Should You Buy Micron Before Sept. 30?

Micron, up 1,300% over five years but roughly 20% below its peak, reports fiscal Q4 results on Sept. 30 amid debate over whether AI-driven memory demand can sustain the current cycle. The stock trades at 22.9x trailing earnings and 6.6x estimated fiscal 2027 earnings, reflecting investor concern that new capacity arriving in 2028-29 and Nvidia's reported shift to lower-spec 4-hi HBM could eventually create oversupply. Offsetting those risks, industry demand is estimated at 2x-10x available supply, Micron has locked in about 20% of DRAM bits and 33% of NAND bits under long-term agreements, and it can resume buybacks from Dec. 9.

Analysis

MU’s setup is less about the reported quarter than whether management can demonstrate that contracted revenue is converting commodity-like spot exposure into visible, multi-year gross-profit durability. The key valuation rerating trigger is not incremental AI demand rhetoric; it is evidence that price floors, volume commitments, and HBM mix can protect margins through the 2028 capacity ramp. If credible, MU can sustain a premium to its historical mid-cycle multiple; if contracts are non-binding, lightly priced, or concentrated in a few hyperscalers, the low forward multiple is correctly discounting a delayed but severe earnings reset.

NVDA’s lower HBM configuration should not be treated as a simple negative read-through for MU. Lower memory content per accelerator may broaden system affordability and unit volumes, while HBM manufacturing intensity constrains fungible DRAM capacity; the net effect depends on HBM bits per GPU multiplied by Rubin shipment volumes, not stack count alone. The more important negative second-order effect is that lower HBM content increases the addressable opportunity for SK Hynix and Samsung to compete on qualified supply, potentially limiting Micron’s pricing power even if aggregate demand remains strong.

For Sept. 30, focus on forward pricing, HBM qualification/yield, customer prepayments, and the fraction of revenue—not bits—under enforceable take-or-pay terms. A buyback authorization after the restriction expires is supportive only if management retains enough balance-sheet flexibility to complete fabs without dilutive financing or materially elevated leverage. Over 1-3 months, a guidance raise plus stronger contract disclosure can close the valuation gap; over 6-18 months, the decisive risk is whether industry capex remains disciplined as new capacity approaches rather than whether current supply is tight.

Consensus likely overweights the binary ‘cycle peak’ narrative but may also be underpricing execution risk in converting scarce HBM capacity into profitable, qualified output. A strong print without hard data on contractual economics is a sell-the-news risk after the run-up. Falsify the constructive thesis if implied 2027 EPS falls materially despite higher AI capex, HBM margins fail to expand sequentially, or competitors signal accelerated qualified HBM supply ahead of demand absorption.

Key Decisions for Investors

• Maintain a modest long MU into earnings only for portfolios able to absorb event volatility; add after results only if forward guidance rises and management quantifies enforceable contract coverage/pricing. Target a 10-15% upside rerating over 1-3 months versus a 12-18% downside on weak forward-margin or qualification commentary.
• Use a defined-risk post-earnings structure rather than naked directional exposure: buy a 3-6 month MU call spread after confirmation of guidance and contract quality. Avoid pre-event premium purchases unless implied volatility is below the expected earnings move.
• Monitor MU/NVDA relative performance for a cleaner AI-memory signal: long MU versus short a beta-adjusted NVDA basket only if Rubin unit-volume commentary offsets lower HBM content and MU confirms HBM qualification gains. Exit if NVDA commentary points to materially lower memory dollars per system rather than lower stack count alone.
• Set an alert for disclosures showing revenue contract coverage below the targeted trajectory, not merely bit coverage. Treat a failure to progress toward durable contracted revenue, or any indication that price floors are weak, as a reason to cut MU exposure regardless of a near-term earnings beat.