SK Hynix Rings the Bell in New York: Nasdaq Crowds Overflowing

Author: Zhou Ailin, Tencent Technology

Editor: Liu Peng

July in New York is extraordinary, with crowds surging in Times Square. In front of the Nasdaq MarketSite, the blue logo of South Korea's SK Hynix intertwines with the green light bands of Nasdaq on a giant LED screen, shining brightly in the morning sun of Manhattan.

On the morning of July 10, local time in the U.S., SK Hynix, known as the "King of HBM," officially listed on Nasdaq. SK Hynix is one of the largest memory semiconductor companies in the world, focusing on the design, manufacture, and sale of advanced storage semiconductors, including DRAM (Dynamic Random Access Memory), NAND flash memory, and high-bandwidth memory (HBM) for artificial intelligence (AI). HBM is currently the most scarce hardware component in the AI ecosystem.

The U.S. stock offering was oversubscribed by seven times, raising $26.5 billion, making it the second-largest stock issuance globally, following SpaceX's listing last month. The company's Q2 financial report showed a net profit of 40.35 trillion Korean won (approximately $27.3 billion) for the quarter, a year-on-year increase of 398% and a quarter-on-quarter increase of 165%. For comparison, Nvidia's net profit for the first quarter was only about $19 billion. A company selling the world's most expensive GPUs had a quarterly net profit that was less than that of SK Hynix.

An SK Hynix employee at the event told Tencent Finance, "I took a special vacation to fly to New York for this listing ceremony." He joked that many young people in South Korea have invested in leveraged ETFs related to storage, and there is a sense of FOMO (fear of missing out) among them.

However, after continuous price increases in the storage industry and a significant rise in the company's stock price, the storage sector has recently entered a consolidation phase. Is SK Hynix's listing on the U.S. stock market a signal of reaching a peak, or is it the beginning of expansion? What does this mean for Changxin Technology, which is about to list on the A-shares?

Asif Suria, founder of the San Francisco asset management firm Inside Arbitrage, told Tencent Finance, "SK Hynix's listing on Nasdaq is clearly aimed at benefiting from the current favorable industry cycle. From the demand side, storage chips are still in short supply, and we have already seen the impact of this supply-demand tension, such as Apple raising prices on some products. However, stock prices usually lead fundamentals, so the current pullback in the storage sector is healthy. I believe that at least until the end of this year, the core logic of this bull market has not changed. Compared to the volatile and leveraged Korean stocks, I prefer to gain exposure to storage through holding Micron Technology."

Nasdaq Marketsite SK Hynix Listing Event at Times Square Photo: Zhou Ailin

01 HBM Dominance in the AI Era

Under this AI super cycle, SK Hynix's king status is highlighted.

The explosive growth of AI large models and high-performance computing (HPC) has led to a surge in demand for high-bandwidth, large-capacity memory. According to Gartner data, from 2025 to 2027, the overall memory semiconductor market is expected to expand at a compound annual growth rate (CAGR) of 86.0%, reaching nearly $748 billion by 2027. HBM is expected to grow particularly rapidly.

At the same time, the average selling prices of the company's DRAM and NAND products are also expected to rise significantly. The industry is shifting from a consumer-driven demand to one dominated by enterprise and data center needs, providing continuous growth opportunities for high-end memory products (such as HBM and enterprise SSDs).

Let's take a look at SK Hynix's industry position and competitive landscape:

Industry Position:

• HBM Field: Global Leader. According to IDC data, in Q1 2026, SK Hynix's revenue share in the HBM market reached 56.4%, ranking first in the world.
• DRAM Field: Global Second. The market share in Q1 2026 was 29.1%.
• NAND Field: Global Second. The market share in Q1 2026 was 18.5%.

Competitive Landscape:

• DRAM Market: Highly concentrated, mainly dominated by Samsung Electronics, SK Hynix, and Micron Technology.
• NAND Market: Major competitors include Samsung Electronics, Kioxia, SK Hynix, and Western Digital.
• Barriers: The industry has extremely high technical and capital barriers, making it difficult for new entrants to pose a threat in the short term.

SK Hynix's ability to maintain high profits is due to the high technical barriers of HBM. SK Hynix is a pioneer in HBM technology, being the first to mass-produce HBM3E in 2024 and successfully developing the next generation HBM4 in 2025. Its mastery of TSV (Through-Silicon Via) packaging technology and MR-MUF (Mass Reflow Molded Underfill) process is its core moat. HBM is crucial because it alleviates the "memory wall" problem that arises when the computational power of high-performance GPUs far exceeds the speed of traditional memory.

