Skip to content
← Back to Blog

Gift, GRAT, or Hold for Step-Up on $6M of Stock With Mortgage Rates Above 7%? The 7.7% Break-Even (September 2026)

Here's a setup I see a lot. You're single with a $22 million estate, and $6 million of it is one stock you bought for $1.5 million. The 2026 federal exemption is $15 million per person, so you're $7 million over the line. That's about $2.8 million of federal estate tax at 40% if you died today.

So do you hold the stock for the step-up in basis, gift it now, or put it in a GRAT?

This week's headlines make the answer harder, not easier. Here's how I'd work through it.

What the September 30 numbers change (and what they don't)

Three sources frame the moment:

  • BLS, "Major Economic Indicators Latest Numbers": CPI +0.4% in August 2026, unemployment 4.1%, payrolls +162,000 (preliminary), average hourly earnings +$0.10 (preliminary).
  • NerdWallet, "Mortgage Rates Today, Wednesday, September 30: Steadily Above 7%": rates are in a holding pattern and inflation is still running hot.
  • Mr. Money Mustache, "Will the AI Bubble Destroy our Retirement?": the market keeps surprising us, in both directions.
Data pointReadingWhat it touches in a transfer plan
CPI (BLS)+0.4% in Aug 2026Annualizes to about 4.9% if repeated for 12 months (1.004¹², my arithmetic, not a forecast). Feeds exemption indexing and rate expectations.
Unemployment and payrolls (BLS)4.1%; +162,000 (p)Not a crisis signal, so no obvious case for rates dropping soon.
Mortgage rates (NerdWallet)Steadily above 7%Same rate pressure that lifts Treasury yields, and with them the IRS 7520 rate.
Stock market (Mr. Money Mustache)Surprises both waysValuation risk at the moment you transfer.

One technical point. The 7520 rate is 120% of the mid-term federal rate, so it tracks Treasury yields, not mortgage rates directly. They tend to move together, though. A higher 7520 rate raises the hurdle a GRAT has to clear. The Voritanel post on what a 0.2-point swing costs a $10 million GRAT vs. IDGT covers that mechanism in depth.

The IRS publishes the actual rate monthly. I'm using 5.0% as an assumed hurdle below, with 4.8% and 5.2% as bookends.

The example: $6 million, three market paths, ten years

Assumptions (this is an example, not a forecast):

  • Basis $1.5 million, so a $4.5 million embedded gain. Selling outright today would trigger $1,071,000 of capital gains tax at 23.8% (20% plus the 3.8% NIIT).
  • The rest of your estate keeps you above the exemption, so every extra dollar is taxed at 40%.
  • Federal only, and your remaining exemption covers any gift.
  • Death (or the decision point) at year 10.

Three paths for the stock:

  • Bull: 12% a year, reaching $18.64M.
  • Base: 7% a year, reaching $11.80M.
  • AI pop: down 30% in year one, then 5% a year, reaching $6.52M.

How each option works:

  • Hold: the estate pays 40% of the full value, but basis resets to market and heirs owe no capital gains.
  • Gift now: the $6M counts in your tax base at its gift-date value (a $2.4M cost), and growth escapes the estate. But heirs take your $1.5M basis.

Ten-year federal tax bill on the position

PathValue in year 10Hold for step-upGift, heirs sellGift, heirs never sell
Bull (12%)$18.64M$7.45M$6.48M$2.40M
Base (7%)$11.80M$4.72M$4.85M$2.40M
AI pop, then 5%$6.52M$2.61M$3.59M$2.40M

If your heirs sell, gifting wins only in the bull case, by about $976,000. It loses by $131,000 in the base case and $988,000 after the pop. That's because you gave up a step-up that would have erased $4.5 million or more of gain.

The last column changes the picture. If heirs hold the stock for life, they get their own step-up, and gifting wins every path. The math is simple, but it rests on a guess about what your kids will do. It also puts the growth into their estates, which is where generation-skipping planning comes in (more below).

How long until a gift pays off?

Gifting beats holding (heirs sell) when:

0.4 × (V − $6M) − 0.238 × (V − $1.5M) > 0, which simplifies to 0.162 × V > $2.043M, or V > $12.61M.

That's 2.1x your starting value. The annual growth needed depends on your horizon:

HorizonRequired annual growth
5 years16.0%
10 years7.7%
20 years3.8%

This explains the base-case loss. A 7% return over ten years falls just short of the 7.7% needed. A shorter horizon makes the gift much harder to justify, and a longer one makes it much easier.

This is the kind of analysis Voritanel runs for you, so you don't have to rebuild the spreadsheet every time your basis, horizon, or growth assumption changes.

For a related comparison at a different position size, see GRAT vs. gifting vs. holding for step-up on $5M of stock.

