Should You Buy Micron Before Sept. 30? 1 Number Could Send the Stock Soaring

Micron is scheduled to report fiscal Q4 earnings on Sept. 30, with management's outlook for 2028 high-bandwidth-memory supply and demand expected to be the key stock catalyst. HBM3E and HBM4 capacity is already fully booked through calendar 2027, while strategic five-year take-or-pay customer agreements cover roughly 20% of DRAM and 33% of NAND volume. The company is also investing $100 billion in a New York manufacturing facility expected to begin output in 2029 or 2030, reinforcing its bullish long-term AI-memory demand outlook.

Analysis

The investable question is not whether Micron's contracted HBM demand supports the next few quarters, but whether management can demonstrate that 2028 demand is converting from customer interest into binding capacity commitments. A positive answer would extend the market's terminal-margin assumptions and support a further rerating; vague commentary would expose MU to a sharp de-rating because consensus already capitalizes unusually elevated memory pricing well beyond the normal cycle. The key distinction is HBM allocation versus broader DRAM/NAND pricing: HBM scarcity alone cannot fully offset a conventional-memory correction if AI capex broadens supply faster than server demand.

Strategic contracts reduce volume downside but may cap upside if pricing bands were negotiated before the latest tightening; investors should seek disclosure on indexation, minimum-volume enforceability, customer concentration, and whether HBM is included. The larger second-order risk is capacity discipline at Samsung and SK Hynix: any evidence of accelerated yields or incremental HBM conversion capacity can compress the scarcity premium before physical supply is visible in reported revenue. Conversely, persistent advanced-packaging constraints would shift more value to memory suppliers and reinforce NVDA's system-level bill-of-material inflation.

Near term, MU is an event-driven volatility trade rather than a clean directional entry after a large run. Over 1-3 months, revised FY27/FY28 gross-margin and capex expectations matter more than the reported quarter. Over 6-18 months, the relevant risk is that hyperscaler AI spending becomes more internally funded and return-constrained, reducing accelerator and HBM intensity; META commentary on AI infrastructure ROI is a more useful demand cross-check than consumer-AI product headlines.

Key Decisions for Investors

• Do not add unhedged MU ahead of earnings solely on booked-capacity commentary. Initiate only if management quantifies 2028 commitments or raises long-term margin/return targets; use a 7-10% post-report pullback without estimate cuts as the preferred entry.
• For existing MU exposure, buy a 1-2 month downside put spread around earnings and finance part of the premium by selling an upside call above a 12-15% post-earnings move. The thesis is that upside requires new 2028 information, while ambiguous guidance can trigger multiple compression.
• Express the structural bottleneck view as long MU / short SOXX in equal beta-adjusted dollars for the next 1-3 months, but only if MU confirms HBM pricing or backlog extension. This isolates memory-specific scarcity from a broader semiconductor drawdown; exit if FY27 gross-margin guidance is reduced or peers signal meaningful HBM supply additions.
• Monitor META's next capex guide and NVDA system-demand commentary as falsification checks. A reduction in AI infrastructure spending, evidence of declining memory content per accelerator, or accelerated Samsung/SK Hynix HBM qualification would invalidate the extended-cycle thesis and warrant cutting MU exposure.