How Long Can a Margin Loan Last? Five Models, Five Different Answers

A margin loan does not have a term. Nobody calls it in on a date, which is exactly why the question "how long can I carry this?" has no answer you can look up — the loan ends when the reading against your collateral reaches a level that forces a sale, and that depends on your rate, your income and what the market does. Put honest assumptions in and you do not get a year. You get a spread decades wide. This post shows the spread on one book, and argues that the width of it is the finding rather than a failure to compute.

The book, and the one line that governs it

The same book throughout, so every figure below is comparable. Illustrative, shaped like mine, single currency: $3,750,000 of gross securities value against a margin loan of $975,000. No cash line — the loan is the negative settled balance.

Our reading is margin loan ÷ gross securities value. Not net of the loan, not including outside cash. That is 26.00%, which sits in Clear, the calmest of the four zones we govern to. The zones themselves, and what each instructs, are set out in the margin pillar on using margin safely. For this post you need only one of them: Forced, at 40% and above, is the end of the road — the zone where a reduction is the only recommendation on the page, and the one zone that does not wait for a second weekly reading.

Worth stating plainly, because everything below is measured against that 40%: it is our line, not a broker's and not a regulator's. A 40% loan against gross securities still leaves 60% equity, against FINRA's general minimum that "Generally, a customer's equity in a margin account that holds margin stocks only must not fall below 25 percent of the current market value of the long securities." Reaching Forced on our reading is not a margin call. It is the point at which we think you have run out of room to choose — and a broker's own house requirement, which can be raised without advance notice, is what actually liquidates anyone.

This post assumes the facility already exists and you are carrying it. If you are still deciding whether to open one, the brokerage family bank covers the underwriting; if you arrived from the estate-planning angle, buy borrow die and the catch in its middle leg is the definitional version of the same problem. And if what you are really asking is whether it should have been a margin loan at all, that comparison is its own post. And if the question is how long the borrowing lasts set against drawing down instead, you can compare the two paths on your own numbers.

At an assumed rate of 6.00%, the first year's interest on $975,000 is $58,500. Who pays it? Not you, by selling — that would defeat the point. The prior question, whether the book pays its own interest, is where this starts. Dividends, coupons and option premium are cash generated inside the book without a sale, and they credit against the debit automatically. Which leaves one line worth knowing exactly, and it is exact, because it is algebra rather than projection:

Break-even yield = your reading × your rate. Below that yield, your debit grows every year on its own.

On this book: 26.00% × 6.00% = 1.56%. That is the income yield, on the whole book, at which realised income exactly meets the interest bill — the same $58,500. It assumes nothing about the market and nothing about time. It is stated before tax; tax on that income raises the yield you actually need, and by how much is a question for your accountant.

One caveat before leaning on premium to clear that line. It is not the same quality of income as a dividend: short options carry margin requirements a loan-to-collateral reading does not capture, a sharp fall is exactly when short puts are assigned and the assignment enlarges the debit, and assignment on a short call is itself a sale. Treat a premium-funded pass as a floor, not a guarantee.

What does break-even actually buy you?

Less than the word suggests. At or above break-even the debit is static or shrinking at the rate you are paying today — and that last clause is the whole caveat, because the rate is variable.

Break-even is a coverage ratio of exactly 1.0, and in our bands that is Thin — the weakest band that clears. Strong starts at 1.5. Clearing the line is not the same as clearing it with room, and the ratio behind it, the Margin Coverage Ratio, is set out with its full bands in the family bank pillar.

Here is what Thin actually buys you. Take a book sitting exactly at break-even, income unchanged in dollars, and move the rate from 6.00% to 7.00%. It reaches Forced in year 24. Nothing else changed — no crash, no extra borrowing, no missed dividend. One point of rate turned a book that was covering itself into a book on a 24-year clock.

How long does a book with no income last?

Below the line the shortfall is added to the debit and next year's interest is charged on the larger balance. Take the harshest honest case first: a book producing no realised income at all, in a market that does nothing, with no withdrawals.

Year	Loan	Securities	Reading	Zone
0 — today	$975,000	$3,750,000	26.00%	Clear
2	$1,095,510	$3,750,000	29.21%	Clear
3	$1,161,241	$3,750,000	30.97%	Harvest
7	$1,466,039	$3,750,000	39.09%	Freeze
8	$1,554,002	$3,750,000	41.44%	Forced

Read the two rows before the end. In year 2 the book reads 29.21% and is still Clear. In year 3 it reads 30.97% and has left it. Nothing happened in between — no decision, no trade, no bad news. That is exactly how this creeps up on somebody: the year you cross out of Clear looks identical to the year before it, and the only thing that changed is a number nobody was reading.

Why year 8 is worth less than it looks

Now hold on to that "year 8," because it is the number a careless version of this post would headline. It buries two assumptions the strategy it serves actively contradicts: a market that never rises, and an owner who never spends. Surface them and the answer moves by decades in both directions.

Five models, one book, one rate. The answers run from five years to never, and not one of them is a cheat — each is a set of assumptions a reasonable person would defend. The last row is the one worth sitting with: if the collateral compounds at the same rate the loan does, the reading never moves at all. That is the entire promise of borrowing against a growing asset, stated as arithmetic, and it is true right up until the year the growth does not arrive.

This is why I will not give you a year, and why you should be wary of anyone who does.

A single number here is not a calculation. It is four hidden assumptions wearing one figure's clothes.

What does the rate do to all of this?

The 6.00% above is explicitly an assumption, not a broker quote, and I will not name anyone's rate. Yours is variable, it moves with the policy rate, and it differs sharply between brokers. Hold everything else on the zero-income track and vary only the rate: 5.00% reaches Forced in year 9, 6.00% in year 8, and 8.00% in year 6.

