Skip to content
STI
← Back to Blog

Why a $9 Million Estate With $0 Federal Tax Due Can Still Owe $947,500 in State Estate Tax

The Question Nobody Runs the Numbers On

NerdWallet ran a piece recently on whether it's worth switching banks just to chase a sign-up bonus — usually a few hundred dollars, for maybe an hour of paperwork and a ding to your account-opening history ("Should I Switch to a New Bank Just to Earn a Bonus?"). People agonize over that math. Spreadsheets get built. Reddit threads get started.

Meanwhile, a lot of those same people are sitting on estates worth $9 million, $12 million, $20 million — and have never run the equivalent five-minute calculation that could be worth six figures, not three.

Here's the gap: the federal estate tax exemption in 2026 is $15,000,000 per person. If your estate is $9 million, your executor can look at the federal return and see a big fat zero. Most people stop there. That's the mistake. The state estate tax bill on that same $9 million doesn't care what the federal exemption says — and depending on your zip code, it can range from nothing to over $900,000.

The $9 Million Estate That Owes Nothing — On Paper

Let's build the example. A single person, no spouse, dies in 2026 with a $9,000,000 estate: $5.5 million in diversified brokerage and retirement accounts, and $3.5 million concentrated in one AI-infrastructure stock that's up roughly 400% since 2023 (original cost basis around $700,000).

Federal math: $9,000,000 estate minus the $15,000,000 exemption = $0 taxable. Federal estate tax owed: $0.

That's the number most families see and stop reading. But the federal exemption only answers the federal question. It says nothing about the other tax return your executor has to file — the state one, if you live in one of the roughly dozen states (plus D.C.) that still impose their own estate tax, usually with an exemption far below $15 million.

Where the Real Bill Comes From: State Estate Tax

Take the exact same $9 million estate and change only the state of residence:

StateExemption (approx., 2026)Top marginal rateEstimated tax on $9M estate
Oregon$1,000,00016%≈ $947,500
New York (cliff state)~$7.16M, but cliff at 105% of exemptionup to 16%≈ $1,015,000
Massachusetts$2,000,000up to 16%≈ $828,000
Florida / TexasNo state estate tax0%$0

(These are illustrative estimates built from published state exemption thresholds and bracket structures — not a filed return. Every state's formula has quirks, and New York's "cliff" rule means crossing the threshold by even 5% can pull the entire estate into taxable territory rather than just the excess.)

Notice what's happening: the federal government says $0. The state government, in three of these four scenarios, says somewhere between $828,000 and just over $1,000,000 — on the identical estate. That's not a rounding error. That's the hidden cost that "check the federal exemption and move on" completely misses. We wrote about this exact trap for a $7M Massachusetts estate that owed about $588,000 despite the federal exemption — the mechanics are the same, just scaled up here.

This is the kind of analysis Voritanel runs for you — plugging in your actual state, your actual asset mix, and your actual exemption use — so you don't have to build this bracket table yourself.

The AI Stock Complicates Everything

Mr. Money Mustache's recent post ("Will the AI Bubble Destroy Our Retirement?") asks the question a lot of people are quietly asking themselves right now: what happens if the market's "super-duper-crazy" run in AI-adjacent names reverses? It's not a hypothetical in this example — $3.5 million of the $9 million estate is sitting in exactly that kind of concentrated position.

That concentration cuts both ways for estate planning:

  • If the position keeps compounding — say 20% a year for five years — $3.5 million becomes roughly $8.7 million. Suddenly the total estate isn't $9 million anymore; it's pushing toward $14 million, closer to that $15 million federal exemption, and the state tax bill (which has no cushion comparable to the federal exemption in low-threshold states) grows right along with it.
  • If the position drops 40% in a bubble-pop scenario, the concentrated stock falls to roughly $2.1 million, the total estate drops to about $7.6 million, and the Oregon-style state tax bill falls by roughly $300,000 in this example — a real number, moving the other direction.

Either way, the estate isn't static, and neither is the tax bill. That's the part flat, one-time estimates always miss: they assume today's numbers hold. Real portfolios don't sit still, especially concentrated ones.

What This Week's Jobs and Inflation Data Have to Do With It

The BLS's latest economic indicators (Consumer Price Index +0.4% in August 2026, unemployment holding at 4.1%, payrolls up 162,000, average hourly earnings up another $0.10) point to an economy that's still running hot enough that the Fed has room to keep rates elevated rather than cut. That matters here for a reason that has nothing to do with mortgages: the IRS 7520 rate — the "hurdle rate" that governs GRATs, CRTs, and other split-interest trusts — moves with the broader rate environment.

