Margin, Options: Two Ways the Same Capital Disappears

You've probably seen how a home loan works. Put down ₹20 lakh on a ₹1 crore flat, borrow the rest, and you own the whole flat — not just the fifth of it you actually paid for. Prices rise 10% and you've gained ₹10 lakh on ₹20 lakh of your own money: a 50% return. Prices fall 10% instead, and you've lost that same ₹10 lakh — also 50%, except now it's half a loss, not half a bonus. And through all of that, nobody can force you to sell. You bought the whole flat. You can hold the whole flat, for as long as you can afford to.

Margin trading and F&O — futures and options — run on that same multiplying logic, just faster, and without the part where nobody can force you out. Two very different mechanisms sit inside that one sentence, and most explanations of "why people lose money in F&O" collapse them into one story. They're not one story. They're two, and a trader can fail either one without ever touching the other.

What margin trading and leveraged F&O actually are

A cash market purchase

• Pays the full price upfront
• Gains or loses exactly what was invested, ₹ for ₹
• Can simply wait out a bad week — nothing forces an exit

A leveraged position — margin trading, futures, or an F&O position bought on margin

• Pays a fraction of the position's value upfront (the margin)
• Gains or loses computed on the full position's value, not the fraction paid
• Must keep that fraction topped up, or the position gets closed without being asked

On the NSE, the margin required to hold one index futures contract is typically a fraction of its full value, set by an exchange risk model called SPAN that widens or narrows the percentage as volatility rises and falls — post that fraction, and a trader controls the whole contract. The same idea runs internationally under different names: US brokers extend Regulation T margin, letting a trader put down as little as 50% of a stock's price and borrow the rest, while CME futures margin — like NSE's SPAN — runs to a single-digit percentage of contract value for the most liquid contracts.

The first way it disappears: the cushion runs out before the bet does

Leverage doesn't just multiply a gain. It multiplies a loss by the identical number, and that number cuts in whichever direction the market decides to go, indifferent to which one the trader was hoping for.

Say a trader posts ₹1,00,000 as margin to control a futures position worth ₹10,00,000 — a leverage of 10x, in the range SPAN margin typically implies on an index future. The underlying moves 5% in the trader's favour: ₹50,000 lands in the account, a 50% return on the ₹1,00,000 actually risked, settled in cash daily rather than saved up for later. The underlying moves 5% the other way instead: the same ₹50,000 is gone — also 50%, except this time it's half the capital, not half a windfall.

Run the identical arithmetic on a CFD — a Contract for Difference, the retail leveraged product popular across Europe and the UK — where, before regulators intervened, a trader could open a position on as little as 3–4% margin: roughly 30 times leverage. A 3% adverse move against a position leveraged 30 times doesn't cost 3% of capital. It costs all of it.

That's the arithmetic. The part that actually forces the loss to happen is separate from the arithmetic, and it's the part most explanations skip.

In India, a broker's Risk Management System — the RMS — watches every client's margin in close to real time. Let that ₹1,00,000 cushion fall below what the exchange requires to keep holding the position, and the RMS doesn't wait for the trader to notice or decide. It squares off the position itself, at whatever price the market is offering in that moment — rarely the price the trader would have chosen to exit at, and often the worst price of the day.

The EU's 2018 CFD rules wrote the identical mechanism into regulation and gave it a name: the margin close-out rule. A provider must automatically close a retail client's position once their margin cushion falls to 50% of what was required to open it — again at market price, again without the client's say.

There's one difference between the two worth naming, because it decides whether a bad trade merely wipes out a trader's capital or actually puts them in debt. Alongside the margin close-out rule, European regulators required CFD providers to offer negative balance protection — a guarantee that a retail client can never lose more than the money already in the account, with the provider absorbing anything beyond that. India's F&O market carries no equivalent guarantee. If a price gaps hard and fast enough — an overnight move, a sudden sharp fall — the RMS may not manage to close the position before the loss has already run past the margin posted, and the shortfall becomes a debit balance the trader owes the broker, not the other way around.

The second way it disappears: the clock was never on your side

The first way runs on capital and force. The second doesn't touch capital at all — it runs on a clock that started the moment the position opened.

None of the above touches the largest share of retail F&O activity in India at all — because most of it isn't leveraged futures or margin-traded stock. It's buying options. A brokerage's analysis of NSE's own public turnover data — not the SEBI loss study itself, but drawn from the exchange's disclosed volumes — found that 76% of individual options traders never once traded a single stock option, running their entire F&O activity through weekly Nifty and Bank Nifty index contracts instead, and that roughly 97% of that index-options turnover happens within the final week before expiry.

