Micron Technology briefly surpassed both Meta Platforms and Tesla in market capitalisation on Thursday, June 25, 2026, after a fiscal third-quarter earnings report so far above expectations that it reset how Wall Street is valuing the memory chip industry in real time. Micron shares surged as much as 18.4% during the session to $1,236, lifting the company’s market capitalisation to approximately $1.398 trillion — narrowly ahead of Meta’s $1.392 trillion and briefly within reach of Tesla’s $1.4 trillion, before all three fluctuated through the remainder of the trading day. It was the first time in Micron’s history that the company had topped either rival.
The scale of the underlying results explains why the market reacted with this much force. Micron reported fiscal Q3 2026 revenue of $41.46 billion, up from $9.36 billion in the same quarter a year earlier — a 346% year-on-year increase that comfortably exceeded Wall Street’s consensus estimate of roughly $35.69 billion. Adjusted earnings per share came in at $25.11 against analyst expectations of $20.49. Gross margin reached a record 84.9%, compared with 37.7% a year earlier — a margin expansion of more than 47 percentage points in twelve months, a move without recent precedent for a company of Micron’s scale in the semiconductor industry.
Why Memory, and Why Now
For most of its history, Micron operated in one of the most notoriously cyclical corners of the technology industry. Memory chips — DRAM (dynamic random-access memory, the working memory computers and AI systems use during active processing) and NAND flash storage — have traditionally followed a boom-bust pattern: prices spike during periods of shortage, prompting manufacturers to expand capacity, which typically overshoots demand and collapses prices a few years later. Investors priced memory makers accordingly, assigning them low, cyclical-sector valuation multiples regardless of how strong any individual quarter looked, because the market correctly anticipated that today’s shortage would become tomorrow’s glut.
The AI buildout has disrupted that pattern in a way that is now visibly reshaping how Wall Street values the sector. High-bandwidth memory, or HBM, is a specialised memory architecture that sits directly alongside advanced AI processors — most notably Nvidia’s GPUs — and is essential to their function at scale. Without sufficient HBM, even the most powerful AI accelerator cannot move data fast enough to operate at full capacity; the memory becomes the bottleneck, not the processor. Micron is the only US-based manufacturer of HBM chips at meaningful scale, with Samsung and South Korea’s SK Hynix as its only large-scale global peers. That narrow competitive set, combined with surging demand from every major AI infrastructure buildout simultaneously, has converted what was historically a commodity business into one with the kind of structural pricing power more commonly associated with companies holding genuine technological moats.
The scale of demand driving this shift is enormous and concentrated. The four largest US technology companies — Amazon, Meta, Microsoft, and Alphabet — are projected to spend a combined $725 billion on AI infrastructure in 2026 alone, and every server in those build-outs requires the memory that Micron and its small set of rivals produce. CEO Sanjay Mehrotra was explicit about the structural nature of this shift in the company’s earnings commentary: “Micron’s record fiscal Q3 financial results and even stronger outlook for Q4 reflect the strategic value of memory in the AI era. Micron is investing at record levels in technology, products and supply to address our customers’ rapidly growing demand.” He added a detail that mattered as much to the market’s reaction as the headline revenue beat itself: “We believe our multi-year Strategic Customer Agreements will significantly enhance the durability and predictability of Micron’s strong financial performance.”
That detail — durability and predictability — is the crux of the re-rating. Alongside its earnings, Micron disclosed that customers had committed $22 billion through long-term supply agreements specifically designed to lock in future memory chip deliveries. For a stock market that has historically discounted memory companies precisely because their earnings were unpredictable and cyclical, a $22 billion book of multi-year, contractually committed demand is the single data point most likely to justify a structurally higher valuation multiple — the same logic, in essence, that has supported premium valuations for Oracle and Broadcom on the back of their own multi-billion-dollar AI infrastructure backlogs.
The Comparison to Nvidia’s Own Trajectory
Wall Street’s reaction to Micron’s results is increasingly being framed in direct comparison to Nvidia’s ascent during the earlier phase of the AI infrastructure cycle — and the comparison, while not exact, illuminates why investors are willing to pay up for Micron now. Nvidia’s graphics processing units became the essential compute layer for training and running AI models, and the market rewarded that position with a valuation that eventually made it the most valuable company in the world. Memory chips, and HBM specifically, have emerged as the next critical bottleneck in that same hardware stack — the component without which even Nvidia’s most powerful processors cannot operate at full scale.
