$MU

Summary

The analyst is bullish on Micron (MU) due to strong earnings beats, record margins, large strategic customer agreements, and structural supply constraints supporting long-term growth.

Reasoning

Micron reported strong revenue and EPS beats, record gross margins, and significant operating cash flow. The company secured 16 strategic customer agreements worth $100B in minimum contracted revenue, de-risking the cyclical nature of memory stocks. Supply constraints are expected to persist, supporting pricing power. Guidance for Q4 FY2026 revenue and margins is very strong, and new fab capacity is coming online earlier than expected. The analyst views MU as a key memory layer in the AI capital cycle with durable demand.

$MU EARNINGS Earnings Breakdown — Not Just a Beat. A Statement. THE NUMBERS ➡️ Revenue: $41.46B vs $35.25B est — beat by ~17% ➡️ Non-GAAP EPS: $25.11 vs $20.28 est — beat by 23.8% . ➡️ Gross margins hit a company record 84.9% ➡️ Operating cash flow: $25.39B vs $11.90B last quarter ➡️ Revenue +346% YoY from $9.3B a year ago ➡️ Stock surged ~13% after hours WHERE THE GROWTH CAME FROM ➡️ Data center revenue climbed more than sevenfold to $11.5B from $1.53B a year ago ➡️ Data center SSD revenue crossed $5B ➡️ Cloud memory surged +300% to $13.77B ➡️ Mobile & client business grew +250% to $11.52B ➡️ Even automotive & embedded memory quadrupled to $4.63B Every single segment lit up. This wasn’t AI carrying the number — AI is pulling everything up with it. THE SCA STORY — THE REAL ALPHA HERE This is the most underappreciated part of today’s report. Micron revealed 16 Strategic Customer Agreements (SCAs) representing approximately $100 billion in minimum contracted revenue. ➡️ The $100B figure is the minimum contractually enforceable revenue — not the full expected value ➡️ $22B in cash deposits and financial commitments tied to those agreements — ~$18B cash deposits + ~$4B letters of credit ➡️ The company will now return 100% of excess cash to shareholders ➡️ Micron is no longer a cyclical memory company. It’s becoming a contracted AI infrastructure provider The SCAs essentially de-risk the classic boom-bust cycle that has plagued memory stocks for decades. HBM4 UPDATE — AHEAD OF SCHEDULE ➡️ HBM4 12-high volume ramp is tracking twice as fast as HBM3E 12-high ➡️ Micron has already shipped over $1B in HBM4 revenue ➡️ HBM4 is in high-volume shipments for their lead customer’s platform (👀 $NVDA Vera Rubin) ➡️ HBM market share strategy: management said they are targeting share close to their overall DRAM share — a deliberate, disciplined approach rather than chasing SK Hynix Q4 FY2026 GUIDANCE — THE EVEN BIGGER STORY Q4 Revenue guide: $50.0B ± $1B (analysts expected ~$42.9B — this blew the roof off) . ➡️ GAAP gross margin guided at ~86% . ➡️ Non-GAAP EPS guide: $31.00 ± $1.00 ➡️ Q4 capex guided at ~$10B, full-year FY2026 capex ~$27B They’re guiding to $50B in ONE quarter. Full year FY2025 revenue was ~$36B. Let that sink in. SUPPLY CAPACITY BUILDOUT ➡️ Taiwan’s Tongluo site (newly acquired) expected to support meaningful shipments in mid-calendar 2027 — pulled in one quarter earlier than prior guidance ➡️ Singapore site becoming a center of excellence for advanced packaging, contributing to HBM capacity from H1 2027 ➡️ Idaho 1 fab also pulled in to mid-2027 THE SUPPLY CONSTRAINT THESIS — STILL INTACT This is the core reason to hold $MU long-term: ➡️ Micron stated they “currently do not have line of sight as to when memory supply will be able to catch up with increasing demand” ➡️ Key customers can only secure 50% to two-thirds of their bit demand requirements ➡️ New fabs won’t deliver meaningful output until fiscal 2028 Quartr The demand wall is structural. The supply response takes years. Pricing power stays. WHAT TO WATCH FROM HERE ➡️ SCA expansion — are more of the 17 contracts being signed? Any mega-hyperscaler (MSFT, AMZN, GOOG) confirmation? ➡️ HBM4 market share vs SK Hynix into late 2026 — can $MU narrow the gap? ➡️ Margin trajectory — can 86%+ gross margins hold or expand into FY2027? ➡️ Capex clarity — $27B in FY2026 capex requires continued demand confidence; watch management commentary on FY2027 spend ➡️ Vera Rubin / Blackwell allocation — $MU is locked into next-gen $NVDA platforms; watch GPU shipment ramp as a lead indicator ➡️ Non-HBM DRAM pricing — DDR5, LPDRAM, server DRAM all quietly printing record ASPs. Underappreciated margin driver $MU is the memory layer of the AI capital cycle. When the hyperscalers build, memory is the one thing you cannot substitute or skip. $MU → $MRVL → $CRDO → NeoCloud → Energy. The cycle continues. Not financial advice.

Return by horizon