Micron Grew 10X in 12 Months. Can It Do It Again?

Micron's fiscal Q3 revenue surged to $41.5B from $9.3B a year ago, while net income jumped to $28.2B from $1.9B, underscoring extraordinary AI-driven demand for HBM and DRAM. Management says HBM output is sold out through 2026 and can meet only about 50%-66% of customer demand, with 16 long-term supply agreements already signed. Despite the stock rising nearly 10x and market cap reaching about $1.31T, the shares still trade at roughly 8x forward earnings versus expected 165% average annual EPS growth over the next five years.

Analysis

The key market implication is that memory has moved from a cyclical input to a strategic choke point in AI capex, and that re-prices the entire stack. If HBM remains rationed, the incremental margin pool shifts from compute vendors to memory suppliers and to the OEMs that can secure supply early; everyone else becomes hostage to allocation rather than demand. The second-order effect is that component inflation can slow unit growth in adjacent hardware categories, forcing cloud providers and OEMs to redesign bill-of-materials and prioritize higher-ROI AI deployments.

The more interesting signal is not just pricing power, but contract durability. Long-dated supply agreements reduce the classic memory-bust setup by pulling future demand forward and lowering near-term inventory risk; that should keep earnings revisions positive for multiple quarters even if spot pricing cools. The flip side is that once customers have locked supply through 2026, the market may underestimate how much volume is already spoken for, which limits downside from a normal cyclical slowdown but also caps upside from a simple multiple rerate.

A contrarian read: consensus is likely extrapolating the shortage too linearly. The real risk is not demand disappearing, but supply normalization arriving faster than expected once new wafer starts hit and customers pre-buy less aggressively. That would compress the forward multiple first, long before revenue rolls over, so the stock is more vulnerable to any hint of capacity additions, mix deterioration, or guidance language implying inventory build in the channel.

For the broader market, consumer electronics pricing pressure is an early warning that memory inflation is now a tax on non-AI hardware margins. If these costs persist for another 2-3 quarters, expect downward earnings revisions in mature device categories while AI-infrastructure names retain estimate momentum. That divergence supports a relative-value trade rather than an outright beta bet.

Key Decisions for Investors

• Stay long MU into the next 1-2 quarters, but express via call spreads rather than outright stock; the thesis is estimate drift from constrained supply, while the main risk is multiple compression if the market starts discounting 2027 normalization.
• Pair trade: long MU / short AAPL over the next 3-6 months. Memory inflation is a direct margin headwind for consumer device makers, while MU is the cleanest beneficiary of the same pricing environment.
• Add a smaller relative long MU / short MSFT basket if AI capex intensity stays elevated; the trade captures the transfer of value from AI consumers to AI infrastructure suppliers, with MSFT more exposed to rising server build costs than market expects.
• Use any 10-15% pullback in MU to build size only if management commentary still supports sold-out HBM through 2026; if guidance weakens, cut quickly because the stock will likely de-rate before fundamentals visibly roll over.
• Avoid chasing downstream hardware names with weak pricing power; instead, rotate toward memory-exposed suppliers and away from OEMs that cannot pass through component inflation within a single quarter.