Listen to the audio version of this article (generated by AI).
Wall Street has a peculiar complaint about Micron (MU). Business is booming, but perhaps it isn’t booming quite spectacularly enough.
Revenue has surged and earnings have exploded, while management’s forecast points to another record quarter. Yet, after the latest report, investors still found reasons to worry.
Margins could be peaking. The earnings beat wasn’t as big as previous beats. Memory is cyclical, after all.
We understand the history, but we think investors are drawing the wrong conclusion from it.
The question is how much earning power Micron can sustain before supply catches up with demand. And the latest evidence suggests that catch-up could take years.
More than 75% of Micron’s fiscal 2027 shipments are already committed. Customer allocation discussions are moving into 2028. Some agreements extend beyond 2030.
Meanwhile, our recent valuation work put the stock at roughly six to seven times forward earnings, depending on the share price and forecast used.
That is a skeptical valuation for a company whose customers are already trying to secure production years down the road.
We believe that disconnect is the opportunity in the AI memory supercycle.
Micron’s Latest Report Raises the Earnings Bar Again
In its Sept. 30 report, Micron posted fiscal fourth-quarter revenue of $54.23 billion, up from $11.32 billion a year earlier. Adjusted earnings reached $33.42 per share, versus $3.03.
For fiscal Q1 2027, management projected revenue of $60 billion to $63 billion and adjusted EPS of $37.15 to $39.15. The midpoints of $61.5 billion and $38.15 point to another sequential increase in sales and profits.
The sticking point was gross margin. Micron guided to an adjusted gross margin of approximately 86.25%, down from 87% in Q4. However, a modest decline in the percentage of revenue retained after production costs does not automatically mean total earnings are rolling over.
Micron’s own guidance illustrates this with slightly lower gross margin, higher revenue, and higher earnings per share.
Investors should watch the trajectory of the whole business before declaring the cycle over.
When we examined Micron in July, we focused on its long-term customer agreements and what they suggested about the scramble for memory supply.
The latest update strengthens that argument.
More than three-quarters of fiscal 2027 shipments are already committed, and customers are discussing allocations for the following year. Management also indicated that it still cannot identify when DRAM supply will catch up with demand.
Think about what that means from a customer’s perspective.
If you expected plentiful supply and falling prices around the corner, you would have less reason to reserve so much capacity in advance. You could wait for suppliers to compete for your business.
Instead, customers are trying to make sure the memory they need will be available when their systems are ready.
Those commitments improve visibility. They do not make every future shipment irrevocable, nor do they mean every committed unit carries a fixed price. Volume commitments and pricing protection are separate things.
But they give us more reason to believe this expansion has room to run.
AI Agents Add Another Layer of Memory Demand
The memory story extends beyond the high-bandwidth memory attached to AI accelerators.
AI agents also need the rest of the server to do useful work: CPUs to execute tasks, system memory to keep information available and storage to retain and retrieve it. Micron specifically identified agents as a growing source of CPU demand, with implications for memory and storage across the system.
As AI moves from answering individual questions to carrying out longer workflows, more of the computing system gets involved. According to management, customers are limiting memory per system to stretch scarce supply across more deployments. That suggests current purchases may understate what customers would install if availability improved.
As supply expands, customers could buy additional systems and increase the memory inside each one. That is a potential second source of demand, although its size will depend on economics and how workloads evolve.
It helps explain why we see a supercycle taking shape, as successive waves of AI deployment keep adding demands on an industry that cannot expand overnight.
The Memory Supercycle Still Has Room to Run
Our bullish view does not require memory to stop being cyclical. It requires this expansion to last longer than investors currently expect.
Micron’s latest update supports that view: customers are committing early, AI agents are creating additional system requirements, and major new capacity will take years to arrive.
We remain constructive on Micron because we see a business with improving demand visibility trading at a valuation that reflects deep skepticism about its staying power.
The wider lesson reaches across AI infrastructure. As demand grows, the companies supplying essential components can gain both pricing power and a clearer view of future orders.
Memory is one of those constraints. Power and the physical infrastructure needed to support computing are others.
For investors willing to look beyond the most familiar AI names, those dependencies open up a much broader field of opportunities.