That comparison has gained real traction among institutional analysts: at least six banks raised their price targets on Micron ahead of the earnings report, each citing a common thesis — that AI memory demand is structurally outrunning supply, with that imbalance expected to persist well into 2028. The rally that followed the earnings beat extended what was already a remarkable run for the stock. Micron is now up roughly 326% so far in 2026, and has gained more than 15% in the period since reporting earnings alone. The company first crossed the $1 trillion market capitalisation milestone in late May 2026, joining a small cohort of technology companies that have used the current AI investment cycle to reach thirteen-figure valuations within a span of months rather than years.
The Sector-Wide Ripple Effect
Micron’s results did not move the stock in isolation — they triggered a broad rally across the memory and storage sector that reflects the market’s recognition that the structural demand thesis extends well beyond a single company. The Philadelphia Semiconductor Index rose 3.2% on the day of Micron’s results and remains on track, according to LSEG data, for its strongest quarterly performance on record. Among direct peers, SanDisk jumped 22%, Western Digital gained 7.4%, and Seagate Technology advanced 4.3%, while Qualcomm climbed on the back of newly outlined growth plans for its own AI-focused data centre business. In a related sign of the same global dynamic, SK Hynix — Micron’s South Korean rival and a fellow HBM supplier to Nvidia — recently overtook Samsung as South Korea’s most valuable listed company, propelled by precisely the same wave of AI memory demand that lifted Micron past Meta and Tesla.
Despite the strength of Micron’s specific results, the broader technology sector finished the session mixed — a reminder that the memory rally, however dramatic, is a distinct and somewhat separate story from the performance of AI software and platform companies, which face a different set of investor concerns around monetisation timelines and capital intensity that do not apply in the same way to a hardware supplier sitting on $22 billion of contracted future demand.
The Cost the Boom Is Already Imposing
The structural memory shortage that has propelled Micron’s valuation is not a frictionless story for the rest of the technology industry — it is actively raising costs for companies that consume memory chips as inputs rather than manufacturing them. Apple’s share price fell nearly 6% in the trading session following Micron’s results, a decline that came after Apple had separately raised prices across its Mac and iPad product lines, attributing the increases directly to the memory shortage that Micron’s own pricing power is helping to create. Apple’s stock decline extended a difficult month for the company, leaving it down more than 11% over the trailing four weeks — a reminder that the same structural dynamic lifting Micron’s valuation is simultaneously compressing margins, or forcing price increases, for any downstream hardware manufacturer that depends on DRAM and NAND as a bill-of-materials input rather than a revenue line.
This is the other side of the memory supercycle that CLSA’s Korea research team had separately flagged in the context of India’s Dixon Technologies and the broader Asian smartphone supply chain: when memory prices rise sharply enough to materially affect device manufacturers’ input costs, the downstream effect compresses volumes, margins, or both for companies on the consuming side of the memory market, even as it dramatically improves the economics for companies like Micron, Samsung, and SK Hynix on the producing side. The same memory supercycle that is producing one of the most dramatic single-company re-ratings of 2026 in Micron’s stock is simultaneously a measurable cost headwind working its way through smartphone, laptop, and tablet pricing globally — from Apple’s Mac and iPad lines in the US to the smartphone original equipment manufacturers across Asia that source DRAM and NAND as a fundamental input.
What Comes Next: Can Micron Catch Nvidia?
The question increasingly being asked across Wall Street commentary — whether Micron’s trajectory could eventually rival Nvidia’s — reflects both the scale of the re-rating that has already occurred and the genuine uncertainty about how durable it will prove. The bull case rests on the multi-year Strategic Customer Agreements management has specifically cited as evidence that this cycle is structurally different from prior memory booms: $22 billion in contracted forward demand provides a visibility into future revenue that memory companies have rarely, if ever, possessed at this scale. If that demand visibility holds and Micron continues converting it into the kind of margin expansion visible in this quarter’s 84.9% gross margin, the higher, less-cyclical valuation multiple the market is now assigning the stock would be justified on fundamentals rather than momentum alone.
The bear case is the one every analyst who has lived through a previous memory cycle will recognise immediately: memory has been here before. Shortages have always eventually attracted enough new capacity investment — from Micron itself, from Samsung, from SK Hynix, and potentially from new entrants drawn by the current margins — that supply eventually catches up to demand and pricing power erodes. The question this cycle poses, and which six major banks have now staked fresh price targets on answering correctly, is whether AI-driven memory demand is large enough and sustained enough through 2028 that the industry cannot build capacity fast enough to restore the old cyclical pattern before the current structural shortage plays out across several more years of elevated pricing and margin.
For now, the market’s verdict, registered in a single trading session that pushed Micron past two of America’s most recognisable technology companies, is a wager that this time is structurally different — and Micron’s own $22 billion in locked-in customer commitments is the most concrete evidence yet supporting that bet.
Disclaimer: Investments in securities markets are subject to market risks. Read all the related documents carefully before investing. The securities quoted are exemplary and are not recommendatory.
