Micron revenue almost triples, tops estimates as demand for memory soars

Micron reported fiscal-quarter revenue of $23.86B (up from $8.05B year-over-year) and adjusted EPS of $12.20, both well above LSEG consensus (revenue ~$20.07B and EPS ~$9.31/$12.05 depending on poll). Management guided the next period to roughly $33.5B in revenue (vs $9.3B a year ago, >200% growth) and adjusted EPS of ~$19.15, far exceeding analyst expectations (~$24.3B revenue, $12.05 EPS). Net income rose to $13.8B ($12.07/share) and cloud and mobile/client revenues climbed to $7.75B and $7.71B respectively, reflecting strong AI-driven demand and a memory supply crunch.

Analysis

This quarter's surprise at the memory OEM level is better read as a demand-concentration event than a pure secular re-rating: a handful of hyperscaler and AI accelerator customers are compressing the industry’s demand curve, pulling forward orders and forcing a multi-year capex response from producers and tool vendors. That front-loading creates two distinct time horizons — a 3–9 month window where pricing and margins stay elevated as inventories are rebuilt, and a 12–36 month window where aggressive capex by Samsung/SK Hynix/Micron and elevated equipment lead times create a meaningful risk of oversupply if OEM demand growth decelerates.

Second-order winners include wafer-equipment and advanced packaging suppliers (ASML/AMAT/LRCX and high-bandwidth packaging vendors) because longer lead times and node migration increase their pricing power; second-order losers are cyclical, memory-adjacent software/platform names whose multiples have been bid down as funds rotate into hardware. Policy and supply-chain levers (export controls, fab siting decisions) can amplify swings — a single trade restriction or a large foundry capacity commitment can re-route demand away from commodity DRAM within quarters.

The immediate investor calculus should separate operating leverage (near-term earnings upside) from structural risk (capital intensity and eventual supply elasticity). Monitor three high-signal metrics weekly: OEM inventory-to-sales (cloud customers), wafer starts and tool backlog, and spot DRAM contract pricing; a normalization in any of these within 6 months is the clearest trigger that could unwind the rally.

Key Decisions for Investors

• Long MU (Micron) — tactical position 2–3% portfolio, horizon 6–12 months. Use a 1-year LEAP call (buy Jan-2027 calls) sized to capture upside while selling 3-month calls to finance premium. Rationale: capture operating leverage from AI memory pull-in; hedge with a 3–6 month protective put if you want capped downside. Target return 40–80% vs max drawdown ~25–30% if inventory correction begins.
• Long LRCX or AMAT (equipment suppliers) — 9–18 month call position (buy 9–15 month ATM calls). Trade captures capex cascade and tool pricing power; limited downside to option premium, asymmetric upside if tool backlogs extend. Risk: capex pull-forward fades — tighten position if weekly tool backlog growth stalls.
• Pair trade: long MU / hedge with short ORCL (Oracle) — equal notional, horizon 3–6 months. Rationale: rotate from mature software multiple into cyclical hardware earnings leverage; short ORCL as a proxy for stretched enterprise multiple repositioning. Risk/reward: expect outperformance of MU vs ORCL of 30–50% if AI memory momentum stays; risk if enterprise SaaS spending re-accelerates.
• Portfolio hedge: buy out-of-the-money DRAM puts or MU puts (3–6 month) sized 1–2% of NAV to protect against a fast capex-driven oversupply and rapid price collapse. This is insurance — cost justified because a >20% spot DRAM price move can erase multiple points of EBITDA across memory names within a quarter.