Standard monthly KOSPI 200 options settle against a single number: the index level produced by the ten-minute closing auction on the second Thursday of the month. That is why Korean expiry days can be quiet for six hours and busy for ten minutes — hedge adjustments and any arbitrage positions timed to that settlement concentrate into the same auction. But it is important to be precise about what does and does not unwind there. Cash-and-carry arbitrage paired with quarterly KOSPI 200 futures ordinarily terminates or rolls at futures expiry, not at every monthly options expiry, and options hedges come in varying structures that do not universally require an exact simultaneous basket unwind. Checking the KOSPI 200 futures basis and the KRX arbitrage program-trade balance the day before gives you a rough read on the potential size and direction of late flows — a useful prior, not a dependable forecast.
Interest in this mechanism tends to spike when the market is already unstable. In early August 2026, the KOSPI closed at 6,296.38 on August 6, a session in which foreign investors net sold about 3.29 trillion won across KOSPI and KOSDAQ combined — with a monthly options expiry landing on August 13, 2026. The rest of this guide sets the news aside and explains the recurring machinery, because the same dynamics return every month.
What Actually Settles on Expiry Day
The KOSPI 200 is the Korea Exchange’s benchmark of 200 large listed companies, and it is the underlying for Korea’s flagship index derivatives. Two features of the contract design explain much of expiry-day behavior:
- Monthly standard options, quarterly futures. The standard monthly KOSPI 200 option series has historically expired on the second Thursday of the month; the exchange has also listed other option series, such as weekly contracts, with their own expiry schedules, so the second-Thursday rule applies to the standard monthly series rather than to every option in existence. KOSPI 200 futures run on a quarterly cycle (March, June, September, December), so most second-Thursday expiries are options-only events.
- Cash settlement against the close. There is no delivery of shares. Each contract pays out in cash against a final settlement index calculated from the closing prices of the 200 constituent stocks — and those closing prices come out of the closing auction. The contract multiplier is roughly 250,000 won per index point, so small index moves translate into large cash flows across open interest.
The practical consequence: a trader whose derivatives leg dies at that settlement, and who wants to keep the hedge exact to the end, has an incentive to deal with the stock leg in the same auction. But many positions are not on that clock at all — futures-linked arbitrage typically lives and dies on the quarterly cycle, and some options hedges are adjusted gradually or carried past expiry in other forms. On a monthly options-only expiry, the auction flow is mostly hedge adjustment rather than wholesale arbitrage liquidation.
Program Trading and Index Arbitrage, Defined
Program maemae (program trading) is the Korean market term for orders that buy or sell a basket of many stocks simultaneously, typically the KOSPI 200 constituents in index weights. KRX splits its daily program-trading statistics into two buckets, and the distinction matters:
- Chaeik georae (arbitrage trading): basket trades explicitly paired against index futures or options. Classic cash-and-carry: if futures trade rich to fair value, an arbitrageur buys the 200-stock basket and sells futures, locking in the spread.
- Bichaeik georae (non-arbitrage trading): basket trades without a paired derivatives leg — index fund creations, portfolio transitions, quant rebalances.
The arbitrage bucket is the one that tends to concentrate around derivatives settlement dates. A cash-and-carry position earns its locked-in spread when its paired derivatives leg settles — for futures-linked positions that means the quarterly futures expiry, unless the position is rolled forward instead. The non-arbitrage bucket spreads across the calendar; the arbitrage bucket clusters around settlement dates, with the heaviest potential unwinds at quarterly expiries. On monthly options-only expiries, arbitrage-related auction flow exists but is typically smaller and less mechanical.
The Basis: A Signal Worth Reading — With Limits
The basis is the futures price minus the spot index level. Its fair value comes from cost of carry — roughly, the interest rate to fund the stock basket minus the dividends the basket pays before expiry. Deviations from that fair value are what feed arbitrage positioning, which means the basis in the days before an expiry offers clues — imperfect ones — about pressure at that expiry:
- Persistently rich basis (futures at a premium): arbitrageurs have likely been buying stock baskets and selling futures. To the extent those long-stock positions are unwound at settlement rather than rolled, they imply potential sell pressure into the relevant expiry auction.
