Margin, Leverage, and Why Blowups Happen

Leverage does not change your average outcome much. It changes your worst outcome enormously, and it removes your ability to wait. This guide works through the margin math, the maintenance call, leveraged ETF decay, and the narrow set of cases where borrowing to invest is defensible.

By Karanvir Singh, founder of Quantix Invest

What leverage actually does

Leverage means investing with borrowed money. If you put up $50,000 of your own and borrow $50,000, you control $100,000 of assets with $50,000 of equity. That is 2x leverage, and it does exactly one thing: it multiplies the percentage change in the assets into a larger percentage change in your equity.

Read the last row carefully. At 2x leverage, a 50% decline in the underlying asset takes your entire stake. The S&P 500 has fallen roughly 50% or more twice in the past twenty five years, in 2000 to 2002 and 2007 to 2009. A 2x levered investor holding through either of those did not lose half. They lost everything, and in practice they lost it earlier, because of the mechanism in the next two sections.

Margin accounts: initial and maintenance

A margin account lets you borrow from your broker using your securities as collateral. Three numbers govern it in the United States.

Regulation T initial requirement: 50%. Set by the Federal Reserve, this caps how much you can borrow to open a new position at half the purchase price. Put up $50,000, buy up to $100,000. That is the maximum leverage on a new stock purchase in a standard margin account: 2x.

FINRA maintenance requirement: 25%. After the trade, your equity must stay at least 25% of the market value of the securities. Drop below and you get a maintenance call.

House requirements: usually higher. Brokers set their own maintenance levels, commonly 30% to 40% for ordinary stocks, and much higher (50%, 75%, sometimes 100%) for volatile, low-priced, or concentrated positions. Critically, brokers can raise these without notice, at any time, including in the middle of a crash, and they do exactly that when volatility spikes. Your margin agreement says so. Many investors discover this clause the week it is used on them.

Two more facts that surprise people. Your broker can sell your positions without contacting you and is not obligated to give you time to deposit funds, even though the phrase "margin call" implies a phone call. And your broker chooses which securities to sell, not you.

The margin call: the formula and a worked example

The price at which a maintenance call is triggered has a clean formula for a single position:

Call price = Loan amount / (Shares x (1 minus maintenance requirement))

Or equivalently, per share: call price = (loan per share) / (1 minus maintenance rate).

Worked example. You have $50,000. You buy $100,000 of a stock at $100 per share, so 1,000 shares, borrowing $50,000. Your broker's maintenance requirement is 30%.

Step 1. Find the trigger price. Call price = $50,000 / (1,000 x (1 minus 0.30)) = $50,000 / 700 = $71.43.

So a 28.6% decline in the stock, which is an ordinary bad year for a single company and a normal bear market for an index, triggers the call. At that price your position is worth $71,430, your debt is still $50,000, and your equity is $21,430, which is exactly 30% of the position.

Step 2. Suppose it keeps falling to $65 before you act. Position value $65,000, debt $50,000, equity $15,000, which is 23.1%. To restore 30% equity you can either deposit cash or let the broker sell.

Cash deposit needed: you need equity of 30% of the position. Depositing cash pays down debt dollar for dollar. Let D be the deposit. Equity becomes $15,000 + D, position stays $65,000, so ($15,000 + D) / $65,000 = 0.30, giving D = $19,500 minus $15,000 = $4,500.

Or, shares sold: selling shares reduces both position value and debt equally, leaving equity unchanged at $15,000. You need the position down to $15,000 / 0.30 = $50,000, so you must sell $15,000 of stock, about 231 shares at $65. Note the leverage in the arithmetic: every dollar of deposited cash does the work of roughly 3.3 dollars of forced selling.

Step 3. The damage. If the stock later recovers to $100, the investor who deposited cash is whole and then some. The investor who was liquidated at $65 owns only 769 shares. At $100 those are worth $76,900 against $35,000 of remaining debt, for equity of $41,900. The unlevered investor who never borrowed would be back at $100,000 on a $50,000 stake, having simply held 500 shares bought with cash and worth $50,000. The forced seller converted a temporary paper loss into a permanent one.

Why forced selling is the real killer

An unlevered investor who is wrong about timing has one option that is enormously valuable: waiting. Markets recover, on average, and the investor who can sit still captures that recovery. Leverage sells that option away. When the maintenance line is crossed, you must produce cash or the position is closed, and by definition this happens after prices have already fallen.

Worse, everyone's maintenance line is crossed at roughly the same time. That is what makes leverage a systemic amplifier rather than an individual one. Falling prices trigger calls, calls trigger forced sales, forced sales push prices lower, which triggers more calls. This loop is the recognizable skeleton of essentially every historic blowup: the 1929 crash (where 10% margin was legal and common, which is why Regulation T exists), Long Term Capital Management in 1998, the mortgage leverage of 2008, the volatility-linked products that collapsed in February 2018, and the family office Archegos in 2021, which lost roughly $20 billion of equity in about two days when swap counterparties liquidated concentrated levered positions.

