Borrowing against your portfolio: buy, borrow, die

The strategy has a memorable name — buy, borrow, die — and a simple engine: appreciated assets are never sold (no capital gains tax), spending is funded by borrowing against them (loans aren't income), and at death the basis step-up erases the embedded gains while the estate repays the debt. It's how billionaires report tiny incomes while spending millions. Scaled-down versions are available to anyone with a meaningful taxable portfolio — along with scaled-down versions of the risks.

The tools

• Margin loans: borrow against securities at your brokerage, instantly, no application. Rates vary wildly by broker — from near the benchmark rate at discount brokers to benchmark-plus-3% or more at big-name firms. Negotiable at size.
• Securities-based lines of credit (SBLOCs): bank-offered lines pledged against your portfolio, often slightly cheaper and outside margin rules — but you can't use proceeds to buy more securities.
• Typical borrowing capacity: 50–70% of a diversified portfolio's value; far less against concentrated single-stock positions.
• Interest is generally NOT deductible when the loan funds personal spending (investment-interest deductions require investment use and itemizing).

The math versus just selling

The fine print that eats people

• Maintenance calls: if the portfolio falls enough, the lender demands cash or sells your positions — at the bottom, without asking which lots, creating exactly the taxable gains you were avoiding, at the worst prices.
• Rates float: the strategy penciled beautifully at 2% borrowing costs; at 7%+ it needs markets to cooperate every year. A decade of 8% loans against 6% returns is slow-motion liquidation.
• Leverage stacks on lifestyle: borrowing to spend means the debt grows while the collateral fluctuates. A 40% crash with 30% borrowed is a forced-selling event; the same crash with 10% borrowed is a Tuesday.
• SBLOCs are demand loans: the bank can cut your line or call the loan when conditions tighten — historically, precisely when you'd least like them to.
• The 'die' step requires actually keeping the assets until death, and estate-tax-exposed households face other tradeoffs between step-up and lifetime gifting.

A sane version for non-billionaires

1. Cap borrowing at 10–20% of the pledged portfolio — the level that survives a 50% crash without a call. The billionaire version works because their ratio is tiny, not because they're special.
2. Use it for bridges, not lifestyles: a house closing before a sale, a tax bill in a bad month, a short-term opportunity — weeks to a couple of years, with a named repayment source.
3. Shop the rate ruthlessly; the spread between brokers on the identical product is often 2–4 percentage points. At $500,000 borrowed, that's $10,000–$20,000/year for the same loan.
4. Pair repayment with tax planning: pay down the line by selling in years when the 0% or 15% capital gains brackets, or harvested losses, absorb the gains.
5. Keep the emergency fund anyway. A credit line secured by the same asset that's crashing is not an emergency fund; it's a correlation.

The three borrowing tools compared

Stress-testing before you borrow

The only borrowing level that matters is the one that survives the worst market you can imagine while the loan is out. The arithmetic: a lender requiring 50% maintenance forces action when the loan exceeds half the collateral's value. Borrow $100,000 against a $500,000 portfolio (20% loan-to-value) and the portfolio must fall below $200,000 — a 60% crash — before a call; that has happened roughly never for a diversified portfolio in a single episode. Borrow $250,000 against the same portfolio (50% LTV) and a 33% decline — an ordinary bear market — triggers forced selling at the bottom. Same tool, same borrower; the ratio is the entire difference between a convenience and a catastrophe. Run the test at double your intended borrowing too, because lines have a way of growing: the renovation overruns, the bridge extends, the repayment year slips. If the doubled number still survives a 50% crash, the plan has real margin. If not, the market will eventually administer the test for you, on its schedule, at its prices.

Finally, put the loan itself on your net worth statement and review it quarterly like any other position. Borrowed-against portfolios have a way of feeling unencumbered — the shares still show up, the dividends still arrive — and that illusion is exactly how a temporary bridge becomes a permanent fixture that compounds against you at whatever rates the next decade brings.

Rates, terms, and lending percentages here are typical 2025–2026 figures and move with the rate cycle — reprice everything before acting.

The bottom line

Buy, borrow, die genuinely works — for balance sheets where the borrowing is a rounding error and the timeline runs to the step-up. For everyone else, portfolio loans are a sharp, useful bridge tool: cheap flexibility at low loan-to-value, catastrophe at high. Keep the ratio small, the purpose specific, the rate shopped, and the exit named — and remember that the tax you're deferring is the second-biggest number in the equation. The biggest is what leverage does to a bad year.