The GRAT: the option that shrugs at the AI-bubble question

A two-year zeroed-out GRAT on $6M at a 5.0% hurdle pays you back about $3.23M a year. That's $6M divided by (1/1.05 + 1/1.05²), or $6M ÷ 1.8594. Whatever the stock earns above 5% passes to heirs gift-tax free. If the stock falls, the annuity payments return it to you, and you've lost only legal fees and some time.

Remainder to heirs after a two-year GRAT on $6M

Growth7520 at 4.8%7520 at 5.0%7520 at 5.2%
12% a year$705K$686K$666K
7% a year$209K$190K$171K
Down 30%, then +5%$0$0$0

Moving the 7520 rate by 0.4 points costs about $39K of remainder. Moving the market from 12% to 7% costs about $496K, roughly 13 times as much. On a two-year term, the market path matters far more than next month's rate print. Longer terms and bigger positions make the rate more important.

GRATs are normally rolled, so I ran five back-to-back two-year GRATs, with each annuity return re-funded into the next one:

  • At 12%: about $7.1M of the eventual $18.6M moves out of the estate, saving roughly $2.8M at 40%.
  • At 7%: about $1.5M of $11.8M moves, saving roughly $617K.
  • After the pop: roughly $0.

That rough model ignores fees and basis, and it assumes you survive each term. If you die during a GRAT term, the assets are pulled back into your estate. Remainders also carry your low basis. For the wider comparison, see GRAT vs. IDGT vs. direct gift on a $10M asset at a 5% hurdle.

Four things that can flip the answer

1. Whether your heirs will actually sell. This single assumption swings the gift result by more than any market path in the first table. Ask before you assume.

2. Inflation and exemption indexing. Each 1% upward adjustment adds $150,000 to a single exemption, about $60,000 of tax shielded at 40%. A hot CPI print helps here. But the indexing formula is lagged and specific, so don't treat one month's 0.4% as a forecast.

3. Gifting at the top. Suppose you gift $6M and the stock sits at $4.2M (down 30%) and never recovers. You've spent $6M of exemption on $4.2M of stock. The extra $1.8M of exemption is worth up to $720,000 at 40%. If the AI-bubble worry is real for your holdings, that's the case a GRAT is built for. Voritanel's 5-question checklist for a $20M estate if the AI bubble pops walks through it.

4. Everything outside the federal line. State estate taxes can apply even when federal tax is $0, as in the $9M estate that still owed $947,500 to its state. If you're married, portability and two exemptions may change whether you're over the line at all. And if gifts go to children outright, the growth lands in their estates later unless you allocate GST exemption to a trust.

A charitable option. If you have real charitable intent, a charitable remainder trust can sell the stock without the $1,071,000 immediate capital gains bill. Voritanel's comparison of a GRAT vs. a CRT on $4M of concentrated stock shows the trade-off.

If the real pressure is a child facing a 7%+ mortgage, the same basis trade-off shows up at small scale. See gifting a $100,000 down payment with mortgage rates above 7%.

Which option leans where

If this describes you…It leans towardMain cost to weigh
Heirs will keep the stock for life, estate well above $15M, long horizonGift (into a trust with GST allocation)Exemption spent at today's valuation
Growth likely below 7.7% a year, or you may need the stock, or horizon is shortHold for step-upGrowth stays in the taxable estate
You want upside without spending exemption and worry about valuationsRolling GRATsTerm mortality risk, fees, carryover basis on remainders
Charitable intent plus a big embedded gainCRTPart of the value goes to charity

These are leanings, not recommendations. Each one flips if a key input changes.

Your numbers will differ

Everything above is a worked example. A $1.5M basis, a 10-year horizon, and a 5.0% hurdle are my assumptions, not yours. Change any of these and the break-even moves:

  • your basis
  • your marital status
  • your state
  • your health
  • how long you expect to hold
  • what your heirs would do with the stock

For instance, a $3M basis instead of $1.5M shrinks the step-up you'd give up, which lowers the gift's break-even growth rate. The estate planning calculations are simple enough to follow. The hard part is that there are a dozen inputs, and they interact.

When I ran my own numbers, the thing that surprised me most was how much the answer hinged on one or two inputs I'd been treating as afterthoughts. That's why I'd rather see the calculation than trust a rule of thumb.

If you'd like to run this on your own position, you can model it at Voritanel. Enter your basis, horizon, growth assumptions, and the current 7520 rate, and compare hold, gift, and GRAT side by side. There's no pressure to pick anything. The point is to see your own numbers before mortgage rates, inflation, or the market make the decision for you.

Sources

Ready to optimize your estate plan?

Optimize Your Estate Plan Free