Two Supply Shocks for August: The SpaceX Lockup Cliff and Korea's Leverage Trap

Two market-structure events in the first week of August share a common thread: both were designed, and both are about to test whether the design holds.

On August 6, two trading days after SpaceX reports its first quarterly earnings as a public company, roughly 911.5 million insider shares become eligible for sale — more than the company’s entire public float at IPO^[1]. It is the first step of a staggered 16-date unlock calendar that could expand tradable supply by roughly 900 percent before October^[2].

Halfway around the world, South Korea is learning the same lesson in reverse. Single-stock leveraged ETFs, introduced on May 27 to enthusiastic retail demand, amplified a semiconductor sell-off into one of the sharpest two-day declines a major national market has recorded in years. The Kospi fell more than 16 percent in two sessions^[3]. Korea’s finance minister apologized^[4]. The regulator capped retail allocations at 20 percent and is considering restricting the products to professional investors^[4].

Neither event is a coincidence. Both were engineered — SpaceX’s lockup staircase deliberately, Korea’s leveraged ETF framework inadvertently — and both are now running ahead of the market’s ability to absorb them.

The SpaceX Lockup Staircase

SpaceX priced 555.6 million shares at $135 on June 11, 2026, raising roughly $75 billion. With the greenshoe exercised in full, gross proceeds reached $85.7 billion — the largest IPO in history, nearly tripling Saudi Aramco’s 2019 record of $29.4 billion^[2]. The stock opened near $150, closed at $161 on day one, and touched $225.64 intraday on June 16, a 67 percent gain in four trading days^[2].

Then it went the other direction. On July 15, SPCX closed below its $135 offer price for the first time. On July 23, it printed an all-time low of $110.85. It closed the week at $115.07 — down 49 percent from peak to trough in 26 trading days^[2].

The drawdown was not mysterious. Three things compounded.

The prospectus broke the narrative. The S-1, filed May 20, disclosed consolidated 2025 revenue of $18.7 billion but a net loss of $4.94 billion^[2]. The AI segment (xAI, X, Grok, Colossus) lost $6.36 billion on $3.2 billion of revenue. Capital expenditure totaled $20.7 billion, of which $12.7 billion went to xAI. In the first quarter of 2026 alone, the AI segment consumed $7.72 billion in capex^[2]. Long-term debt stood at $29.1 billion as of March 2026^[2]. Investors who spent years waiting to buy a rocket company found themselves reading a prospectus for a data-center buildout with a rocket company attached.

Four percent of a company is not a market. The 555.6 million shares sold represented about 4 percent of SpaceX’s roughly 13 billion shares outstanding^[2]. The underwriting syndicate steered an unusual 20 percent of the deal to retail — roughly $15 billion — and order books reportedly reached $150 billion against a $75 billion raise^[2]. Enormous demand meeting deliberately starved supply drove the stock to $225.64. When the demand thinned, there was nothing underneath it.

The unlock calendar starts on August 6. SpaceX did not write a standard 180-day lockup. It designed a staircase — a 16-date release schedule engineered to distribute selling pressure over time. The first step is the largest: 20 percent of restricted shares, roughly 911.5 million, become sellable two trading days after the Q2 earnings release^[1]. At recent prices near $124, that block is worth roughly $123 billion — more than the estimated $86 billion in shares currently in public float^[1].

Subsequent tranches follow every two to three weeks: 7 percent on August 21, another 7 percent on September 10, and so on through October 25. A larger release of approximately 28 percent is tied to the third-quarter earnings report. All remaining shares free up on December 8, the 180-day mark^[1]. By then, roughly 40 percent of SpaceX’s total shares will be freely tradable.

A performance condition that would have released an additional 10 percent will not be met. That tranche required SPCX to close at or above $175.50 (30 percent above the IPO price) on at least 5 of 10 consecutive trading days before the Q2 earnings date. The stock trades around $124. The condition is effectively dead^[1].

Elon Musk’s approximately 6.4 billion shares carry a 366-day lockup running to approximately June 2027, with no performance-based early release^[1]. His stake — about 42 percent of the equity and 93.6 percent of the Class B super-voting shares — is the largest single block of locked stock in any public company^[2]. Its removal from the selling pool is the one structural feature protecting other shareholders from a CEO-sale signal.

Wall Street’s Unresolved Argument

When the underwriters’ research quiet period expired, 18 banks published price targets. The spread is remarkable: Raymond James at $800, Stifel at $190, a $225 median around which JPMorgan, Deutsche Bank, and others clustered^[2]. Later initiations were more cautious. MoffettNathanson started at Neutral with a $131 target, arguing SpaceX lacks a credible financial model to support a valuation near $2 trillion^[2]. HSBC initiated at Hold with $115 — which turned out to be the accurate call. SPCX closed at $115.07 on July 24, within a rounding error of HSBC’s target published the day before^[2].

