SKHY SMH Inclusion Meets Quarterly OPEX at the Close

The verified event is large, but the 60 million-share figure does not reconcile cleanly

SK hynix American Depositary Shares enter a major semiconductor benchmark at the September 18 close. MarketVector lists September 18 as the implementation date for the Q3 2026 review of the MVIS US Listed Semiconductor 25 Index, with the new composition effective the following trading day. SMH seeks to replicate that index.

Korean financial reporting says the review adds SKHY at roughly a 5% weight and removes Skyworks Solutions. The same reporting estimates approximately $3.4 billion of passive allocation based on SMH assets near $68.7 billion.

That is the part of the catalyst that can be supported by opened sources.

The more aggressive claim that SMH alone requires roughly 25 million SKHY shares, plus another 33 million shares from a separate September rebalance for a total near 60 million, does not reconcile with the public numbers reviewed here. At an intraday SKHY price around $185 on September 18, a $3.4 billion allocation corresponds to roughly 18.5 million ADSs. Even using somewhat higher SMH assets would still leave the implied share count below 25 million at that price.

No opened primary or reputable secondary source independently confirmed the additional 33 million-share component. The better options question is therefore not whether exactly 60 million shares must trade. It is how a clearly large index-linked allocation interacts with quarterly expiration, the closing auction, and an unusually young U.S. ADR market.

The index event is concentrated into one closing mechanism

MarketVector describes the semiconductor benchmark as a modified market-capitalization-weighted index containing 25 large and liquid U.S.-listed semiconductor companies. Its published review calendar shows the September review using August 31 index data, an announcement date of September 11, and implementation on September 18.

That timetable matters because index-tracking funds generally need their portfolios aligned with the new benchmark composition when the revised index becomes effective. For SMH, the practical concentration point is the September 18 closing process rather than an undefined multi-day corporate catalyst.

VanEck reported SMH net assets near $70 billion in the days preceding the rebalance. A 5% target weight therefore represents several billion dollars of economic exposure. Korean reporting estimated approximately $3.43 billion using September 17 fund assets.

The event is unusually large relative to the history of SKHY as a U.S.-listed instrument. SK hynix only completed its Nasdaq ADS offering in July. The SEC prospectus shows that the company offered 177.9 million ADSs at $149 each, with every ADS representing one-tenth of a Korean common share.

Using the reported $3.43 billion allocation estimate and a stock price near $185 produces an illustrative requirement of about 18.5 million ADSs. That is more than 10% of the 177.9 million ADSs issued in the July offering. The exact amount that changes hands at the close can differ because index funds may already hold positions, assets change during the day, prices move, and implementation can involve multiple portfolio and liquidity mechanisms.

The magnitude is still sufficient to create a meaningful market-structure test without relying on the unverified 60 million-share headline.

Quarterly expiration makes the attribution problem harder

September 18 is not only an index implementation date. It is also a major quarterly options expiration.

That overlap creates the central research tension. A large closing move in SKHY could be associated with index-linked portfolio adjustment, options hedging, expiration-related position closure, ordinary semiconductor-sector movement, or some combination of those forces.

The options chain was already unusually active before the close.

OptiView's September 17 snapshot showed SKHY 30-day at-the-money implied volatility at 57.1%. The most active listed contracts included the September 18 190 call with more than 12,000 contracts of volume, the 200 call with more than 8,000, the 185 call with more than 8,000, and the 175 put with more than 7,000. The same dataset reported more than one million contracts of aggregate call and put open interest across expirations.

On September 18 itself, the near-expiration chain continued to trade heavily. Around 18:45 GMT, with SKHY near $185.24, the 187.50 call had traded more than 12,000 contracts while the 190 call had traded nearly 8,000. Open interest from the prior effective date was also concentrated around several nearby strikes.

Those observations do not reveal investor intent. They do show that the same closing window contains both a benchmark rebalance and a dense expiration structure.

Near-expiration implied volatility is a poor standalone measure here

The same-day chain also illustrates why annualized implied-volatility readings can become difficult to interpret in the final hours before expiration.

At the September 18 observation, implied volatility around the 187.50 and 190 strikes had fallen sharply from the prior day's one-day readings as remaining time collapsed. A small change in option premium can produce a large change in annualized implied volatility when only hours remain.

