‘Ride It Until 2028,’ Says Investor About Micron Stock

Micron’s AI-driven demand remains exceptionally strong, with products sold out through 2027 and only 50% to 70% of ordered volumes currently being delivered. The article cites projected EPS growth of 600% in FY2026 and another 75% in FY2027, while the stock still screens at 12.9x forward P/E versus a 15-year average of 21.6x. Despite near-term upside, the investor warns that capacity expansion could eventually normalize pricing and advises exiting by late 2027/early 2028.

Analysis

The near-term setup is structurally supportive for memory makers because pricing is no longer being set by marginal consumer PC/phone demand, but by hyperscaler capex with much longer planning horizons and far less tolerance for supply interruptions. That shifts Micron from a classic cyclical semi to a quasi-strategic AI infrastructure bottleneck, which can justify a valuation premium for several quarters even if the broader semiconductor tape weakens.

The key second-order effect is that supply discipline can persist longer than skeptics expect because new capacity is expensive, slow to qualify, and risky to bring on before visibility deteriorates. The market is underestimating how much of the current margin expansion could flow through to free cash flow rather than just headline EPS, which matters because balance sheet repair and buybacks can amplify upside into 2026-2027. The bigger risk is not demand rolling over immediately; it is simultaneous capex normalization across hyperscalers plus incremental capacity from Micron and peers creating a rapid inflection in pricing power.

Consensus looks directionally right on the upcycle but may be too casual about the exit window. Memory tends to peak well before revenue does, and when pricing inflects, earnings estimates usually lag by multiple quarters, which is exactly why forward multiples often look cheap at the top. The article’s central miss is that the stock may still work higher even as the probability of a 12-18 month drawdown rises materially late in 2027 if capacity additions and inventory rebuilding collide.

For now, the trade is less about whether the AI buildout is real and more about how long the bottleneck lasts. The best risk/reward is to own the secular squeeze while avoiding complacency about terminal value: this is a momentum-and-fundamentals long, not a forever hold.

Key Decisions for Investors

• Long MU into the next 2-4 quarters on pullbacks; use the name as a relative-value long against weaker semiconductor cyclicals where end-demand is less AI-anchored. Risk/reward remains favorable while hyperscaler ordering visibility stays intact.
• Buy MU call spreads dated 6-12 months out to express upside with defined downside, since implied volatility should remain supported but outright equity still carries late-cycle risk. Structure for a 2:1 to 3:1 payoff if the market rerates the multiple on continued supply tightness.
• Start planning a staged exit or hedge program for late 2027 using collars or put spreads, because the highest-risk window is when new capacity and customer inventory normalization converge. The trade should be treated as time-limited rather than permanent.
• Pair long MU against a basket of non-AI memory or broad hardware names that lack pricing power, to isolate the AI-supply bottleneck premium. This reduces beta and monetizes the relative scarcity of Micron’s product mix.
• If MU rallies sharply on guidance revisions, trim strength rather than wait for the cycle to turn. Memory inflections are usually obvious only in hindsight, and the market can re-rate quickly once estimates stop moving up.