In terms of process technology, SK Hynix also maintains an absolute advantage. For example, in the DRAM field, the company has 1c generation (6th generation 10nm level) process technology; in the NAND field, it is transitioning from 238 layers to 321 layers of high-density flash products, continuously improving storage density and energy efficiency. As AI server architectures continue to evolve, the role of CPUs in systems is becoming increasingly important, which means that the demand for server DRAM is also increasing at each node.

The business model is particularly critical for an excellent company. "Dual-drive" is the key term. The company generates stable cash flow by mass-producing standard memory chips (such as DDR5, SSD) while obtaining excess profits through customized, high-value AI-specific memory (such as HBM, CMM).

Thanks to its strong cash generation ability, the company benefits from the upward cycle of memory price increases, allowing it to support massive capital expenditures (Capex). Against the backdrop of HBM being in high demand, the company's determination to expand capacity has also been evident, which is one of the main purposes of SK Hynix's financing in the U.S. The company is building a large-scale semiconductor cluster in Yongin, South Korea, and advanced packaging plants in Cheongju and Indiana, USA, ensuring capacity supply for the next 5-10 years.

02 The Cyclical Battle

The storage industry has always been a highly cyclical industry. The million-dollar question now is ------ when will the cycle return, and when will stock prices react in advance? Recently, the South Korean storage sector has continued to pull back, bringing this question to the forefront.

In AI servers, the storage value accounts for as much as 40%, due to the continuous expansion of the number of tokens processed by AI, leading to a nearly fourfold increase in DRAM capacity demand, with prices rising due to supply shortages. At the same time, cloud vendors and storage manufacturers have successively signed long-term supply agreements (LTA), actively smoothing out past cyclical fluctuations. For at least the next two years, the traditional "storage cycle" has basically disappeared.

"But this does not mean the cycle will never return." Stuart Rumble, Chief Investment Strategist for Fidelity International in the Asia-Pacific region, previously told Tencent Finance, "Historically, every wave of technology infrastructure ultimately experiences a phase of excessive investment, and storage will not be an exception. Major storage companies are expanding production, but new capacity is expected to come online only between 2028 and 2029. The capital market typically trades fundamental changes 12 to 18 months in advance ------ this means that the pressure of capacity becoming a market theme may not occur until 2027.

Even though strong profits have led to a contraction in SK Hynix's valuation (i.e., becoming cheaper), Asif Suria told Tencent Finance, "The contraction of valuation multiples may actually be a bearish signal because, for cyclical stocks, it often appears that valuations look cheapest when the industry cycle peaks and profits are at their best."

03 The Three Core Controversies in the Current Storage Sector

Regarding the recent pullback in the storage sector, Morgan Stanley has pointed out three core controversies in the market:

Controversy 1: Did Meta's Sale of Computing Power Indicate an AI Bubble Burst?

An unverified rumor circulated last week in the market: one of the world's largest AI computing power buyers intends to sell surplus computing resources. The bearish interpretation is obvious ------ if large-scale cloud vendors have excess computing power, has the entire AI construction become oversupplied?

Morgan Stanley's judgment is: the real answer will wait for the Q2 financial report.

Selling computing power does not equal excess; it is more likely a business behavior by cloud vendors to optimize capital returns. The real judgment criterion is whether large-scale cloud vendors maintain or raise capital expenditures in their Q2 financial reports ------ if they raise, storage stocks will welcome an excellent buying window; if they lower, the narrative of excess will continue to ferment.

At the same time, new variables have emerged in token economics: many companies previously encouraged employees to consume as many tokens as possible, but this led to IT budget overruns, and companies are now starting to cut token usage and turn to cheaper alternative models ------ companies are building an "orchestration layer": simple queries are handled by open-source models, while complex queries go to flagship models. This means that the AI supply chain performed well in Q2, but uncertainty in guidance for the second half of the year is rising.

Controversy 2: Why Haven't Long-Term Supply Agreements (LTA) Driven Valuation Reassessment?

The signing of LTAs by storage stocks should be a significant positive, but the stock price reaction has been tepid. Morgan Stanley provided a direct answer:

The market is rational ------ it remembers that LTAs have been renegotiated in the past or ultimately forced customers to accept unwanted inventory (similar to semiconductor companies during the COVID period). The market's indifference to LTAs is essentially a reasonable doubt about whether the "structural nature of AI demand" can truly persist. Morgan Stanley believes that current storage LTAs do indeed have structural characteristics (as long as AI demand remains strong), but whether profit expectations can continue to exceed expectations is the biggest uncertainty facing the bulls.

