Why selling stock to buy a house costs more than you think

TL;DR: Selling appreciated stock to fund a large purchase triggers capital gains tax. That part most people know. What fewer people realize is that depending on your income and where you live, the total tax bill can run far higher than the headline rate. In California, a high earner selling a sizable position can see combined federal and state taxes take 40% or more of their gains. Borrowing against the portfolio instead can defer all of that.

---

The 2-minute version

Consider Tony. He and his partner bring in $300,000 a year between their salaries. They want to buy a $2 million house in the Bay Area. They have been saving for years: a mix of vested RSUs and S&P 500 index funds that has grown to about $1.1 million. The cost basis on that position is roughly $300,000, meaning they have $800,000 in long-term gains sitting in the account.

The plan: sell enough of the portfolio to cover the 25% down payment of $500,000.

Here is what actually happens.

To net $500,000 after taxes, they need to sell roughly $700,000 of the portfolio. That sale generates about $509,000 in long-term capital gains (the proportional cost basis on the sold shares is around $191,000). Combined with their $300,000 salary, total income for the year hits $809,000. California taxes long-term gains at the same rate as ordinary income: 13.3% at the top. The federal rate on long-term gains at this income level is 20%, plus the 3.8% Net Investment Income Tax. And then there is a hidden wrinkle in the current tax code that hits households in exactly this income range.

The total tax bill on that sale comes to roughly $199,000. They net about $500,000 from the $700,000 sale — but they have now liquidated nearly two-thirds of their portfolio. They had $1.1 million. After the sale, $400,000 is left.

---

The nerdy version

The SALT torpedo

Congress passed a tax law in 2025 called the One Big Beautiful Budget Act, which set the cap on state and local tax (SALT) deductions at $40,000. For most taxpayers this was an improvement over the prior $10,000 cap. But it included a phase-out: households with income between $500,000 and $600,000 lose the SALT deduction at the rate of 30 cents per dollar of income over $500,000.

Here is how the trap works. A taxpayer at $499,000 of income gets the full $40,000 SALT deduction. A taxpayer at $600,000 gets $10,000. Each dollar of income between $500,000 and $600,000 costs 30 cents of SALT deduction. Those 30 cents of lost deduction, taxed at the 37% federal bracket, translate to an additional 11.1 cents of tax per dollar of income in that band.

For a California resident in the long-term capital gains zone, the base rate on those gains is already around 37% (20% federal, 3.8% NIIT, 13.3% California). The SALT torpedo adds another 11.1% in that band, pushing the effective marginal rate to roughly 48%.

Tony enters the year at $300,000 of W-2 income. When he sells, his gains push him directly through the $500,000 to $600,000 band. That $100,000 of gains in the torpedo zone costs him about 11 cents more per dollar than he expected.

Running the actual numbers

Simplified illustration for Tony: roughly $509,000 in long-term capital gains on top of $300,000 in W-2 income.

Layer	Amount	Rate	Tax
First $200K of gains (income $300K to $500K)	$200,000	~37% (20% fed + 3.8% NIIT + 13.3% CA)	~$74,000
SALT torpedo zone ($500K to $600K)	$100,000	~48%	~$48,000
Remaining gains (above $600K)	$209,000	~37%	~$77,000
Total estimated tax			~$199,000

This is a simplified illustration, not tax advice. Actual liability depends on filing status, deductions, and other factors. Consult a CPA.

Tony sells $700,000, nets roughly $500,000, and the portfolio drops from $1.1 million to $400,000.

What compounding costs on top of taxes

That is just the immediate tax cost. There is also the opportunity cost of the assets that are no longer compounding.

$1.1 million growing at 7% annually becomes roughly $2.16 million in ten years. If Tony borrowed $500,000 against the portfolio instead of selling, paying 4% in annual interest, the portfolio grows to $2.16 million while the outstanding loan grows to approximately $740,000. Net equity in the portfolio after ten years: roughly $1.42 million.

Compare that to the selling path: $400,000 left in the portfolio after the sale grows to roughly $787,000 over the same period. The difference in portfolio value alone is more than $600,000 — before accounting for the $199,000 in taxes avoided upfront.

One piece we are setting aside for now: mortgage interest is generally deductible up to certain limits, and box-spread interest has its own deductibility treatment that changes the effective cost. That comparison is worth its own article, and we will come back to it.

That is not a free lunch. A $500,000 loan outstanding has to be managed. But the math illustrates why people who have done this before treat “never sell appreciated assets” as a standing rule.

The Buy, Borrow, Die framework

Finance writers have described this strategy as “Buy, Borrow, Die.” Buy appreciating assets. Borrow against them instead of selling. At death, heirs inherit at a stepped-up cost basis. The embedded capital gains are permanently erased, and the loan gets repaid from the estate.

The reason the strategy is not universal is friction: most people cannot borrow at 4% against their portfolio. Banks charge 8-13% on margin loans. That spread is too wide to make the math work comfortably. The box-spread approach closes most of that gap.

It also does not have to be all-or-nothing. A household that needs to cover a gap rather than the entire down payment might borrow $100,000 or $200,000 while funding the rest from savings or a conventional mortgage. The rate advantage exists at any loan size. The core point is that the box-spread rate is meaningfully better than margin loan rates — and at the moment, better than mortgage rates as well.

---

What this isn’t

This is not advice to never pay taxes. Taxes are eventually paid: when the loan is repaid, when assets are sold, or at death. The strategy defers, not eliminates.

The SALT torpedo may change. The $40,000 cap and phase-out are statutory and subject to amendment. A taxpayer depending on the torpedo math should run scenarios with and without that benefit.

This does not work for everyone. At low marginal rates, the math is less compelling. Someone in a no-income-tax state with modest investment gains sees much less benefit than Tony. The tax series covers the four scenarios in detail: see “Short-term gains, long-term gains, or none”.

Carrying debt has psychological costs. A large loan outstanding on a portfolio is genuinely stressful, and stress has real value. The financial math can be sound while the emotional math comes out differently.

---

If you want to actually do this

Understanding why selling is expensive is the first step. The next step is understanding how to borrow instead. Continue with “Borrow, don’t sell: the basic idea”.

When ready to talk to a firm that executes portfolio-backed borrowing, request a referral. I’ll send a referral code and an intro. Talk to your CPA first. See full disclosure.

---

Sources

1. IRC §1256 — Section 1256 contract treatment
2. IRS Topic No. 409: Capital Gains and Losses — Capital gains rate overview
3. IRS Net Investment Income Tax — 3.8% NIIT explanation
4. California Franchise Tax Board: Capital Gains — CA treats LTCG as ordinary income