Original Author: Jia Liu, Zhang Sheng Beatz
This round of memory shortage has finally passed on to consumers.
On June 17, Cook was still complaining to The Wall Street Journal about cost pressures. He said that when consumers needed devices, the supply decreased, and memory manufacturers were passing the huge price increase pressure down the chain. Memory pricing and supply must return to a level that consumer products can bear.
Less than a week later, on June 25, Cook again accepted an interview with The Wall Street Journal, describing this round of cost shocks as "a once-in-a-century flood." He said that in his more than 40-year career, he had never seen a situation like this in any field. Musk, who had just become the first person in human history to exceed a trillion-dollar valuation, immediately echoed on social media, saying it was one of the most severe price spikes he had ever seen.
Then he announced that Apple would raise prices.
In retrospect, these two interviews seem more like a public opinion preparation for Apple to adjust prices.
Apple's explanation is straightforward: it is not that they want to raise prices, but that the rapid expansion of AI data centers has driven up the demand for memory and storage chips. The company had previously tried to avoid passing costs onto consumers, but now it couldn't sustain it any longer.
This timing is very subtle. The day before Apple announced the price increase, Micron had just released an exaggerated financial report: a gross margin of 84.9%, revenue quadrupled year-on-year, and the stock price surged after hours. The next day, Apple's stock price plummeted, with a market value evaporating by more than $100 billion. Upstream memory manufacturers were raking in cyclical profits, while downstream consumer electronics manufacturers began passing the bills to consumers.
This is the most genuine transmission chain of this round of AI infrastructure expansion: cloud giants are competing for GPUs and HBMs, data centers are fighting for power and servers, memory manufacturers are locking in long-term contracts and raising prices, and finally, Mac, iPad, Xbox, and the iPhone, which might raise prices in the future, will foot the bill.
Moreover, Apple is not an isolated case. Microsoft also announced that it would raise the prices of Xbox consoles starting August 1, with the 512GB model increasing by $100 and the 1TB model increasing by $150, citing the reason that storage and memory prices for consoles have risen more than 2.5 times and could double again by the fall of 2027.
But Micron does not quite recognize this account from Apple.
Micron's Chief Business Officer Sumit Sadana did not directly name Apple, but he told The Wall Street Journal that in the past, some clients were very aggressive on price, causing the entire industry to shut down many investment plans in 2023 due to low prices and low gross margins. Tom's Guide directly interpreted this statement as Micron implying long-term price pressure from major clients like Apple.
Thus, a verbal war began between Apple and Micron: Apple stated, "It's not that I want to raise prices; memory is too expensive"; Micron believes, "In the past, you suppressed prices to a point where no one wanted to expand; now that there is a shortage, don't blame the suppliers."
As consumers, growth in consumption is inevitable, so we need to earn from other places: what will be the next trend driven by this change in storage?
Micron Locks in Profits, Korean Duo Forced to Step Up
To understand how intense this shortage is, first look at the exaggerated financial report just released by Micron, the competitor to Hynix and Samsung.
For Q3 of fiscal year 2026, Micron reported revenue of $41.456 billion, up from $9.301 billion in the same period last year; non-GAAP EPS of $25.11; and a gross margin of 84.9%. The Q4 guidance is even more exaggerated, with a median revenue of $50 billion, a gross margin of around 86%, and a median EPS of $31.
This is no longer the gross margin structure that a cyclical commodity company should have.
More noteworthy than the financial report is Micron's long-term contracts. The company has signed 16 Strategic Customer Agreements, typically covering the period from 2026 to 2030.
Among them, 14 agreements are calculated at the minimum price, with a total income of about $100 billion, and it is expected to receive about $22 billion in customer deposits or financial commitments. The agreements are filled with take-or-pay, minimum purchase quantities, and price floors, with some also setting price ceilings. Micron emphasizes that even with price range agreements, the floor price can bring about gross margins higher than any peak in previous cycles.
While Apple and Micron are arguing, the most significant signal of this verbal fight is actually sent to the two Korean giants.
If they do not expand production, the most profitable orders in the coming years will be locked in by competitors; however, once large-scale expansion occurs, they will have to pay for the next round of cyclical reversal. As a result, both Samsung and SK Hynix have entered an investment race.
At yesterday's press conference, South Korean President Yoon Suk-yeol announced plans to invest 800 trillion won to build four chip factories, with Samsung and Hynix each constructing two new factories. South Korea is betting the next round of national industry on the AI hardware gateway.
The semiconductor industry is already a vital enterprise for South Korea's fortune. Exports, chaebol profits, exchange rates, employment, and stock market valuations are all tied to Samsung and SK Hynix. In the last electronic cycle, South Korea benefited from global digitalization via memory and smartphones; this time in AI, South Korea aims not just for application gateways but for HBM, advanced packaging, and storage supply that every AI server from NVIDIA, AMD, Google, and Microsoft cannot bypass.
