Micron (MU) Earnings Report Analysis: Record Revenue of $41.4 Billion, SCA Strategy Reshapes Storage Industry Valuation Logic

Written by Xiaobing

In one quarter, $41.4 billion in revenue, a staggering 346% year-over-year increase. Gross margin of 84.9%, net profit of $28 billion.

These numbers would be shocking for any tech company, and almost absurd for Micron. In the same quarter two years ago, the company's revenue was $9.3 billion, gross margin was 39%, and net profit was less than $1.9 billion.

But the 15% jump in after-hours stock price shows that the market's excitement is no longer centered on Q3 itself. What truly ignited sentiment was the Q4 guidance: $50 billion in revenue (midpoint), gross margin around 86%, and earnings per share of $31.

The storage super cycle is underway.

Behind the Numbers: Dissecting a Money-Making Machine

Breaking down Micron's earnings report, every business segment speaks with multiples growth.

DRAM contributed $31.3 billion in revenue, accounting for 76% of total revenue, with average selling prices surging over 60% quarter-over-quarter. NAND recorded $9.9 billion, also well above expectations. However, the most explosive growth was hidden within business units: core data center revenue exceeded $25 billion in the single quarter, annualizing to over $100 billion, more than 7 times the $1.53 billion from the same period last year. Data center SSD revenue exceeded $5 billion, doubling quarter-over-quarter. The automotive and embedded business also recorded $4.63 billion, up more than three times year-over-year.

Operating cash flow was $25.39 billion, with adjusted free cash flow of $18.3 billion. Cash and investments on the balance sheet totaled $30.2 billion, with net cash of $24.4 billion. Debt was reduced by $4.4 billion this quarter, and all three major rating agencies simultaneously upgraded it to BBB+.

This is an earnings report that makes the old saying "storage is a cyclical industry" seem outdated.

16 SCAs: Micron is Rewriting Its Own Business DNA

Numbers can be explained by supply-demand imbalances, but the mutation of the business model cannot.

Micron announced during the earnings call that it has signed 16 Strategic Customer Agreements (SCAs). These are "take-or-pay" binding contracts, extending from 2026 to the end of 2030, covering approximately 20% of DRAM shipments and one-third of NAND shipments. Among them are 4 hyperscale customers and 3 mid-sized customers, with the rest coming from the automotive industry.

CFO Mark Murphy disclosed a new metric for the first time: Remaining Performance Obligations (RPO). As of the end of Q3, RPO was $5 billion; after including new agreements signed after the quarter ended, this number jumped to approximately $100 billion. Management clearly stated that actual revenue will "far exceed" the contract floor corresponding to the RPO.

More critical is the goal: Micron plans to increase the revenue share covered by SCAs to over 50%.

What does this mean?

The storage industry's business model over the past 40 years has been based on spot pricing and short-term contracts, with prices frequently skyrocketing or crashing, causing stock valuations to be suppressed by "cyclical discounting." The essence of SCAs is to transform storage chips from commodities into pre-sold infrastructure resources, exactly the same logic as cloud computing companies signing long-term contracts to buy electricity or fiber optics.

If Micron can achieve 50% SCA coverage by 2027, its revenue predictability will approach that of an enterprise software company. Yet its current forward P/E ratio is only slightly over 10x. Among trillion-dollar market cap companies, there is no cheaper one than this.

Supply Can't Catch Up with Demand

CEO Sanjay Mehrotra's statement during the earnings call is worth savoring repeatedly: Micron currently "does not see a point in time where supply will catch up with growing demand." He expects DRAM and NAND shortages to continue beyond 2027.

This is not empty talk. Micron's entire HBM capacity for 2026 has already been fully sold out in terms of price and volume. The production ramp for 12-layer HBM4 is twice as fast as the previous generation HBM3E, and it has already contributed over $1 billion in revenue. Management predicts the total HBM market will grow at a compound annual growth rate of about 40%, from $35 billion in 2025 to $100 billion by 2028, two years earlier than previously expected.

Supply-side constraints are physical. Building an advanced storage fab takes 3-4 years and tens of billions of dollars in investment. Micron has raised its fiscal 2026 capital expenditure to approximately $27 billion (after government subsidies), with new factories in Idaho and Japan ramping up, adding $100-200 million in startup costs each quarter. But capacity release is slow, while AI data centers' thirst for storage is exponential.

Global hyperscale cloud vendors' AI data center capital expenditure for 2026 totals over $725 billion, and all this money will eventually flow through storage chips.

The Timing of the Anthropic Deal is Worth Pondering

Two days before the earnings release, Micron announced a strategic agreement with Anthropic, covering joint design of storage architecture, a multi-year supply contract, deployment of Claude within Micron, and a strategic investment in Anthropic's Series H funding. With this, all three global HBM suppliers (Samsung, SK Hynix, Micron) have become strategic investors in Anthropic's Series H.

No AI lab has ever locked in all three manufacturers of the global HBM supply chain. The significance of this deal goes far beyond the supply contract; it marks the beginning of AI companies treating the storage supply chain as a strategic asset. As the efficiency of model training and inference increasingly depends on storage subsystem performance, the bargaining power of storage vendors will only continue to rise.

Concerns Remain, but Their Weight is Shifting

The historical lesson of the storage industry is clear: every super cycle is followed by a crash caused by overcapacity. Will the synchronized expansion by Samsung, SK Hynix, and Micron ultimately repeat the stories of 2018 or 2022?

The difference lies in the structural change in demand. Past storage cycles were driven by consumer electronics; once smartphone and PC shipments peaked, demand would plummet. But AI data centers' storage demand is continuously cumulative: each generation of models is larger, each inference request consumes more tokens, and each agent requires a longer context window. Add to that AI PCs raising standard memory from 16GB to 32GB, and flagship phone DRAM demand rising simultaneously, the floor of storage demand has been structurally elevated.

The SCA system is Micron's institutional hedge against this cyclical risk. Even if demand growth slows, take-or-pay contracts can cap the downside of revenue. This cannot eliminate the cycle, but it can significantly reduce the amplitude of fluctuations.

Micron's earnings season is over, but the questions it raises are just beginning: When storage transforms from a commodity into a strategic resource that needs to be reserved, should the valuation language of this industry be rewritten?

Based on the Q4 guidance of $31 EPS annualized, Micron's current forward P/E ratio is around 10x. NVIDIA, which also benefits from AI and faces supply shortages, has a forward P/E ratio of over 30x. In this gap lies both the market's pricing of the "cycle will eventually return" narrative and an old story that is being gradually disproven by the SCA system.