That makes a single end-of-day volatility number less useful than three other measurements.

The first is realized price movement during the final hour and closing auction. The second is trading volume relative to ordinary SKHY closing activity. The third is the behavior of the next surviving expiration after the September contracts disappear.

If the closing event produces large turnover but limited price displacement, the rebalance may primarily be a liquidity event. If the stock experiences an unusually large final-hour move, the next question is whether that movement persists after the mechanical index alignment has passed.

The post-expiration curve is therefore more informative than the expiring contract itself.

The ADR structure makes passive-flow arithmetic unusually important

SKHY is not simply another long-established U.S. semiconductor constituent.

The July prospectus shows that 177.9 million ADSs were created in the U.S. offering. Because the Nasdaq listing is recent, the market has had only a short period to develop stable patterns in liquidity, institutional ownership, derivatives positioning, and the relationship between the U.S. ADS and the Korean ordinary shares.

That makes share-count arithmetic especially important when evaluating flow claims.

A 5% SMH weight does not automatically translate into a fixed number of ADSs. The number depends on the ETF's assets, the implementation price, existing holdings, creations or redemptions, and any portfolio adjustments already completed before the closing auction.

The reported $3.43 billion estimate is therefore more defensible as an approximate economic exposure than a precise share instruction.

This distinction becomes even more important for the additional 33 million-share claim. Without an identified benchmark, official review document, or fund-level calculation, combining that number with the SMH estimate risks double-counting or mixing separate methodologies.

For options research, uncertain flow arithmetic should not be converted into a directional assumption.

SMH itself provides a second observable market

The event can also be studied from the ETF side.

Before the review, VanEck's published holdings showed a portfolio dominated by Nvidia, Taiwan Semiconductor Manufacturing, Broadcom, Micron, AMD, ASML, Intel, and other large semiconductor names. Adding SKHY near a 5% target requires weight reductions elsewhere in a fully invested benchmark portfolio.

That means the event is not only about SKHY demand. It is also a redistribution across the semiconductor basket.

Nvidia is particularly relevant because Korean reporting noted that the index methodology can reduce very large constituent weights at scheduled reviews. Other constituents can also experience mechanical weight changes even when their company-specific information has not changed.

A pure SKHY explanation for every semiconductor move around the close would therefore be too narrow. The better cross-market test compares SKHY with SMH and major constituents during the final hour, then again after the revised index is active.

If SKHY moves independently while SMH remains comparatively stable, the inclusion mechanism becomes a stronger explanation. If the entire semiconductor basket moves together, broader sector conditions become more plausible.

The cleanest test begins after the close

The most useful evidence arrives in two stages.

The first stage is the September 18 closing auction. Relevant observations include SKHY volume, the size and direction of the closing print relative to the pre-close market, and whether SMH shows an unusual divergence from its semiconductor components.

The second stage begins on September 21, when MarketVector says the revised index composition is effective.

Once the scheduled implementation has passed and the expiring September options are gone, the market loses two important mechanical forces at the same time. That creates a natural post-event comparison.

If SKHY volatility and turnover fall quickly while the price remains near the closing-auction level, the event would look primarily like a liquidity and index-alignment episode. If substantial volatility persists in later expirations, another source of uncertainty is likely contributing, such as memory-cycle expectations, company-specific news, ADR pricing dynamics, or broader semiconductor volatility.

The distinction matters because the headline describes a one-session mechanical event, while SKHY options price distributions across many future sessions.

The event is stronger as a volatility experiment than as a flow forecast

The SMH inclusion is real, the implementation date is identifiable, and the estimated economic allocation is large. Those facts make September 18 an unusually useful options-market experiment.

The unsupported part is the precise 60 million-share total.

Based on opened sources, the most defensible arithmetic begins with a roughly 5% target weight and approximately $3.4 billion of estimated SMH-linked exposure. At a stock price near $185, that implies an illustrative share count closer to 18 million to 20 million ADSs, subject to fund assets, price, existing positions, and implementation mechanics.

That is already a substantial event relative to the size and short trading history of the U.S. ADS program.

The research value comes from what happens when that index flow meets quarterly expiration at the same close. Comparing the closing auction with the next surviving options curve can separate a temporary market-structure shock from a more persistent change in SKHY volatility. That is a cleaner test than treating an unverified share-count estimate as the conclusion.