There’s a financial strategy that gets a lot of breathless coverage and a lot of bad explanations. ProPublica wrote a long feature on it. Twitter (or X, depending on the year) brings it up every time a tech founder shows up in a Forbes list and pays no income tax. The shorthand is “buy, borrow, die.”
It’s real, and the math actually works, but only above a certain portfolio size and only when three other things line up. Below that size or without those conditions, BBD is worse than just selling stock and paying capital gains. People don’t usually walk through where the line is. So here.
The mechanics, in three sentences
You buy assets that appreciate (stocks, real estate, sometimes a concentrated position). When you need cash, you don’t sell. You borrow against the assets through a securities-based line of credit, a margin loan, or a pledged asset line, and you live on the borrowed money. When you die, your heirs inherit the assets with a stepped-up cost basis under IRC §1014, meaning the capital gains that accumulated during your lifetime are permanently erased.
The trick isn’t avoiding tax during your life. It’s compounding for forty years on assets you never had to sell, and then handing them over with the gain wiped to zero.
Why this only works above ~$5M
Below a certain portfolio size, BBD is a worse deal than just selling and paying long-term capital gains. There’s no fixed threshold (it depends on asset composition, state of residence, and other things), but as a rough heuristic, the math starts to work somewhere around $5M in pledgeable assets.
Three reasons.
1. The interest rate has to be lower than asset appreciation, after tax.
A securities-based line of credit (SBLOC) typically prices at SOFR plus 1-2%, so call it 5-7% in a normal-rate environment. If your portfolio compounds at 7-9% (a reasonable equity expectation), you’re net positive on the carry, but the gap is thin. If your assets are dragging at 4% or rates spike to 9%, the carry flips negative and you’re losing money to live on borrowed cash.
Below $5M, the loan amounts are small enough that you’re usually paying margin rates instead of negotiated SBLOC rates. Margin rates often run 10-13%. At those rates the entire strategy is upside-down on day one.
2. The capital gains avoided have to be larger than the interest paid.
If you sell $100k of stock with a $40k cost basis, you owe long-term capital gains on $60k. Federal LTCG tops out at 20% + 3.8% Net Investment Income Tax. Add California, New York, or Oregon and the marginal rate can hit 33-37%. So selling that $100k actually costs roughly $20k in tax and leaves $80k.
Borrowing $100k at 6% for ten years costs roughly $34k in interest if you never amortize. So the math says: if you can defer that sale long enough that compounding on the unsold $100k outpaces the interest, you win. At higher portfolio sizes you can also negotiate your borrow rate down to the floor, widening the spread.
3. You need enough liquidity to service the loan without forced sale.
This is where most armchair BBD plans fall apart. The strategy works only if you can pay interest and meet potential margin calls from somewhere other than the pledged assets themselves. Otherwise a market drawdown forces you to sell at the worst possible time, which is exactly the outcome BBD was supposed to prevent.
Below $5M, you typically don’t have a separate income stream large enough to service the loan in a 30% drawdown year. Above $5M, you’re usually pledging only 30-40% of total assets, leaving slack for service and a buffer for margin call risk.
Where people get it wrong
“Margin calls won’t happen to me.”
They will. Not most years. But across a 30-40 year retirement, you will live through at least one drawdown that triggers a margin call somewhere in your portfolio. The question isn’t whether, it’s whether you have a plan when it does. The people who blow up doing this almost always blow up because they pledged too much, not because the strategy itself failed.
“I’m living for free.”
You’re not. You’re paying interest. The interest is usually tax-deductible if the borrowed money is used for investment (and almost never deductible if used for personal spending), but you’re paying it. The strategy doesn’t eliminate the cost of your lifestyle. It defers it, with compounding on your side, into a future tax event that may never come (the step-up).
“The step-up is forever.”
The step-up exists today and has existed since the modern code, but it has been on the policy chopping block multiple times. Plans that depend on the step-up surviving for 40 more years are making a political bet, not just a financial one. A serious BBD plan models what happens if step-up is repealed mid-strategy.
“I should put everything in the pledged account.”
The opposite. Concentration is the killer. The right BBD setup usually pledges a fraction of total assets and keeps significant liquidity outside the pledge to handle margin events. The classic family-office structure runs a 30-40% loan-to-value at most, with a separate cash reserve sized to cover 12-24 months of expected interest plus a margin call buffer.
What the math looks like at scale
How Thermal models this
Buy-Borrow-Die is one of the eight withdrawal strategies in Thermal’s retirement engine. It models loan accrual, interest deduction (where applicable), and the all-important margin call probability across 10,000 Monte Carlo scenarios. The success rate it reports already includes the margin call events that wipe out a percentage of runs. That number is the one most online BBD discussions skip, and it’s the most important one.
I write more about strategies for compounded high-earner portfolios at High Earner Playbook. If your portfolio is in the range where BBD math actually works, that’s where the longer-form pieces live.