-inclusion risk.

IDGT vs. non-grantor trust vs. GRAT

Three estate-planning trusts show up in almost every founder pre-exit conversation. They sound similar. They are not interchangeable.

  • Income tax. IDGT: grantor pays (intentional defect). NGT: trust pays (Form 1041). GRAT: grantor pays.
  • Separate QSBS taxpayer. IDGT: no. NGT: yes. GRAT: no.
  • Outside the grantor's estate. IDGT: yes (if grantor outlives the terms). NGT: yes (if structured properly). GRAT: only if the grantor survives the GRAT term.
  • Uses gift exemption. IDGT: yes (initial seed gift) or no (sale path). NGT: yes (gift). GRAT: minimal (zeroed-out GRATs).
  • Best for. IDGT: estate freeze + basis-burn growth assets. NGT: QSBS stacking + asset protection. GRAT: short-term estate freeze with low gift cost.
  • Typical funding. IDGT: gift or installment sale. NGT: gift. GRAT: gift with annuity payback.

When an IDGT beats a GRAT

A GRAT is a 2- to 10-year mortality bet. If the grantor dies during the term, the entire trust value is included in the estate — a complete failure of the structure. An IDGT has no comparable term-limited mortality risk.

IDGTs also accommodate longer-horizon assets — illiquid pre-IPO equity, long-hold real estate — better than a GRAT's annuity-payback structure.

When a GRAT beats an IDGT

GRATs use almost no gift exemption when properly zeroed-out. For founders who've already used their $15M lifetime exemption, a GRAT can move appreciation without consuming additional exemption.

When an NGT beats an IDGT

For QSBS stacking, full stop. Asset protection is also stronger in a non-grantor structure (the grantor isn't deemed to retain economic ownership for any purpose).

How to set up an IDGT

An IDGT is not a self-serve product. The components:

  • Trust agreement drafted by a tax attorney with the §675(4)(C) swap power explicitly included, exercisable in a non-fiduciary capacity. Alternative grantor-trust triggers — §674 (power to control beneficial enjoyment), §677 (income for grantor's spouse) — are available but less common.
  • Independent trustee — required so that the swap power isn't held by a fiduciary (which would defeat the §675(4)(C) trigger).
  • Jurisdiction — Nevada, South Dakota, and Delaware are common choices for the same reasons that drive non-grantor trust situs: directed-trustee statutes, no state income tax on trusts, robust asset protection.
  • Funding decision — gift versus sale-to-IDGT, with installment note at the current AFR if sale.
  • Annual mechanics — the grantor reports trust income on personal Form 1040; the trust files an informational return; basis is tracked separately for swap-power purposes.

Promissory's core focus is QSBS-focused non-grantor trust planning. For IDGT and broader estate work, we coordinate with QSBS-aware estate attorneys. Schedule a free consultation to scope which structure — or combination — is right for your situation.

The three mistakes that kill IDGT plans

The most expensive failures we see come from three patterns:

  • Funding an IDGT with QSBS expecting stacking. This is the single most expensive mistake — covered in detail above. Use a non-grantor trust for stacking; reserve the IDGT for non-QSBS assets or post-cap QSBS appreciation. A founder who funds a $20M slug of QSBS into an IDGT expecting an additional $15M exclusion ends up with one exclusion shared between grantor and trust — a ~$3.57M federal-tax surprise.
  • AFR mispricing on an installment sale. If the note rate is below the applicable federal rate at closing, the IRS recharacterizes part of the sale as a gift. The fix is straightforward: lock in the right AFR, document the date, and use an independent valuation for the underlying asset. Sloppy AFR documentation is a frequent audit trigger.
  • Reciprocal trust collapse on spousal IDGTs. Spouses each setting up an IDGT naming the other as beneficiary, with substantially identical terms and same-day funding, triggers the reciprocal trust doctrine. Courts have unwound these structures repeatedly. Differentiate beneficiaries, terms, funding dates, and amounts — or skip the symmetry and use a different structure for the second spouse.

The bottom line

Three takeaways:

  1. An IDGT is a grantor trust for income tax purposes and a completed gift for transfer tax purposes. That mismatch — engineered through the §675(4)(C) swap power — is the value proposition.
  2. The grantor pays the trust's income tax, and that payment is a tax-free gift to beneficiaries. Combined with the estate-freeze effect of the structure, an IDGT can move tens of millions of dollars of appreciation outside the taxable estate over time.
  3. An IDGT is not a QSBS stacking tool. If your goal is multiplying the §1202 exclusion, use a non-grantor trust. If your goal is freezing post-cap QSBS appreciation or non-QSBS growth assets out of your estate, the IDGT is the right structure.

Promissory builds QSBS-focused non-grantor trusts on a fixed-fee model and partners with QSBS-aware estate attorneys for IDGT and broader estate work. If you have founder stock and a horizon to exit, schedule a free consultation to scope the right structure for your situation. See our pricing for transparent fixed fees.

‍