Micron Stock Is Up 700% Over the Past Year. Its Shares Still Look Relatively Cheap.

Micron reported fiscal Q2 sales of $23.9 billion, up 196% year over year, and non-GAAP EPS of $12.20, up 682%, as AI data center demand drives a surge in memory chip sales. Management said it received its first-ever five-year memory contract and sees additional upside from robotics and ongoing AI capex expansion, including higher spending from Meta and Alphabet. The stock has already risen 710% over 12 months and still trades at about 27x trailing earnings, but the article warns future gains may be harder to repeat.

Analysis

MU’s move is no longer just a cyclical upturn; the market is starting to price memory as a scarce strategic input with quasi-contractual demand visibility. The second-order effect is that supply discipline across DRAM/HBM should improve because the biggest customers are effectively underwriting capacity, which reduces the usual boom-bust lag that kills pricing power. That is bullish not only for MU, but also for the broader memory ecosystem, while putting pressure on compute vendors that cannot pass through the higher bill-of-materials cleanly.

The key risk is that the current narrative assumes AI capex remains linear, but the base is now so large that incremental growth becomes mathematically harder to sustain. Any pause in hyperscaler capex would likely hit MU first because expectations are stretched and the stock’s multiple has expanded into a “good news must persist” regime. The timeline matters: near-term the setup is momentum-driven over weeks to months, but the real fundamental test is whether the next 2-3 quarters show pricing stability rather than just unit growth.

The contrarian miss is that the market may be underestimating memory intensity across adjacent AI applications beyond data centers. If robotics or edge inference becomes real, demand could broaden from a single end-market into a multi-year secular layer, extending the cycle and supporting a structurally higher earnings floor. On the flip side, if those optionality narratives fail to monetize, MU could rerate sharply lower even while fundamentals remain solid because the stock has already capitalized years of expected growth.

Key Decisions for Investors

• Stay long MU on pullbacks, but treat it as a momentum/earnings-quality long rather than a deep-value holding; use 6-12 month horizon and size down versus normal semiconductor longs because implied expectations are now elevated.
• Buy MU Jan-2026 call spreads instead of outright stock to capture continued AI-memory upside while defining downside in case hyperscaler capex slows; best risk/reward if entered on post-earnings consolidation.
• Pair trade: long MU / short a basket of AI capex beneficiaries with lower pricing power or more execution risk (e.g., equipment or generic infrastructure names) to isolate memory scarcity as the cleaner expression.
• If MU gaps higher into earnings on no new information, consider selling 1-2 month covered calls against existing longs; the stock should remain supported, but upside volatility is likely to compress after a large move.
• Monitor hyperscaler capex commentary from GOOGL, AMZN, MSFT, and META as the leading indicator; if aggregate capex guidance stops rising for two consecutive quarters, reduce MU exposure immediately because the de-rating risk will arrive before the fundamental slowdown shows up.