Think of a block of ice bought to cool a drink. It doesn't matter whether the drink gets poured in the next five minutes or the next five hours — the ice melts at roughly the same pace either way, whether or not it's actually doing its job yet. By the time you're ready to use it, part of what you paid for is already gone, and no amount of being right about needing ice later gets that part back.

An option buyer's loss looks nothing like a margin call. The buyer's worst case is capped at the premium paid, full stop — there's no RMS, no forced square-off, no debit balance. But that cap doesn't mean the odds are even, because a premium is quietly doing exactly what that ice does. It decays a little every single day the position is held, whether the price moves or not — so a buyer isn't just betting on direction. They're betting the move happens far enough, and fast enough, to outrun a cost that's melting away regardless. Buy a contract expiring in three days, guess the direction correctly, and still lose money, because the move wasn't large enough to clear the premium and decay combined before expiry arrived.

Academic research on the same behaviour in US retail markets — buyers of short-dated calls, often around earnings announcements — found retail investors systematically bidding option prices above what the underlying's actual realized volatility would justify, then losing money on the position even when their direction was right, because they overpaid for the option in the first place and the clock never stopped running against that overpayment. Different country, different index, same trio of mistakes: short-dated contracts, a decaying premium, and a price paid that assumed more movement than the market was likely to deliver.

What the data actually shows

SEBI's first study of individual traders in the equity F&O segment, released in January 2023, found 89% of them lost money in FY22 alone, at an average loss of ₹1.1 lakh — against an 11% minority who profited, averaging ₹1.5 lakh each. An updated study released in September 2024, extending the window to FY22–FY24, found the losing share had climbed to 93%, and the combined net loss across those three years came to more than ₹1.8 lakh crore.

Europe's regulators found almost the identical shape in a completely different leveraged product. Reviewing national data across EU jurisdictions ahead of its 2018 intervention, ESMA found that 74–89% of retail CFD accounts lose money, with average losses per client ranging from €1,600 to €29,000 depending on the jurisdiction and provider — the finding that led regulators to cut maximum retail leverage from as high as 30:1 down to as low as 2:1 depending on the underlying's volatility, and to mandate the close-out rule and negative balance protection described above.

Two products, two countries, two mechanisms that don't otherwise resemble each other, tangled together inside each number. And in both places, regulators looked at the data and concluded the same large majority of retail participants were losing.

Why this matters for a Business Analyst

Think of two insurance policies sitting in the same folder

One has a fixed, known deductible — whatever happens, the most you're out is that number, agreed in advance. The other has no cap on the payout at all — it can run to any size the underlying event turns out to demand. File them both under "insurance" in the same tracking sheet, and the sheet stops telling you anything useful about which policyholder is actually exposed to catastrophic loss.

"The desk's F&O book lost ₹2 crore this quarter" is the same kind of sentence. It tells you the total. It doesn't tell you whether that total came from option buyers whose worst case was always capped at the premium they paid, or from leveraged futures and margin positions where the loss had no ceiling until an RMS forcibly found one. A risk or P&L system that reports F&O exposure as a single blended number — instead of separating capped, premium-at-risk positions from uncapped, margin-at-risk ones — will understate exactly the risk that matters most: how much of that ₹2 crore could still be growing the moment the report was generated, versus how much was already finished the day the premium was paid.

Build the margin engine on top of that same blur, and it inherits the error twice over — once in how it classifies risk, and again in how it models the speed of a loss. A margin check built to catch an ordinary, orderly price move has time to find a shortfall and route an exit order before the loss compounds. It has no such time when a price gaps past the margin cushion in a single tick, faster than any exit order can land near the price the system last checked. A client-facing statement that reports "position auto-squared, loss capped at margin" needs an honest, separate field for the case where it wasn't — a debit balance carried forward, not absorbed by anything the client had already posted.

Lighthouse Insight

Go back to that flat.

Nobody can force you to sell it on a bad week. You bought the whole thing; you can hold the whole thing, for as long as you can afford to, waiting however long a bad week needs to turn into a good year.

A leveraged F&O position was never given that patience. The moment its cushion runs out, the choice to wait stops being the trader's — an RMS decides, at whatever price the market is offering, not the one the trader would have picked. An option bought outright keeps that patience for exactly as long as its premium survives the decay ticking against it — right up until expiry arrives and decides the question anyway, whether or not the trader was ever wrong about direction.

That's the answer to how people really lose in F&O and margin trading. Not one mistake, repeated by nine traders out of ten. Two entirely different mechanisms — one that takes away the choice of when to exit, one that was never going to wait — arriving, in both India and markets an ocean away, at almost exactly the same result.