Mizuho raises Micron stock price target to $1,150 on memory demand

Mizuho raised its Micron price target to $1,150 from $800 while keeping an Outperform rating, citing 70% fiscal 2027 revenue growth and 85% EPS growth driven by DRAM, NAND, and AI-related demand. The firm sees HBM reaching 23% of fiscal 2028 revenue and expects pricing tailwinds into 2027, with its fiscal 2028 EPS estimate now 41% above consensus. The note is constructive for Micron, though the article also flags the stock as already sharply rerated and potentially overvalued near $928.50.

Analysis

The key takeaway is not that Micron is expensive, but that the market is still underpricing the duration of the memory upcycle. When supply is structurally constrained and AI-driven demand is lumpy but persistent, the earnings power can re-rate faster than consensus models, and memory names can stay “overvalued” on current-year multiples for much longer than traditional semis. The second-order effect is that the entire AI capex stack becomes more fragile: if HBM and DRAM pricing stay tight, accelerator vendors and hyperscalers face a hidden input-cost inflation that can pressure near-term hardware margins even as AI demand remains strong.

The more interesting competitive dynamic is that this is increasingly a supply discipline story, not just an AI demand story. If no major capacity ramps arrive through 2027, the winners are the vendors with the cleanest mix of HBM and leading-edge DRAM exposure; the losers are downstream OEMs and module buyers forced to absorb higher memory cost or delay purchases. That also creates a setup where NAND can surprise to the upside later than the market expects, because the industry’s reflexive tendency is to chase HBM returns while leaving non-AI end markets chronically under-supplied.

The main risk is timing: the bullish thesis likely plays out over months to years, but the stock is now pricing in a very clean 2027-2028 path. Any evidence of capex acceleration, yield improvement, or customer inventory digestion could trigger a sharp multiple reset even if fundamentals remain solid. The contrarian view is that consensus may be too focused on EPS growth and too complacent about peak-cycle framing; a small slowdown in AI order growth could matter disproportionately when expectations already imply sustained scarcity and above-cycle margins.

Key Decisions for Investors

• Stay long MU on pullbacks rather than strength-chasing; use 3-6 month horizons and prefer entries on any 8-12% drawdown, with upside driven by continued estimate revisions and a potential multi-quarter supply squeeze.
• Pair trade: long MU / short a basket of AI hardware beneficiaries with weaker pricing power or higher memory exposure over 6-12 months; the thesis is that memory scarcity compresses margins for downstream names faster than it does for the supplier.
• Buy MU call spreads 9-12 months out to express upside to the 2027 EPS revision cycle while capping premium risk; this is a cleaner way to participate if the market keeps rerating the stock before fundamentals fully flow through.
• Fade overextended semiconductor beta only if there is evidence of memory capex re-acceleration; until then, avoid shorting MU outright because scarcity regimes can sustain elevated multiples much longer than historical valuation anchors suggest.