Settlement Mechanisms: Cash vs. Physical

The rule in the Indian market is clean and worth memorising in exactly this form: index derivatives are cash-settled, single-stock derivatives are physically settled. Buy a Nifty or Sensex option and hold it to expiry, and the exchange computes the difference between the strike and the final settlement price, multiplies by the lot size, and moves rupees between the two accounts. Nothing else happens. You never own anything and you never have to fund anything beyond the premium you already paid.

Buy a call on an individual company in the F&O list and hold it to expiry in the money, and the exchange does not hand you a difference. It hands you a purchase. You are obliged to buy the shares at the strike price, in the full lot quantity, and pay for all of them. The same applies in reverse to a put buyer, who is obliged to deliver shares, and to both sides of a stock futures contract, where the long takes delivery and the short gives it. NSE moved all single-stock derivatives to compulsory physical settlement in a phased rollout completed in 2019.

The reason for the split is partly logistical and partly regulatory. An index cannot be delivered — settling a Nifty contract physically would mean transferring weighted fractions of fifty companies, which is not a thing that can be done. Physical settlement on single stocks, by contrast, was introduced deliberately: it ties the derivatives price back to a real transferable asset and removes the incentive to run large speculative positions in a name with no intention of ever engaging with the underlying. It works. It also means an expiry that used to be a cash event is now an ownership event.

A settlement needs one agreed number, and it is not the last traded tick. A single trade at 3:29:58 could be off the true level, and on a contract with lakhs of crores of open interest, a manipulable closing tick is an invitation. Exchanges therefore derive the final settlement price from an averaging window at the end of the expiry session rather than from a single print. For the index, the closing value is computed as a weighted average of the index over a defined closing window rather than the last quote; the exact window and method are stated in the contract specification.

Two practical consequences follow. First, the number you can see on your screen at 3:30 pm is not necessarily the number your contract settles at. An index can print 24,512 at the close and settle at 24,498, and on a strike sitting right at that boundary the difference decides whether your option is worth something or nothing. Second, the averaging window is exactly why the last half hour of an expiry session behaves the way it does — the settlement price is being formed in real time, and every large participant with a position at a nearby strike has a reason to be active in it.

For single-stock contracts the same logic applies to determine whether a contract is in the money, but the settlement itself is then physical. The final settlement price of the stock decides which contracts get exercised and at what reference; the delivery obligation that follows is struck at the strike price, not at the settlement price. Do not assume the two are interchangeable. The expiry-day trading module works through how the closing window shapes price behaviour in that final stretch.

Take an illustrative case. Suppose SBIN is trading near ₹820 and its F&O lot size is 750 units — both figures are made up here to show the arithmetic, and the real lot size must be read off the NSE contract specification. You buy one lot of the ₹820 call for a premium of ₹14. Cash out of your account: 14 × 750 = ₹10,500. That ₹10,500 is what the trade cost you, and for most of the contract’s life it is also the entire extent of your involvement.

Now let it expire in the money by ₹2. The intrinsic value is 2 × 750 = ₹1,500 — a small win on paper. But the contract does not pay you ₹1,500. It obliges you to buy 750 shares of SBIN at ₹820, which is 750 × 820 = ₹6,15,000. That is roughly fifty-eight times the premium you paid, and it must be funded. If the money is not in the account, the position becomes a settlement shortfall with penalties attached. A trade sized as a ₹10,500 punt has become a ₹6.15 lakh obligation overnight, on a ₹2 move.

The way out costs nothing and takes one order. Square the option off in the market before the close on expiry day and no delivery obligation is ever created. You capture whatever the option is worth, the contract is extinguished, and the exchange has nothing to deliver to you. This is why experienced traders in single-stock derivatives treat the final sessions as a hard deadline rather than a decision point. The obligation is not a risk you manage; it is an event you avoid.

Because the clearing corporation can see the obligation coming. In the last stretch before a physically settled contract expires, positions that may convert into delivery attract an additional requirement — commonly called delivery margin — which is layered on top of the usual SPAN and exposure margin described in the margin rules module. It does not appear all at once. It steps up across the final sessions, climbing toward the point where the account is holding something meaningful against the full contract value.

The effect on a trader is that a position which was comfortably funded on Monday can be short of margin by Wednesday with the price unchanged. Nothing was traded. The margin schedule simply advanced. If the account cannot fund the increase, the broker squares the position off, and it does so in a contract whose order book is already thinning as everyone else exits for the same reason. That combination — a forced exit into a thin book — is the specific way this rule costs money.