The bear case is a sum-of-the-parts exercise: value Starlink and the launch business on comparable multiples for infrastructure and connectivity assets, assign the AI segment something defensible for a unit losing $6 billion a year, and the total comes out near $900 billion, or roughly $69 a share^[2]. Even after the drawdown, the market is paying a substantial premium over that — around 80 times trailing sales, with no earnings to put in a P/E field^[2].

None of this will be settled on August 4, when SpaceX posts Q2 results after the close with Musk on the 4:30 p.m. ET call^[2]. But three narrower questions will be answered: whether Starlink subscribers crossed 11.5 million, whether ARPU decline from $99 to $66 continues at that slope, and what management says about AI capex trajectory after a $7.7 billion quarterly burn annualizes to over $30 billion^[2].

Two days later, the selling window opens. The market is front-running its own supply.

Korea’s Leverage Trap

On the same week SpaceX was testing its lows, South Korea provided a real-time demonstration of how structured products can amplify a selloff when the underlying turns.

Single-stock leveraged ETFs were introduced in Korea on May 27, 2026^[4]. Retail investors piled in. Net purchases by Korean retail investors reached 14 trillion won ($9.7 billion), compared with roughly 2 trillion won from foreign investors^[4]. The products were popular because the Kospi had been one of the world’s hottest markets, driven by the AI semiconductor rally. Samsung Electronics and SK Hynix dominate the index, and leveraged products tracking both had surged.

Then the trigger hit: a report that Chinese manufacturers had begun mass-producing deep ultraviolet (DUV) lithography equipment — the machines used to print circuits onto silicon wafers^[3]. That threatens the assumption underpinning Korean memory profitability, which is that new supply cannot arrive quickly because the equipment is too difficult to build. Samsung Electronics fell 13.4 percent and SK Hynix 14.7 percent in the initial session^[3].

The leveraged ETFs made it worse. A 2x single-stock ETF delivers twice the daily move of one company. When the underlying falls, the fund must sell exposure to maintain its leverage ratio — selling into a falling market, which pushes prices down further, which forces more selling the next day^[3]. The KODEX SK Hynix Single Stock Leverage ETF fell more than 80 percent from its June 23 peak^[4]. The equivalent Samsung-tracking product fell nearly 75 percent from its June 3 peak^[4]. The Kospi plunged more than 16 percent in two sessions^[3].

On July 30, the government capped retail allocation in leveraged single-stock ETFs at 20 percent^[3]. The Financial Services Commission ordered asset managers to stagger rebalancing throughout the trading day rather than concentrating it near the close^[5]. Finance Minister Koo Yun-cheol apologized in a parliamentary session, saying the products were introduced “without careful consideration”^[4]. The FSC’s Lee Eog-weon told lawmakers the regulator could lower the leverage multiple from 2x if legislation is prepared, and is considering restricting the products to professional investors only^[4].

The intervention addresses the mechanism, not the cause. Leveraged funds made the decline faster and deeper, but they did not create the question about Chinese lithography capability that started it^[3]. The Nikkei 225 rose 0.93 percent on the same day the Kospi fell further^[3] — evidence the market is treating this as a Korean memory problem rather than a general Asian technology problem.

The Broader Pipeline

The US IPO market has been active in 2026, with roughly $251 billion raised across 86 deals through late July — already ahead of all of 2025^[6]. New S-1 filings continue to arrive at the SEC, including registrations from Global Interactive Technologies (filed July 29), Sinda Ltd. (filed July 27), and Lyntris Inc. (filed July 23)^[6]. The Renaissance Capital calendar lists upcoming deals including River City Bank (RCBC)^[7].

On the buyback side, BASF announced it will begin a new €1 billion share repurchase program in August 2026, running through April 2027, as part of a larger €4 billion buyback announced in September 2024^[8]. Shell declared a Q2 2026 interim dividend and launched a multi-billion share buyback^[8]. Corporate liquidity support continues even as new issuance tests the market’s absorptive capacity.

Meanwhile, the CFTC proposed new rules on affiliations among regulated entities in U.S. derivatives markets, covering amendments to Parts 37, 38, and 39^[9] — a signal that regulators are watching the growing links among trading venues, clearing houses, and intermediaries that the Korean episode just made vivid.

What to Watch Next

The pattern is the same in both stories. A product was designed — a lockup staircase in one case, a leveraged ETF framework in the other — to manage a known risk. In Korea, the design failed under stress within seven weeks of launch. At SpaceX, the design has not been tested yet; August 6 is the first real trial.

What makes both worth watching is that the mechanism is the message. SpaceX’s 4 percent float produced a price that was an auction result, not a valuation. Korea’s 2x ETFs produced a decline that was an amplification, not just a correction. In both cases, the structure — not the fundamentals — is what the market is about to discover the price of.