A higher 7520 rate makes a GRAT (Grantor Retained Annuity Trust) less efficient, because more of the trust's growth has to clear the hurdle before anything passes to heirs tax-free. An IDGT (Intentionally Defective Grantor Trust) sale, by contrast, is largely indifferent to the 7520 rate — it uses a different interest rate benchmark and structure entirely. We've walked through this exact GRAT-vs-IDGT sensitivity in detail elsewhere, including how a jump toward a 5% hurdle rate can cost a GRAT hundreds of thousands more than an IDGT on a comparable estate, and what a rising-rate environment does to the GRAT vs. IDGT break-even math.

For the concentrated AI stock in this example, that rate environment is the deciding factor in whether a GRAT is even the right tool, or whether an IDGT sale makes more sense given where the 7520 rate is sitting this month.

Do Nothing vs. Gift Now: Running Both Paths

Back to the core decision on the $3.5 million AI stock position. There are really two paths:

Path 1 — Hold until death, take the step-up in basis. The stock stays in the estate. At death, its cost basis resets to fair market value (whatever it's worth on the date of death), erasing the embedded capital gain entirely. Heirs could sell the next day and owe no capital gains tax. The cost: the full $3.5 million (or whatever it's grown to) stays in the taxable estate, exposed to whatever state estate tax applies.

Path 2 — Gift the stock now (using lifetime gift tax exemption, since it's well above the $19,000 annual exclusion). The stock leaves the taxable estate immediately, removing it — and all its future growth — from the state estate tax calculation entirely. The cost: the $700,000 original basis carries over to the recipient. If they eventually sell at $3.5 million, they owe capital gains tax on the $2.8 million gain — roughly 23.8% federal (20% long-term rate plus the 3.8% net investment income tax) plus state capital gains tax, which in a state like Oregon (taxed as ordinary income, up to 9.9%) pushes the combined rate close to 34%. On a $2.8 million gain, that's approximately $943,600 in capital gains tax whenever the recipient sells.

Line those two numbers up: holding for step-up avoids roughly $947,500 in Oregon state estate tax exposure on the appreciated piece, but the capital gains bill if gifted now runs almost identically — about $943,600. In this specific example, it's close to a wash on the built-in gain alone. The tiebreaker becomes future growth: every dollar the stock appreciates after a gift is made escapes the donor's estate completely, gift tax and state estate tax both. Every dollar it appreciates while still owned by the donor stays exposed. If you believe the AI trade still has room to run, gifting sooner locks in today's value and lets the growth happen outside the taxable estate. If you think a correction is coming, there's less urgency — and holding for the step-up avoids irreversibly giving up the basis reset on a position that might be worth less next year anyway.

This is exactly the kind of scenario where the "right" answer flips based on your growth assumption, your state, and your actual cost basis. You can model this for your specific situation at Voritanel rather than guessing which side of the wash the math falls on.

So What Should You Actually Do?

Three questions determine which path makes sense, and none of them have a universal answer:

  1. What state do you live in — and will you still live there at death? The difference between Florida ($0) and New York (potentially over $1 million) on an identical $9 million estate is the single largest variable in this entire analysis, bigger than any trust structure.
  2. How much of your estate is concentrated, low-basis stock? The lower the basis relative to current value, the more the step-up matters, and the more a bubble-pop scenario changes the calculus in your favor if you wait.
  3. Where is the IRS 7520 rate right now, and where is it headed? With inflation still running warm and unemployment steady at 4.1%, the rate environment favors different tools than it did a year ago — see the decision framework for GRAT vs. IDGT vs. portability for how to weigh that against your specific term length and growth assumptions.

None of this is a reason to panic, and none of it points to one obviously correct move — the Oregon-versus-Florida gap alone should make clear that "estate planning" isn't one-size-fits-all advice, it's a function of your specific state, basis, and risk tolerance. But it is a reason to actually run the numbers instead of stopping at the federal exemption and assuming you're covered. The bank-bonus math gets more attention than this does in a lot of households, and the estate math is worth a hundred times more.

Run your own numbers — your state, your basis, your portfolio mix, and today's actual 7520 rate — at Voritanel.

Sources

Ready to optimize your estate plan?

Optimize Your Estate Plan Free