「Making money in the A-share market proves you have skill, luck, courage, competence, foresight, insight, and patience.
Making money in the US stock market only proves you have money in the US stock market.」
This has become the reality for most US stock traders in 2026.
Those who quietly bought US memory stocks, uninstalled the app, and went to relax, logged back in one day to find their accounts had multiplied several times over.
Spurred by the US market, A-share memory stocks are also taking off.
Crypto enthusiasts have shifted from discussing memecoins and altcoins to talking about US stocks: 「I live every day in fear of US stocks rising and BTC falling」.
New account holders pose the soul-searching question in group chats: Why is making money in US stocks so easy?
1. Who's Actually Rising in the US Stock Market?
The clear capital theme in the 2026 global market is storage.
Justin Sun was among the first to tout the storage sector in late 2025.
As netizens calculated, if you had bought US memory-related stocks at the time Sun made his call:
If you bought Micron, you'd be up +222%; if you bought Seagate, up +256%; if you bought Western Digital, up +280%; if you bought SanDisk, up +515%.
If you had invested 500,000 yuan in SanDisk stock a year ago, you would now have 15 million yuan.
What exactly is 'storage'?
Memory chips are the components in computers and phones responsible for 'remembering' things, divided into two types: DRAM handles short-term memory, used to temporarily store data while programs run; NAND handles long-term memory, where your phone's photos and files are stored. Choosing between 128G and 256G for a phone—that capacity is NAND.
Globally, the companies capable of producing these two things number no more than five.
The stocks of these five companies have surged this way over the past year:
SanDisk, spun off from Western Digital in February 2025, an old company making USB drives and SSDs, saw its stock price peak at a 22-fold increase.
Micron, a cyclical stock shunned by fund managers for a decade, rose over 550% in a year, with gross margins pulled from 18% to 56%. Apple's gross margin is about 43%, already considered a high-water mark for profitability in tech. Micron is now higher.
SK Hynix is up 123% this year. Samsung is up 94%.
Seagate, Western Digital, both at all-time highs.
Then there's South Korea.
Samsung and SK Hynix together account for over 30% of the weighting of South Korea's KOSPI index. In 2025, they drove the entire Korean stock market up 76%, clinching the annual champion title among major global indices.
Two memory makers' explosive performance lifted an entire nation's stock market.
On the price side, it's even more direct. DDR4 memory chips, $1.45 at the start of 2025, peaked at $17 in February 2026, an almost 12-fold increase in a year. Kingston 16GB memory sticks in Huaqiangbei rose from 200 yuan to 800 yuan. Part of the reason your recent phone or computer purchase was more expensive lies in these stocks you didn't buy.
SK Hynix's Q1 2026 net profit exploded by 398%, with an operating margin of 72%. Samsung Electronics' overall operating profit skyrocketed 755% year-on-year.
Sell 100 yuan worth of memory, 72 yuan is profit, 28 yuan is cost. This isn't doing business anymore; this is mining.
2. Institutions Have Lost Their Minds Even More Than Retail
In a typical market, institutions are the ones in sharp suits, expressionlessly saying "we are bullish long-term on fundamentals," while retail investors are the ones shouting "to the moon!" in group chats.
In the 2025-2026 storage sector, the institutions went crazy first.
Google, Microsoft, Amazon began placing "price-unlimited, quantity-unlimited" open orders with Micron.
The term "price-unlimited" is worth pondering. It means you name your price, we pay it, no haggling. This procurement method usually appears in wartime scenarios when governments buy military supplies.
In 2025-2026, it appeared in tech companies buying memory sticks.
Broadcom locked in supply through 2028.
SK Hynix said at an investor conference: "Our 2026 HBM production capacity is completely sold out."
Completely. The entire year.
HBM is high-end memory specifically designed to work with AI chips. For every AI chip Nvidia sells, a piece of HBM must accompany it. Globally, only SK Hynix, Samsung, and Micron can make HBM, with SK Hynix holding about 57% market share. "Completely sold out" means one of the most critical components in global AI infrastructure has no spares for the entire year of 2026.
Then came the analysts.
Within three months, Wall Street's consensus estimate for SanDisk's 2026 EPS was revised up by 172%. Citigroup predicted a 144% year-on-year increase in the average price of server DRAM in 2026. Nomura said the supercycle would last at least until 2027, with meaningful supply increases earliest in 2028. Melius upgraded Micron to Buy after the stock had already risen hundreds of percent, adding "still 41% upside over the next 12 months," without blushing or losing breath.