- Persistently weak basis (futures at a discount, or backwardation): reverse cash-and-carry builds short-stock positions, which imply potential buy pressure at settlement if they are closed rather than rolled.
- A basis that flips near fair value: arbitrage balances tend to be small, and the expiry close is more likely to be orderly.
You do not have to guess at the accumulated position. KRX publishes program-trading and arbitrage balance data on its data portal, and Naver Finance surfaces daily program-trade net figures in Korean-language tables that are readable with minimal vocabulary. Two caveats: the common mistake is watching only the day-of basis, since by expiry morning the informative positioning has already been built — read the trend over the preceding week or two; and even then, the published balances are an incomplete picture of positioning, so treat them as one input, not a prediction.
Why the Final Ten Minutes Can Get Wild
Korean cash equities trade continuously from 09:00 to 15:30 KST, but the last ten minutes, 15:20 to 15:30, are a single-price call auction — dongsihoga — in which orders accumulate and every stock prints one closing price. Because the options settlement index is computed from exactly those auction prices, unwinds that are timed to settlement cannot be spread out; they must be submitted into that one auction to keep the hedge exact. Three consequences follow:
- Order imbalance can be structural, not accidental. If the portion of the arbitrage balance that is unwound at settlement is one-sided, the auction receives a one-sided basket flow by design.
- The last print can detach from the day’s trend. A market drifting up all day can still close down hard if net settlement-linked unwinding is on the sell side — and vice versa.
- Tail events are possible. The most famous case is the November 2010 expiry, when a massive foreign sell program in the final minutes knocked the KOSPI 200 down abruptly and prompted tighter disclosure and monitoring rules for expiry-day program orders. Korea also operates a sidecar mechanism that briefly restrains program orders after sharp futures moves — a speed bump, not a prevention.
Reading the Flow Data Beforehand: A Worked Example — and Its Limits
Foreign investors dominate KOSPI 200 futures trading, so their cash-market flows around an expiry are worth reading. But be clear about what this data is: the series below is aggregate foreign net flow in the KOSPI and KOSDAQ cash markets combined. It is not a KOSPI 200 program-arbitrage balance, and it cannot establish the size, direction, or even the existence of an expiry-linked position. Nor can it distinguish program baskets from the aggregation of many stock-specific trades — at most it raises a hypothesis that basket-scale money might be moving, one that has to be tested against the KRX program-trading breakdown. The figures are in units of 100 million won (eok won), from August 2026:
| Session (2026) | Foreign net flow (100M won) | In trillions of won |
|---|---|---|
| Jul 27 | -28,811 | -2.88 |
| Jul 28 | -45,009 | -4.50 |
| Jul 29 | -12,502 | -1.25 |
| Jul 30 | +13,280 | +1.33 |
| Jul 31 | +72,410 | +7.24 |
| Aug 3 | -28,220 | -2.82 |
| Aug 4 | -3,716 | -0.37 |
| Aug 5 | +14,464 | +1.45 |
| Aug 6 | -32,893 | -3.29 |
Step by step, using the 2026-08-06 figure of -32,893:
- Convert the unit. One eok is 100 million won, so divide by 10,000 to get trillions: -32,893 eok won is about -3.29 trillion won of net foreign selling in one session.
- Judge the magnitude against the series. Within nine sessions, flows ranged from -4.50 trillion to +7.24 trillion won. Single-day swings of several trillion won in either direction are large enough to be consistent with basket-scale activity — but because this is an aggregate figure, it could equally reflect the sum of many unrelated stock-specific trades. Treat it as a hypothesis to be confirmed in step 4, not a conclusion.
- Net the window. Summing all nine sessions gives roughly -5.1 trillion won. Despite one enormous buy day, the direction of travel into the mid-August expiry was net selling. Violent alternation around a negative net is a very different setup from steady accumulation.