Controversy 3: Cycle Peak or Cycle Extension?

Morgan Stanley clearly distinguishes between two concepts: "rate of change peaking" ≠ "cycle ending".

The issue is that the year-on-year price increase has narrowed significantly from over 100% in Q1 to single-digit to low double-digit ranges in Q3 ------ this is an intuitive manifestation of "rate of change peaking".

Currently, institutions still maintain a long-term bullish stance on the storage industry, citing expected profit growth of 35-40% in 2027 and the wave of large-scale adoption of AI agents. However, they provide three clear signals for the short term:

• Stock prices will continue to be under pressure before the earnings season, and market reactions will depend on how large-scale cloud vendors interpret AI capital expenditures.
• Sector preference: DRAM > traditional memory >> NAND, with the least favorable view on storage module manufacturers.
• In terms of positioning: historically high long positions are difficult to maintain in the current volatile environment, and investors have begun to diversify positions and seek opportunities in stagnation (such as MLCC, semiconductor equipment, etc.).

04 Changxin's IPO "Dream Link"

One is listed on Nasdaq in New York, and the other is targeting the Shanghai STAR Market; the "dream link" between SK Hynix and Changxin Technology has attracted attention.

Changxin Technology will start its IPO on July 16, with an expected market value exceeding 2 trillion yuan. A semiconductor investment banker from a leading Chinese brokerage told Tencent Finance that Changxin's profit logic is to reap the "HBM squeeze dividend" ------ Samsung, SK Hynix, and Micron will lock up over 90% of their production capacity in the high-profit HBM sector, creating a structural supply vacuum in traditional DRAM. The direct result is a cliff-like reduction in the effective supply of general-purpose DRAM such as DDR4, DDR5, and LPDDR5 ------ as the only large-scale manufacturer focused on general-purpose DRAM, Changxin is not making money due to technological leadership but because its competitors are "withdrawing from the market." This is a historic window where "others target high-end, while I collect low-end premiums."

Samsung and SK Hynix have announced a gradual exit from DDR4 production, with some models seeing price increases of nearly 50% within a month in 2025, and an annual cumulative increase approaching 800%. TrendForce predicts that ordinary DRAM profits are expected to surpass HBM3e in 2026 ------ this is the first time in history that such a reversal has occurred.

The image shows Changxin's financial data

Wang Ying, a Chinese stock strategist at Morgan Stanley, told Tencent Finance that regarding market concerns about liquidity squeeze, the listing of quality giants will instead actively promote retail participation, attracting more off-market funds into the market with strong fundamentals and a large float. Additionally, since the beginning of the year, the national team has sold over $150 billion worth of A-shares, and this capital is fully capable of entering the market to stabilize fluctuations during short-term IPO subscriptions.

The impressive performance is the confidence of Changxin at present. Although it does not yet hold sway over HBM, Changxin's financial report remains explosive, with gross margins jumping 36 percentage points year-on-year in 2025, directly stating the reason in the prospectus as "the sales price of DRAM products has continued to rise since the second half of 2025":

• Q1 2026 single-quarter net profit is 5.7 times that of the entire year of 2025.
• Revenue compound growth rate of 160.78% from 2022 to 2025.
• The company originally expected to be profitable in 2026-2027, but actually achieved this more than a year ahead of schedule.
• Guidance for the first half of the year: revenue of 110-120 billion yuan, a year-on-year increase of over six times.

In the short term (now to 2027), Changxin can undoubtedly benefit from "windfall money," as the HBM squeeze effect continues; in the medium term (2027-2030), the company will profit from domestic substitution, as China is the largest DRAM consumer market, but Changxin's current global market share is only 7.67%, and its penetration rate in the domestic market is far from saturated; in the long term (after 2030), the company will need to profit from high-end storage. The prospectus clearly outlines the technical route: first pursue general-purpose DRAM, then attack HBM. Of the 29.5 billion yuan raised, 13 billion is allocated for DRAM technology upgrades, and 9 billion for forward-looking technology research and development ------ this is to prepare for future capacity and technical reserves for entering HBM. However, whether this path can be successfully navigated remains to be seen, as the three mountains of sanction walls, yield walls, and packaging walls still exist.