DeepMind CEO Demis Hassabis publicly stated that the overall memory supply chain is constrained, hampering massive AI deployment. Intel CEO Dr. Randhir Thakur (Note: The article uses "陈立武", likely a mistake or alias; actual CEO at the time is Pat Gelsinger, but the translation follows the text) said memory shortages won't ease before 2028.
Then SK Hynix secretly filed with the SEC to issue ADRs in the US, aiming to raise up to $15 billion. A company with fully sold-out capacity and a 72% profit margin decided to go raise more money in New York, reasoning that the Korean market gives too low a valuation, while US investors understand AI better and are willing to pay a higher price.
The A-share market followed suit.
DeMingLi hit limit-up, BIWIN Storage surged, Longsys rose 41%. Small player Shannon Semi's Q1 net profit was preliminarily estimated to grow 6714% to 8747%, a quadruple-digit increase. Topics in finance groups shifted from "Is CSI 300 still worth buying?" to "Which to buy, Micron or SK Hynix?" People who didn't know how to spell HBM two months ago started explaining how High Bandwidth Memory works in group chats.
Even many matchmaking groups started discussing memory stocks.
3. The Most Ironic Scene
On February 24, 2026, Citron Research announced it was shorting SanDisk, giving three reasons.
First, memory is cyclical. 2008, 2012, 2018—each period of high profitability ended in a crash. Existing capacity is already double the 2018 peak. Supply release is just a matter of time.
Second, SanDisk sells a commodity.
"Nvidia has a moat; SanDisk is just a commodity." Nvidia's moat is its CUDA software ecosystem; almost all global AI models run on it, making switching costs extremely high.
SanDisk's SSDs? Samsung could make an identical one tomorrow, possibly cheaper.
Third, major shareholder Western Digital is massively selling SanDisk shares at a 25% discount to the market price.
Selling their own stock at a 25% off. Needing cash urgently is one possibility; another is thinking it will be cheaper later. Neither scenario is called being bullish on future prospects.
Two trading days later, SanDisk rebounded and later continued to hit new all-time highs. Citron's report circulated in various finance groups, becoming meme material.
One question was skipped by everyone: Whose accounts did those shares sold at a 25% discount ultimately end up in?
4. Making Money in US Stocks, As Easy as Breathing?
The world's three most profitable memory companies, at the peak of their profits, collectively chose not to expand production.
SK Hynix's 2025 capital expenditure related to HBM fell 50% year-on-year, officially citing concerns about oversupply in 2027. Samsung's 2026 DRAM capacity growth is only about 5%, far below demand growth.
The entire industry's capital expenditure growth is only 14%, while historically during each expansion phase it's typically 30% to 50%.
Three companies controlling 92% of global DRAM capacity simultaneously choosing not to expand production has a name in any other commodity market: supply-side coordination. OPEC did this with oil, resulting in the 1973 oil crisis. The concentration in the memory chip market is even higher than OPEC's; the combined market share of these three is something thirteen oil-producing nations can't achieve.
Investors interpreting "manufacturers' restraint in expansion" as a positive is logically not wrong; prices can indeed be sustained longer. But what this structure means for the buyers on the other end isn't in any analyst report.
Two plausible narratives can be told about this rally.
The first: AI's demand for memory is a structural shift. AI models in the inference era need to remember increasingly long contexts, requiring memory in orders of magnitude more. The three major memory makers control 92% of capacity, new fabs won't be operational until 2027 at the earliest, and the gap won't disappear before then.
The second: It's just like every other time in history. The 2000 dot-com bubble narrative was "the internet changes everything," which was true. The 2008 subprime narrative was "housing prices won't fall nationwide," which also held up based on historical data back then. The real problem is never whether the story is right, but whether the price has already discounted that story.
The memory industry has an ironclad rule, unbroken for 30 years: Prices rise slowly, but fall fast.
During the 2018 supercycle, from peak to halving took less than two quarters.
No one knows which day this cycle's peak will be. Including those selling shares at a 25% discount, or rather, especially those selling shares at a 25% discount, because they are selling chips—and selling chips to those who are currently believing the story is the most efficient.
The last time you bought a phone, upgrading the memory from 128G to 256G cost you an extra three or four hundred yuan. That three or four hundred yuan traveled through an industry chain, distributing profits layer by layer, and a tiny portion of it ultimately appeared in SK Hynix's 72% operating margin, in Samsung's 755% profit growth, and in all those stocks you didn't buy.
It also, of course, finally aggregates into the moment you open any social media app and see others' soul-searching question: Why is making money in US stocks so easy?