- Cross-check the composition — this step is essential, not optional. Because these figures mix KOSPI 200 and non-index names, program and non-program orders, and arbitrage and non-arbitrage motives, they say nothing by themselves about program trading or expiry positioning. That is what the KRX program-trading breakdown is for: only if a meaningful share of the activity shows up in the chaeik (arbitrage) column is there a basis for expecting settlement-linked auction flow — and even then, the direction and timing remain uncertain.
Monthly Expiry vs Quadruple Witching
Four times a year — the second Thursday of March, June, September, and December — options expiry coincides with the expiry of KOSPI 200 futures, single-stock futures, and single-stock options. Korean media calls this ne manyeo-ui nal, literally the day of the four witches: quadruple witching. The differences from an ordinary monthly expiry are practical, not cosmetic:
- Larger balances. These are the dates when futures-linked cash-and-carry positions actually mature. Quarterly futures accumulate three months of arbitrage and hedging positions, versus one month of options positioning for options-only expiries, so auction flows are typically heavier.
- Roll activity in the preceding week. Futures holders who want to stay positioned must roll to the next quarterly contract, which itself moves the basis and can generate program flow days before the expiry itself.
- Single-name distortion. Single-stock derivatives settlement can push individual large-cap closes around independently of the index effect.
A Pre-Expiry Checklist
- Mark every second Thursday, and flag March, June, September, and December as quadruple-witching dates — the quarterly dates are where futures-arbitrage liquidation concentrates.
- Over the prior one to two weeks, track whether the KOSPI 200 futures basis has been persistently rich or weak — that suggests, without guaranteeing, the direction of potential auction pressure.
- Check the KRX arbitrage program-trade balance to size the accumulated position, and recent foreign flow data (as in the worked example) for unusually large flows — remembering that the aggregate flow data alone cannot confirm basket-scale or expiry-linked positioning; only the KRX program-trading breakdown can.
- Note the domestic liquidity backdrop for context: as of 2026-08-05, investor deposits at brokerages (tujaja yetakgeum, the cash retail investors keep ready to trade) stood at about 103.2 trillion won, and margin lending balances (sinyong yungja) at about 28.4 trillion won — both published by KOFIA. Whether such levels represent a large or small buffer requires comparison against their own history, which is beyond a single snapshot.
- If you do not want auction risk, avoid market-on-close orders on expiry day; the single print can gap away from the 15:19 price.
- Be cautious about extrapolating the expiry close into the next morning. Settlement flow is mechanical rather than informational, and it can reverse at the following open.
FAQ
What time does expiry-day volatility usually hit?
Most often in the closing auction, 15:20 to 15:30 KST, because that is where the settlement index is determined. The continuous session can be entirely unremarkable beforehand, which lulls newcomers into thinking the event passed quietly.
Is expiry day systematically bullish or bearish for the KOSPI?
No. The direction of any closing pressure depends on the sign and size of the settlement-linked portion of the accumulated arbitrage position, which flips from cycle to cycle with the basis. Treat expiry as a potential volatility event whose direction the basis and balance data can only partially suggest — not a calendar edge in either direction.
Where can I see program-trading and basis data without a Bloomberg terminal?
The KRX data portal publishes program-trading statistics, arbitrage balances, and derivatives data free of charge; Naver Finance carries daily program-trade and investor-flow tables; KOFIA publishes investor deposits and margin balances. All are Korean-language sites, but the tables follow fixed layouts that are easy to learn.
Does the currency matter on expiry days?
Indirectly. Foreign arbitrage and hedging flows are won-denominated, so a volatile won — it closed at 1,421.16 per dollar on August 6, 2026 — changes the attractiveness of carrying hedged Korean positions and can amplify foreign flow swings around expiries.
Sources
- KRX data portal — derivatives, program trading, and arbitrage balance statistics (data.krx.co.kr)
- Naver Finance — daily program-trade and investor flow tables (finance.naver.com)
- KOFIA statistics — investor deposits and margin lending balances (freesis.kofia.or.kr)
- Bank of Korea ECOS — rates and FX reference data (ecos.bok.or.kr)
Disclaimer: This is an information and analysis publication, not investment advice. See our Methodology for data sources, standards, and our corrections policy.