Macro Events and Expiry: Read the Right Clocks

A scheduled release time is also not proof of when the news was actually published or received. A real reconstruction needs the publisher’s timestamp and timezone, the data version, the underlying and quote timestamps, and any fill timestamps. A calendar shows when news is expected. It does not show what the news will say, or when your screen receives it. These differences matter even when several items share one date.

Once the final reference is fixed, later news cannot change this contract’s ordinary payoff computed from it. Earlier anticipation may already have moved its premium, and later news can still affect a surviving contract or a hedge. An official correction of a wrong reference is a separate process. Fixing the payoff also does not say when the settlement cash becomes available.

In this synthetic model, total remaining log variance is ordinary diffusion variance, which shrinks with time, plus the unchanged event variance. Annualising that total means dividing by the time left and taking the square root. The equivalent annual IV rises from about 34.47% to 59.61%, while the Call’s model mark falls from about 73.77 to 63.60 points. Less total uncertainty and a higher annualised number can coexist, because the time in the denominator has become smaller.

This builds on the Gaussian scheduled-earnings model studied by Tim Leung and Marco Santoli. The NIFTY inputs and the dividend-yield extension are our teaching assumptions. Federal Reserve research also finds higher option insurance costs around selected US macro releases, which supports the narrower idea that scheduled news can be priced in advance. Neither source gives a NIFTY intraday coefficient or a universal IV path. Once an event resolves, the market can revise remaining uncertainty up or down, so the after-release examples next allow both.

Say NIFTY rises only 25 points to 25,025 and IV falls to 20%. The pair is marked at about 47.28 points. With exactly the same small move but IV at 60%, it is about 129.51 points. That is higher than the lower-IV case, yet still below the original 170.49. Annual IV rose from 40% to 60%, but the model’s scale for the time left fell from about 0.855% to 0.641% as time shortened. A higher annual IV does not guarantee that a bought pair gains when other inputs change too.

A bigger rise of 200 points, to 25,200, tells a different story. Even with IV at 20%, the pair mark is about 200.15 points, above its starting model value. The Call now carries substantial intrinsic value, and the Put’s tiny pre-expiry mark stays positive. So “IV fell, so every option buyer lost” leaves out the move in the underlying, the contract and the time horizon.

The changes shown for 65 units compare unrounded model values with the unrounded opening model value. They are not realised profits, and the opening model mark is not the fictional fill used in the next section. A real early exit needs actual entry and exit fills, the quantities filled and charges. Bid and ask, depth, stale quotes or partial fills can make fills differ from a model estimate. Rounded Call and Put cells can also differ by 0.01 from the rounded pair, since the calculations keep full precision.

At a final reference of 25,000, both legs pay nothing and the long pair loses the ₹11,082.50 entry payment. At 25,025, the Call pays ₹1,625.00 and the Put pays zero, leaving a pair loss of ₹9,457.50. The direction was up, but the distance was too small to recover both premiums. At either 24,800 or 25,200, exercise cash is ₹13,000.00 and the pair gains ₹1,917.50 before the excluded costs and financing. The chart’s zero line separates a cash loss from a profit after premium.

For a matching short pair with the same fills and exclusions, every terminal result has the opposite sign. Keeping the opening premium at 25,000 is one scenario. It is not evidence of dependable income or of a seller’s winning probability. The two sides also carry different risk. An uncovered short Call has unlimited loss as the final reference rises without limit. An uncovered short Put’s terminal loss is finite, because the reference cannot fall below zero. The short pair inherits the unlimited Call tail, and the chart’s finite window is not a loss cap.

In this example the bought pair’s largest terminal cash loss is its entry premium, before the excluded amounts. These cash-settled calculations do not estimate an early-exit fill, margin needs or physical stock-delivery funding. A realised result needs the recorded fills or the official final reference and the settlement ledger. The event story and the model give neither, and these alternatives cannot say which position wins more often.

The separate cases below use 65 assumed option units. A bought Call’s supplied Delta rises from +0.50 to +0.80, so the book’s exposure rises from +32.5 to +52.0 equivalent units. The target hedge moves from −32.5 to −52.0: sell another 19.5 equivalent units. Selling the same Call reverses the book exposure and calls for buying 19.5. These are assumed exposure snapshots, not observed NIFTY trades.

For the falling-underlying Put cases, a bought Put’s Delta becomes more negative, from −0.50 to −0.80, so its target hedge becomes more positive and calls for a purchase. The short Put has the opposite, positive Delta, and its target adjustment is a sale. Call and Put name the contract. Whether you bought or sold it decides the signed exposure.

Equivalent units express sensitivity. NIFTY is an index, so there is no single NIFTY share to buy or deliver. A practical hedge may use whole futures contracts, other options or a basket of stocks. Contract multipliers, lot rounding, gaps between futures and index prices, costs and incomplete fills can leave exposure behind. Time and IV change Delta too, so the target arithmetic does not mean continuous neutrality or a hedging profit.