How many sessions the ramp covers, and what proportion of contract value is blocked at each step, are set out in exchange and clearing corporation circulars and have been revised. This page does not quote them, because a stale schedule would be worse than none. Read the current delivery margin schedule for the instrument you are trading before the final week begins, and hold free cash against it rather than discovering it on the morning it lands.

Options on NSE are European style, which means they cannot be exercised early — only at expiry. That removes one whole category of surprise: no counterparty can assign you in the middle of the contract’s life. It also removes any element of choice. There is no button to press. Every contract that finishes in the money is exercised automatically by the exchange, whether the holder wanted the delivery or not, and whether they were watching the screen or not.

From there the sequence is mechanical. The final settlement price is determined from the closing window. All in-the-money contracts are exercised automatically. Positions are netted across your account, so a long call and a short call at the same strike in the same expiry offset rather than generating two separate obligations. The resulting net delivery obligations are then struck and passed into the equity settlement cycle, which in India now runs on a T+1 basis. Shares or funds move on the following settlement day.

That compression is the part traders underestimate. Under a T+1 cycle there is very little slack between the obligation being struck and the money or shares being required. There is no comfortable week in which to arrange funds. If the cash is not there, the shortfall is identified almost immediately and the auction machinery described in the next section engages. Plan funding before expiry, not after it.

The dangerous version of this is the in-the-money put buyer who does not own the stock. A put gives the right to sell; exercised, it becomes an obligation to deliver shares. If the shares are not in your demat account, you have a short delivery. The exchange does not simply cancel the trade — it has already guaranteed the counterparty. It goes into the market and buys the shares on your behalf through a separate auction session, and hands you the bill.

That bill is not the market price you were watching. Auction settlement prices are struck under rules designed to protect the party who was owed the shares, not the party who failed to deliver, and they can land well above the price that prevailed at expiry. On top of that sits a penalty on the shortfall. The combined outcome is regularly worse than the entire notional profit the option was showing, which is how a position that expired in the money produces a net loss substantially larger than the premium originally at risk.

The mirror case is the call buyer who is assigned shares he cannot pay for. Here the broker generally liquidates the delivered position at the next opportunity to recover the funds, at whatever the market offers on that day, plus charges. Neither outcome is exotic and neither is rare in expiry week. Both are entirely avoidable by squaring the contract off in the market, which costs one order and the prevailing spread.

Pin risk is the uncertainty that exists when the underlying finishes almost exactly at your strike. One rupee either side flips the contract between worthless and exercised, and you cannot know which until the settlement price is published after the close. For an index option the consequence is a small cash difference and nothing more. For a stock option it is the difference between a contract that lapses and a several-lakh delivery obligation that lands overnight.

It is worse for the writer than for the buyer, because the writer has no control at all. A short call sitting at the strike may or may not be assigned; if it is, shares must be delivered. A short put at the strike may or may not convert into a purchase obligation. The writer collected a premium and now carries a binary outcome decided by an averaging window they cannot influence, on a position they may already regard as effectively expired. This is the gamma behaviour covered in the gamma module, translated into a settlement obligation.

The defence is unglamorous. Do not hold a single-stock position near the strike into the close on expiry day. If the contract is at or near the money in the final session, close it. The residual value being given up on a near-the-money option in the last hour is small; the obligation being avoided is not. Traders who make an exception "just this once" because the premium looks like free money are the ones who discover the auction market.

It can, and the reason is that an exercised contract and a squared-off contract are charged on different bases. When you sell an option in the market, Securities Transaction Tax is levied on the premium. When an option is exercised at expiry, STT is levied on a settlement-linked base instead. Because that base is far larger than a small premium, an option that finishes only marginally in the money can generate charges that comfortably exceed the intrinsic value it delivers, turning a nominal win into a real loss on the ledger.

The precise base and rate have themselves been amended more than once, and the shape of the trap has changed with them. That makes it a genuinely poor thing to memorise. What is stable is the principle: exercise and square-off are charged differently, exercise is charged on the larger base, and the gap is dangerous specifically when intrinsic value is small. Confirm the current STT treatment for exercised options against the applicable Finance Act provisions or with a qualified chartered accountant before relying on any figure, including any you have read elsewhere. The expiry-day trading module examines the mechanics of that final session in more detail.

Three further specifications are worth confirming for whatever you trade rather than carrying in your head, because all three have moved: the lot size of the contract, the expiry weekday, and the delivery margin schedule. Nifty’s lot size stood at 75 units when this page was written, and both exchanges fix a single weekly expiry weekday which each has already revised at least once since the 2024 framework. None of that is a permanent fact. The contract specification page on the exchange site is the only source that is current by definition.