Micron Stock Is Up Over 1,000% Since 2025 and Is Now a $1 Trillion Company. Is It Still a Buy?

Micron is benefiting from a memory chip shortage tied to AI-driven data center build-out, with DRAM and NAND both in tight supply. The article says the stock now trades at about 16x forward earnings this fiscal year and just 9x FY2027 earnings, while Wall Street expects roughly 60% revenue growth in FY2027. Overall, the piece argues Micron still has upside from current levels, though the biggest gains are likely behind it.

Analysis

The setup is less about a single-name rerating and more about the entire memory complex entering a supply-discipline phase. When DRAM and NAND both tighten simultaneously, the usual cyclical fade is delayed because producers have less ability to shift output to the weaker segment, which supports gross margins longer than the market typically underwrites. That creates a second-order beneficiary in AI hardware spend: GPU shipments can keep rising, but the real bottleneck increasingly migrates to board-level memory allocation, so hyperscalers may end up paying up for capacity sooner than current capex plans assume.

The market is still pricing MU like a normal cyclical, but the risk is that this cycle behaves more like a constrained industrial oligopoly for several quarters. If AI data-center demand stays sticky, the earnings runway can extend beyond the current fiscal-year optics and force estimate revisions to step up again, which matters more than the headline multiple. The key tell will be whether lead times and contract pricing remain firm into the next budgeting season; if they do, sell-side numbers are likely still too low for FY27 and beyond.

The contrarian risk is that consensus may be extrapolating AI demand too cleanly into memory profits without accounting for customer inventory behavior. If hyperscalers temporarily over-order to secure allocation, MU can enjoy a strong near-term earnings phase that later normalizes into a digestion period, and the stock could de-rate even while fundamentals remain okay. Another risk is capex response: once pricing visibly improves, capacity additions eventually come back, and memory cycles can turn fast once new wafer starts hit the market; that is a 12-24 month risk, not a next-quarter risk.

Relative to the broader AI basket, MU looks like a cleaner way to express infrastructure scarcity than NVDA at this point, because the market is already assuming extraordinary GPU economics while underappreciating memory’s pricing power. The best trade may be to own MU against a more expensive AI winner where earnings already reflect perfection, or use options to capture a continuation of the rerate without taking full cycle-risk equity exposure. The opportunity is still there, but the margin of safety is now tied to how long pricing stays tight, not whether demand exists.

Key Decisions for Investors

• Go long MU on a pullback over the next 1-2 weeks, targeting a 6-12 month hold; thesis is continued estimate revisions and sustained memory tightness, with upside if FY27 numbers keep moving higher. Risk: rapid supply response or customer inventory digestion.
• Pair trade: long MU / short NVDA for 3-6 months as a relative-value expression of AI infrastructure scarcity; MU has more underwritten upside from pricing power, while NVDA already embeds a higher perfection bar. Stop if memory pricing softens or GPU order momentum reaccelerates materially.
• Buy MU call spreads 6-9 months out to express upside without full cycle exposure; best if you expect continued rerating but want defined downside in case the market starts pricing in a later-cycle normalization.
• Trim or hedge if MU starts trading above the multiple implied by FY27 earnings before consensus revisions catch up; that would signal the market is pulling forward peak optimism and compressing future return potential.
• Watch hyperscaler capex updates and memory contract pricing closely; if either turns less supportive, reduce exposure quickly because memory cycles can reverse in a much shorter window than the current narrative suggests.