Public open interest cannot show who holds the long or short side, what else they hold, their hedge policy or the trades already done. Every outstanding contract has a long and a short. The same displayed OI can therefore sit beside opposite dealer Gamma signs. Cboe’s explanation for the US market supports the hedge mechanism, not a finding about today’s NIFTY dealer book. Whether hedge flow moves a market also depends on its size and the liquidity available.

This separate comparison keeps the NIFTY strike at 25,000 and annual IV at 20%, and changes only the model’s time left, from four hours to one hour, at each spot level. At spot 25,000, Gamma per unit rises from about 0.00373376 to 0.00746772 Delta per index point. At spot 25,200, it falls from about 0.00064199 to 0.00000698. The narrower, taller peak shows why distance from the strike matters.

The curves are alternative model snapshots, not an intraday price path. Their 20% IV is a fixed assumption, separate from the changing-IV event cases earlier. A real release can move spot and IV together, and a portfolio can mix positive and negative Gamma. A high-Gamma snapshot does not show pending hedge orders and does not predict the size of the next move.

At expiry, use the final intrinsic payoff directly. Its slope changes abruptly at the strike, a kink, so there is no single ordinary slope there and no finite smooth Gamma. The usual formula for positive time cannot be stretched by setting time to zero. The fixed-IV limit also differs from a model in which unresolved event risk makes annualised IV itself rise as time shrinks.

The extra expiry-day extreme loss margin (ELM), checked on 8 October 2026, applies to expiring short index options. NCL/CMPT/64639, dated 21 October 2024 and effective 20 November 2024, covers shorts already open at the start of the day and those opened during it. The extra 2% is charged on underlying notional based on the last available underlying close, not on the option premium or the margin already blocked.

With the fictional quantity of 65 and an assumed underlying closing reference of 25,000, that notional is ₹16,25,000 and the extra ELM is ₹32,500. This is only the extra ELM part. Other margin and settlement needs remain, so the amount is neither a loss limit nor a full funding estimate. The assumed quantity is not a verified live lot size.

A calendar spread uses different expiries. Dated SEBI rules remove the calendar-spread margin offset that involves an expiring leg on its expiry day, and leave pairs of later expiries under their existing treatment. The 1 October 2024 circular covers index contracts. The 5 February 2026 circular extends the same treatment to single-stock derivatives, effective three months from its date. This does not mean every same-expiry vertical spread loses its margin benefit. A vertical’s bounded final payoff still does not guarantee an exit fill, fixed collateral or enough funds for stock delivery.

The contract you hold, current clearing requirements and your broker’s account limits decide what you can actually do. Fixing the final reference, last trading, automatic exercise and settlement payment are separate processes. Neither this example nor a generic webpage gives a universal retail square-off time or a promise about when money becomes withdrawable. Start lesson 32 on margin covers the mechanics in full.

A real, now-past event shows the clock conversion. The Fed released its FOMC statement on 28 October 2026 at 2:00 p.m. EDT (read on the Federal Reserve’s own site on 6 November 2026). On that date Eastern Time was EDT, UTC−04:00, and IST is UTC+05:30, so the release came at 23:30 IST on 28 October. The Fed’s calendar had also scheduled a press conference half an hour later, at 00:00 IST on 29 October. The statement held the target range at 3¾ to 4 percent, with two members preferring a quarter-point rise. We use it only for the clock arithmetic. Compare any such time with the rules of your own option. This example does not invent a matching NSE expiry date or fixing hour, and it says nothing about how NIFTY options priced the news.

Keep the original release and its information vintage, the version available when you decided. The BLS, for example, revises payroll estimates as new information comes in and keeps vintage tables. A revised value downloaded later is not the earlier information. Actual minus the previous reading measures a change. Actual minus a pre-release expectation measures something else. Without a recorded source for the expectation, the size of the surprise stays unknown.

Record the timezone, the source timestamp and your own receipt time separately. A data provider can publish or refresh later than the original publisher. FRED’s release calendar uses US Central Time and separates source releases from availability on FRED. A quote’s timestamp, a last trade’s timestamp and a model calculation’s timestamp are also different evidence. Bid, ask and available quantity describe a quote, and only an actual fill shows what was executed.

Record requested and filled quantities, every fill price and time, partial or rejected orders, and charges counted once. An order submission is not a fill, and a chart or model mark does not show execution at your size. If there is no fill evidence, write “unknown” and do not present a realised result. The fictional premiums and the 65-unit paper ledger in this lesson stay labelled as assumptions.

Reconcile an expiry outcome with the official final reference and the settlement statement, keeping premium, dealing costs, collateral and delivery funding apart. A what-if such as holding longer needs the same starting position and a stated horizon, price convention and costs. Label it a counterfactual, not an observed trade. Spot, IV, time passing, other news and execution can all matter. One later outcome does not prove a single cause or a winning frequency for buyers or sellers.

The next lesson uses position-sizing maths to connect a contract’s exposure with the capital you have. Carry forward the contract, the filled quantity, the funding duties and the reconciled outcome, because they define what is being sized. The aim here was to see which clocks and exposures interact. It is not a rule to buy, sell, wait a fixed time or always exit because an event and an expiry share a date.