Original | Odaily Planet Daily (@OdailyChina)
Author | Azuma (@azuma_eth)
On July 29, Beijing time, SK Hynix announced its second-quarter 2026 financial results.
Financial results show that SK Hynix recorded revenue of KRW 7.932 trillion in the second quarter, representing a 257% year-over-year increase and a 51% quarter-over-quarter increase; operating profit reached KRW 6.054 trillion, up 557% year-over-year and 61% quarter-over-quarter, with the operating profit margin rising further to 76%, a historical high. Including a one-time investment gain of KRW 6.2166 trillion from the sale of a portion of its stake in Kioxia, the company’s net profit reached KRW 9.392 trillion.
In any industry, this would be considered a stunning performance that shocks the market.
However, the capital market’s initial reaction was the opposite. Due to revenue (actual: 79.32 trillion KRW, market expectation: 84 trillion KRW) and operating profit (actual: 60.54 trillion KRW, market expectation: 64 trillion KRW) both slightly missing prior estimates, combined with SK Hynix’s stock having already declined over 40% in the preceding month, pessimism prevailed—causing the company’s U.S. ADR to drop approximately 9% in after-hours trading following the earnings release (after having fallen nearly 9% in regular trading the previous day). However, as investors gradually absorbed the earnings details, the stock quickly recovered all of its losses and even turned positive.
Meanwhile, after the opening of the Korean stock market this morning, SK Hynix's stock price initially opened higher and rose by up to 4%, but then gradually weakened, falling more than 9% by 10:00.
A record-breaking earnings report—why did it first face frantic selling, then quickly recover its losses, only to plunge again? The answer may lie in the fact that the market cares far more than just how much SK Hynix earned in the second quarter; it’s about how to reprice its future growth potential—and bulls and bears clearly haven’t reached a consensus on this point.
Why was the most profitable quarter still below expectations?
Numerically, SK Hyness is still nearly at its most profitable stage.
In the second quarter, the company achieved a gross profit margin of 83% and an operating profit margin of 76%, meaning that approximately 76 won of every 100 won in sales was converted into operating profit—a profitability level surpassing that of the vast majority of global semiconductor manufacturers. Meanwhile, the company’s cash and short-term financial assets continued to grow rapidly to 87.96 trillion won, further expanding its net cash position and providing ample resources for future capacity expansion.
However, the issue is that the market had already set expectations even higher. Previously, the consensus forecast for SK Hynix’s second-quarter revenue was approximately 84 trillion KRW, with an operating profit of around 64 trillion KRW; the actual results ultimately fell about 5% and 6% below these estimates, respectively.
For a typical company, such a deviation is not significant, but for SK Hynix, which has been labeled as the "biggest beneficiary of AI" and whose valuation is based on high growth expectations, any data below expectations will be amplified by the market.
Upon closer examination of the financial report, it becomes clear that this "missed expectation" was not due to a deterioration in market demand, but rather primarily resulted from changes in the profit structure.
A counterintuitive point is that the continuous increase in the proportion of HBM products has actually weakened profit elasticity. Over the past several quarters, a key driver behind the rapid profit expansion in the memory industry has been the sustained rise in spot prices for traditional DRAM and NAND. However, since SK Hynix’s revenue mix from HBM is significantly higher than that of its peers, and HBM is predominantly priced under long-term supply agreements (LTAs), it cannot fully benefit from the rapid spot price increases in the same way that standard DRAM can.
In addition, SK Hynix disclosed that the average selling price of standard DRAM in the second quarter increased by approximately 30% quarter-over-quarter, remaining positive but significantly lower than in the first quarter; the average selling price of NAND increased by 50%–55% quarter-over-quarter, also showing a noticeable slowdown compared to the first quarter.
In other words, AI products are selling more, but the price increases for traditional products have slowed; long-term contracts lock in future revenue but limit short-term profit flexibility. This is why record profits still fell short of the numbers the market had already “imagined.”
Is the storage supercycle still ongoing? How do earnings reports answer this?
If the operating data answers how much SK Hynix earned in the second quarter, then the information provided by management in the earnings report and subsequent conference call addresses another question that the market cares more about—whether the AI storage supercycle has begun to cool down.
At present, SK Hyness's response remains optimistic.
First, in terms of demand outlook, the company did not signal any notable caution, as markets had feared. SK Hynix expects global DRAM demand to grow by a mid-20% year-over-year in 2026, and NAND demand to increase by a high-teens percentage year-over-year. Management also stated on the post-earnings call that they have not observed any signs of slowing AI investment and anticipate continued robust growth in AI infrastructure spending beyond 2027.
Second, another key point of interest is the further advancement of long-term supply agreements (LTAs). SK Hynix disclosed that it has finalized LTA negotiations with approximately ten customers and continues to engage in ongoing discussions with other major industry clients. The next-generation LTAs will incorporate pricing mechanisms designed to address price volatility, along with corresponding financial safeguards to ensure contract performance, thereby enhancing the stability and predictability of future demand.
This shift holds significant meaning for the storage industry. In the past, products such as DRAM and NAND relied heavily on spot market pricing, and their volatile prices kept the entire industry labeled as cyclical stocks. However, as the share of HBM products continues to grow in the AI era, an increasing number of major cloud providers are securing multi-year supply commitments in advance. The supply-demand relationship is gradually evolving from short-term market fluctuations toward more long-term, stable partnerships. Although long-term agreements may temporarily compress profit flexibility during periods of rapid spot price increases—such as this quarter—they bring significantly higher revenue certainty for the coming years.
Additionally, the development timeline for SK Hynix’s next-generation products proceeded without any surprises. The financial report shows that SK Hynix began shipping HBM4 products in the second quarter and plans to ramp up production fully in the second half of the year. Sampling of the next-generation HBM4E has already been completed with key customers in the first half of the year. Furthermore, the SOCAMM2 product based on the 1cnm process has officially begun shipping.
This means SK Hynix continues to maintain a leading pace in its product roadmap for the next-generation AI GPU platform. Given that HBM4 will be a critical memory component for next-generation AI platforms such as NVIDIA’s Rubin, its successful mass production also indicates that the company remains firmly positioned as a market leader in high-end AI memory.
Finally, regarding capital expenditures (CapEx)—the metric that best reflects management’s true judgment—SK Hynix has maintained its expectation of CapEx at the high end of the ₩40 trillion range for 2026, plans to accelerate the mass production of the M15X facility, speed up the construction of Phase 1 of the Yongin Fab, and continue advancing mid- to long-term projects such as P&T7, M17, and the new Korean semiconductor cluster.
For a manufacturer that has undergone multiple storage cycles, such an aggressive expansion plan is itself a statement—management still believes that AI storage demand over the coming years will be sufficient to absorb this new capacity.
The focal point of the long-short battle
Today, SK Hyness has become the central asset in the battle between bulls and bears in the AI storage cycle.
For bulls, record profits, sustained growth in HBM demand, and the AI infrastructure investment cycle continue to support the company’s long-term growth thesis; for bears, disappointing earnings, valuation pressures, and concerns over the sustainability of AI capital expenditures are intensifying short-term adjustment pressures. Bulls bet that AI infrastructure expansion will persist, while bears worry that the market has already priced in future growth.
To wear the crown, you must bear its weight. SK Hynix, having enjoyed the valuation of an industry leader, must also shoulder the pressures that come with it—when the market already believes your story, strong performance is no longer enough; only consistent outperformance of even higher expectations can drive further valuation growth.