The pattern is always the same and it does not depend on the asset. It depends on the borrowing.

The cost of carry

Margin loans are not free, and the rate is often much worse than advertised. Broker margin schedules are tiered, with the best rates reserved for very large balances. Small balances commonly pay a spread of several percentage points over a benchmark rate, and rates float, so a rate that looked tolerable when short rates were near zero becomes punishing when the Federal Reserve raises them, which is exactly what happened between 2022 and 2024.

Worked example. A $50,000 margin loan at 9% costs $4,500 a year. On $50,000 of your own equity, that is a 9% headwind on your capital before the investment does anything at all. If your portfolio returns the long-run stock average of roughly 10% nominal, the levered position earns 10% on $100,000, which is $10,000, minus $4,500 interest, leaving $5,500 on $50,000 of equity, an 11% return. You took double the risk of ruin to gain about one percentage point.

That is the whole trade in one calculation. Leverage pays you the spread between the asset's return and your borrowing cost, multiplied by how much you borrowed. When your borrowing cost is close to your expected return, there is almost no spread to collect, and you are absorbing all of the downside for very little compensation. Institutions that use leverage well borrow at rates retail investors cannot get, on assets with far lower volatility than stocks.

Path dependence: the same average, a different ending

Unlevered compounding already punishes volatility: lose 50% and gain 50% and you are at 0.5 x 1.5 = 0.75, down 25%. Leverage magnifies that arithmetic penalty rather than the average return, which is why levered returns fall short of the multiple you expect in choppy markets.

Consider two days: the asset falls 10%, then rises 11.11%, ending exactly where it started.

• Unlevered: 0.90 x 1.1111 = 1.000. Flat, as expected.
• 2x daily rebalanced: day one you lose 20%, day two you gain 22.22%. So 0.80 x 1.2222 = 0.978. Down 2.2% on an asset that did not move.
• 3x daily rebalanced: 0.70 x 1.3333 = 0.933. Down 6.7%.

Nothing went wrong. No fees, no interest, no bad luck. This is pure arithmetic, and it is the single most important fact about daily-reset leveraged products.

Leveraged and inverse ETFs

A leveraged ETF promises a multiple of an index's return for a single day, and then rebalances its exposure at the close so that it starts the next day at the stated multiple again. This is stated plainly in every prospectus. It is also the part almost nobody internalizes.

The daily reset means the fund must buy after up days and sell after down days, mechanically, to maintain its ratio. That is a systematic buy-high sell-low policy imposed by the product's design. Over a single day it delivers what it promises. Over any longer period, the result depends on the path the index took, not just its start and end points.

They also carry real costs on top of decay: expense ratios frequently in the 0.90% to 1.00% range versus roughly 0.03% for a plain index fund, plus the embedded financing cost of the swaps and futures that create the exposure, which rises with short-term interest rates.

Inverse funds have all the same issues plus one more: the thing they are short (the stock market) has a positive long-run expected return, so their expected value trends toward zero over long holding periods regardless of decay. A "minus 1x" fund is not a hedge you can set and forget. It is a decaying short position.

Volatility decay, worked out

Take a realistic bad year. The index falls 40% over six months, then recovers all of it over the next six months, ending flat for the year. Assume, for a clean illustration, two clean halves rather than daily moves, which understates the decay but shows the shape.

Index: 0.60 then x 1.6667 = 1.00. Flat.

3x, tracked in two steps: down 40% means the 3x product is down 120%, which is impossible, so a real fund's daily reset prevents total wipeout by shrinking exposure continuously. Modeled properly with daily moves, a 40% drawdown at 3x typically leaves the fund down roughly 75% to 85%, depending on the path. Suppose it ends the first half down 80%, at 0.20. It then needs the second half's 66.7% index rise, tripled to about 200%, to get to 0.20 x 3.0 = 0.60.

The index is flat for the year. The 3x fund is down 40% and needs a 67% gain just to break even from there. This is not a hypothetical: real 3x products have repeatedly gone through year-long stretches where the underlying index finished higher and the fund finished lower, and several have executed reverse splits after multi-year drawdowns of 90% or more.

Other leverage in disguise

Margin is the obvious form. Several other things are leverage wearing different clothes.

Options. Buying a call gives you exposure to many shares for a fraction of their price, which is leverage by definition. The added feature is a hard expiration: you can be directionally right and still lose 100% because the move arrived a week late. Selling naked options is worse, converting a defined premium into an undefined loss. Options have legitimate uses for hedging and income, but a long call bought for upside is a levered, expiring bet, and it should be sized like one.

Futures. Standard futures contracts carry leverage ratios far above what Regulation T permits in a stock account, with initial margins often in the 3% to 12% range of notional value. Futures are marked to market daily with cash settlement, so an adverse move produces an immediate cash demand rather than a paper loss.