Ordinary investors can understand HBM as high-speed memory stacks next to AI chips. AI training and inference require not only GPUs but also memories to continuously feed data. Whoever can get certified earlier and deliver faster will secure the most scarce orders in this round of AI infrastructure.
This is also why Samsung and SK Hynix are being pushed to the table.
The most crucial variable here is the time difference. Expanding production is not just about investing money today and shipping tomorrow. SK Group Chairman Chey Tae-won has already stated that the memory shortage may last until 2030 because new wafer production capacity takes at least four to five years. HBM is even more complex than regular DRAM, involving not just front-end wafers but also TSV, thinning, stacking, bonding, advanced packaging, and an entire set of testing.
In other words, the money South Korea invests today addresses supply issues in 2027, 2028, and 2029 and may not immediately relieve the price pressures currently faced by Apple, PC manufacturers, and server manufacturers.
After Expansion, What Investment Directions Will Fund Flow Into?
Having clarified the expansion, we need to examine the beneficiaries of the expansion narrative. The orders for building factories by these storage companies will first fall on equipment, materials, packaging, facility management, and power systems. They will receive the money first.
Yesterday, JAK KJ hit the daily limit, signaling the market's agreement.
JAK KJ is a supplier of upstream materials for major storage companies like SK Hynix. The market is not betting on its ability to manufacture HBM but rather its position in the material segment in front of HBM and DRAM expansions. As long as SK Hynix, Samsung, and Micron increase wafer production, precursor materials, specialty gases, CMP, photoresists, and wet electronic chemicals will follow in volume.
This is why the storage market typically does not stop at Micron, Hynix, and Samsung but quickly spills over to "upstream storage."
Expansion primarily corresponds to orders for front-end wafer plants. EUV and DUV lithography, deposition, etching, cleaning, ion implantation, CMP, and measurement all require rescheduling.
In US and European stocks, ASML corresponds to lithography and EUV; AMAT Applied Materials covers deposition, materials engineering, and CMP; LRCX Lam Research corresponds to etching, deposition, and cleaning; KLAC KLA corresponds to measurement and inspection; TER Teradyne and COHU correspond to testing; MKSI MKS Instruments and Ichor correspond to vacuum, power supply, and gas-liquid delivery subsystems. These stocks performed much better than storage last night in the US market.
If mapped to A-shares, funds typically seek domestic semiconductor equipment chains: Northern HC, Zhongwei GS, Tuojing KJ, Shengmei SH, Huahai QK, Xinyuan W, Jingce DZ, and others.
The logic is not that they all directly supply Samsung or Hynix but that global storage expansions raise equipment sector sentiment, while domestic Changxin, Changjiang, and others’ expansion and domestic substitution expectations will be re-evaluated together.
Materials and consumables are also the easiest directions for funds to repeatedly explore in this round.
Equipment orders are one-time, while material consumption is ongoing. As long as the production line is up and running, every wafer will consume precursors, electronic specialty gases, photoresists, wet electronic chemicals, CMP slurry and pads, target materials, silicon wafers, quartz components, and filtration materials. HBM is more complex than regular DRAM, with more processing steps and higher material consumption intensity.
JAK KJ is being speculated upon at its core. The market interprets it as an upstream reflection of SK Hynix’s expansion: the increase in Hynix's HBM production drives demand for storage wafer production and high-end process materials, giving suppliers of precursors and other materials a chance to reprice.
Under the same logic, A-share funds will also look for targets along several lines: for precursors and electronic materials, look at JAK KJ, Nanda GD; for electronic specialty gases, look at Huate QT, Jinhong QT; for CMP materials, look at Anji KJ, Dinglong GF, etc.
In the overseas materials chain, ENTG Entegris corresponds to filtration, chemicals, material handling, and CMP-related consumables; LIN Linde and APD Air Products correspond to electronic specialty gases and industrial gases; GLW Corning can serve as an indirect observation point of the glass and materials chain but is not strictly a pure memory material target.
The elasticity of such companies arises from three questions: are they in the supply chain of major manufacturers, can the single wafer material value be enhanced, and will price increases and volume truly reflect on gross margins?
Additionally favorable are domestic substitutions and China's storage capacity.
If the conflict between Apple and Micron continues to ferment, end manufacturers will more actively seek alternative supplies. Changxin Storage and Yangtze Memory may not provide a complete answer for simply replacing Samsung, Hynix, and Micron, but they will become new negotiating chips in supply chain discussions.
As long as domestic DRAM, NAND, and HBM capacities are repriced, A-share equipment and material suppliers will be rediscovered by funds. The addition of Zhaoyi to the DRAM ETF is the best illustration.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