In my experience the rate is also the field most people cannot fill in from memory — which is a remarkable thing to be true of a cost that compounds against you. If you have to go and look yours up, that is not a failure. It is the finding.

A convention worth stating, since it runs the other way: the models above compound annually, and brokers typically accrue monthly, which reaches the line marginally sooner. The years above are, if anything, generous.

What number goes on the other side?

The denominator I use is the one the reading already uses: gross securities value. Not equity net of the loan, not the income-producing part of the book on its own. Mix the bases and the percentage lies to you while the dollars sit still. A book realising $45,000 reads 1.20% on gross, under the 1.56% line, short of the $58,500 interest bill by $13,500. Move only the income to equity net of the loan — $2,775,000 — and it reads 1.62%, which looks like it clears. It does not: on that basis the reading is 35.14% and the line is 2.11%, so the same $45,000 still fails, and the shortfall is still $13,500. Either basis gives the same dollar answer. Only a mismatched one gives a different verdict.

The numerator is realised cash over the trailing twelve months, and it is narrower than it sounds. Dividends and coupons on the date they are paid. Option premium once the contract is finished — at expiry, buyback or assignment, not when it was written. Realised capital gains do not belong in it at all: selling is the thing this line exists to avoid, so counting the proceeds as income would let a book pass by doing the one thing that would have ended the argument. (Trailing, not planned — the Margin Coverage Ratio asks the forward question.)

And both sides are stated before tax, because the line above is. Tax is a question for your accountant; what matters here is that the two figures are stated on the same basis.

Two limits, both mine. Twelve months of income divided by today's value is a flow over a snapshot — a book that grew through the year is measured harshly and one that shrank is flattered. And clearing the line is still only Thin. It also moves against you: a 20% fall takes this book's reading to 32.50% and its break-even to 1.95%, so the hurdle rises in exactly the year the income is least likely to.

What a governed book does instead

None of the five models is a forecast, and none of them is what I actually run on. The point of laying them side by side is that the spread — five years to never — is wider than any decision you would make on a single one of them. So the operating answer is not to pick the best model. It is to stop needing one.

A reading taken on a cadence replaces the forecast. You do not need to know which year the book reaches Forced if you can see the reading move, in the same format, every week — because the thing that makes this dangerous is not the arithmetic, which is trivial, but that the arithmetic runs silently while nobody looks. Year 2 at 29.21% and year 3 at 30.97% feel identical from inside. They are not.

Two honest limits on that, both mine. A weekly reading is weekly: it cannot see an intraday move, and a book that gaps through a zone between two Fridays will be read after the fact, not during. And the two failure paths are correlated — a deep fall is also when dividends get cut and premium-writing capacity shrinks, so the model that assumes a flat market and steady income is describing a world where the two are independent, and they are not.

Frequently asked questions

How long can a margin loan last?

There is no term on a margin loan — it lasts until the loan measured against your collateral reaches a level that forces a sale, which depends on your interest rate, the income your holdings realise, and what the market does. On one illustrative book of $3,750,000 in securities against a $975,000 loan at 6.00%, five defensible sets of assumptions give year 5, year 8, year 16, year 25, and never. The spread is the honest answer; a single year is four hidden assumptions wearing one figure's clothes.

What is break-even yield on a margin loan?

Break-even yield = your reading × your rate, where the reading is the margin loan divided by gross securities value. It is the income yield, across the whole book, at which realised income exactly meets the annual interest bill. On a book reading 26.00% at a 6.00% rate it is 1.56%, before tax. Below that yield the shortfall is added to the debit and next year's interest is charged on the larger balance, so the reading climbs even in a completely flat market.

Can dividends and option premium cover margin interest?

Often, yes — dividends, coupons and realised option premium are cash generated inside the book without a sale, and they credit against the debit automatically, so carrying a loan does not force the interest to capitalise. But premium is the weaker form of that income: short options carry margin requirements a loan-to-collateral reading does not capture, a sharp fall is exactly when short puts are assigned and the assignment enlarges the debit, and assignment on a short call is itself a sale. Treat a premium-funded pass as a floor, not a guarantee.

Is covering the interest enough?

It clears the line, but only just. Break-even is a coverage ratio of exactly 1.0, which in Incomestead's bands is Thin — the weakest band that clears. Strong starts at 1.5. Because the rate is variable, a book sitting exactly at break-even is covering itself only at the rate it pays today: on the same illustrative book, moving the rate from 6.00% to 7.00% with income unchanged puts it on a path to Forced in year 24, with nothing else having changed.

How do I work out my book's income yield?

The way I do it: divide the cash the holdings realised over the trailing twelve months by gross securities value — the same denominator the margin reading uses. Dividends and coupons count on the date paid; option premium counts only once the contract is finished, not when it was written. Realised capital gains do not count at all — selling is the thing the line exists to avoid. Use gross rather than equity net of the loan: mixing the bases changes the percentage but not the dollars, and it is the percentage people read. On the illustrative book above, $45,000 of realised income is $13,500 short of the $58,500 interest bill either way. I am a practitioner, not an adviser; this is how I measure my own book.

Stefano Starkel is a practitioner, not a licensed adviser, and runs a leveraged portfolio under the governance process described here — which is also what Incomestead's software does; weigh the argument accordingly. This article is education, not personalised financial, tax or legal advice. Margin borrowing amplifies losses as well as gains and can result in the forced sale of your securities without prior notice; in a sharp decline it is possible to lose more than you invested and to owe the broker the difference. Every figure here is illustrative and computed for this article; the interest rate, the income yield and the growth rates are assumptions, not quoted or forecast figures, and no model here is a prediction. Nothing here is a recommendation to borrow any particular amount, or at all.

Incomestead recommends. You decide.