Portfolio margin. Available to approved accounts above a balance threshold, it sets requirements based on modeled portfolio risk rather than fixed percentages, which for a diversified book can permit leverage well above 2x. It is a professional tool that works precisely until correlations move to one in a crisis, at which point the model that granted the leverage revises its opinion overnight.

Crypto perpetual futures. Offshore venues have advertised leverage of 50x and 100x. At 100x, a 1% adverse move liquidates the position. These are not investments in any meaningful sense; they are levered bets with fee structures and liquidation engines that profit from the churn.

Your mortgage, and borrowing against your home. A mortgage is leverage on a house, and for most households it is the largest leverage they will ever take. That is generally reasonable because the loan is long-term, fixed-rate if you choose, non-callable, and secured by an asset you use. A home equity loan taken out to buy securities is a different animal entirely: you have converted a stable claim on shelter into a levered market position, and the downside includes your house.

Company stock plus a concentrated job. Not borrowing, but the same fragility. If your salary, your bonus, your options, and your retirement account all depend on one employer, a bad quarter hits every account at once. Correlated exposure is functionally leverage on a single risk.

Short selling and unlimited loss

Short selling borrows shares, sells them, and hopes to buy them back cheaper. It is inherently a margin activity, and it has an asymmetry that runs the wrong way. A long position can fall at most 100%. A short position can lose an unlimited amount, because a stock's price has no ceiling.

Three additional mechanics make it harder than it looks. You pay a borrow fee to whoever lends you the shares, and for the most heavily shorted names that fee can run tens of percent per year. You owe any dividends paid while short. And your lender can recall the shares at any time, forcing you to close at a moment you did not choose.

The short squeeze dynamic is what happens when all of that fires at once. Rising prices generate margin calls on shorts, forced covering means buying, buying pushes prices higher, and the loop runs until the shorts are out. The January 2021 episode in a heavily shorted retailer produced losses of several billion dollars at individual funds within days. The correct lesson is not that short sellers are villains or victims, but that a strategy with unlimited loss and a financing cost requires a level of position sizing and risk management that almost no individual has in place.

When leverage is ever sane

Blanket rules are usually wrong, so here is the honest version. Leverage is defensible when several conditions hold at once, and it is dangerous when even one is missing.

1. The loan cannot be called. A fixed-rate mortgage cannot be called because prices fell. A margin loan can. This single distinction explains why leveraged homeownership is normal and leveraged stock speculation is not.
2. The borrowing cost is well below the expected return. A 3% mortgage against a diversified portfolio has a plausible spread. A 9% margin loan against equities does not.
3. The leverage is modest and the asset is diversified. Academic proposals for lifecycle leverage, notably the work of Ayres and Nalebuff on diversifying across time, argue that a young investor with decades of future savings could reasonably use leverage capped around 2x on a broad index early in life, precisely because their real exposure to markets is small compared to their lifetime savings. Even those authors emphasize a hard cap and a rapid unwind. The strategy also had brutal real-world stretches, and the version people actually implemented in 2008 was not the one in the paper.
4. You can service the debt from income regardless of the market. If the market falling means you cannot pay, you are not leveraged, you are exposed.
5. The position is small relative to net worth. Leverage on 10% of your assets is a decision. Leverage on all of them is a wager on not experiencing a bad decade.

Most retail leverage fails two or three of these tests simultaneously: a callable loan, at a high floating rate, on a volatile or concentrated asset, sized large. That combination has a well-documented ending.

It is also worth saying the obvious alternative. If you want more expected return, you can usually get most of the way there by holding a higher stock allocation without any borrowing at all. Going from 60% stocks to 90% stocks increases expected return and risk substantially, has no interest cost, no margin call, no daily reset, and no possibility of losing more than you invested. Ask honestly whether the thing you want from leverage can be had that way first.

Common mistakes

Treating a 2x leveraged fund as "the index, but more." It is a different instrument with a different payoff function. Path matters.

Assuming a margin call comes with a phone call and a grace period. Brokers may liquidate immediately, without notice, and choose the positions themselves.

Forgetting that maintenance requirements can be raised mid-crisis. The rules you were levered under are not the rules you will be liquidated under.

Using margin for "just a few days" and then not closing it. Short-term margin use has a way of becoming a permanent balance that quietly compounds interest for years.

Confusing leverage with conviction. Being more certain does not change the probability distribution. It changes only how much you will lose when you are wrong.

Ignoring the correlation between your leverage and your income. Margin calls arrive in recessions. So do layoffs. Assuming you can deposit cash in a downturn assumes the downturn spares you personally.

Believing you will "just get out" before it gets bad. Every levered blowup features people who intended to exit early. The mechanism that stops them is the same one that created the losses: prices gap, liquidity vanishes, and the broker acts first.

This guide is educational material, not individualized financial advice, and nothing here is a recommendation to use margin, options, or leveraged products.

Related guides: How Markets Work, Behavioral Investing, Asset Allocation and Diversification, Rebalancing Your Portfolio