KeyBanc raises Micron stock price target to $600 on strong demand

Micron reported fiscal Q2 EPS of $12.20 vs $8.79 expected (38.79% surprise) and revenue of $23.86B vs $19.19B expected (24.34% surprise). Data center strength drove Cloud and Core DC growth of 163% and 211% YoY, DRAM pricing rose ~mid-60% q/q and NAND ~high-70% q/q, and Micron disclosed multi-year (five-year) supply agreement plus volume HBM4 shipments for Nvidia. KeyBanc raised its price target to $600 (from $450) and Barclays to $675 (from $450); management expects free cash flow to double to $13.7B in fiscal Q3 from $5.5B despite ~ $7B capex, underscoring materially improved fundamentals and sector implications.

Analysis

The market is re-pricing memory as a strategic bottleneck rather than a commoditized input. That elevates suppliers with HBM4 and DRAM scale into quasi-duopoly pricing power for high-performance AI stacks, which should compress the payback period on incremental FCF into a much shorter window (quarters rather than years) and force downstream OEMs to prioritize supplier allocation over spot-channel buying.

Second-order beneficiaries will be server OEMs and systems integrators who can secure multi-quarter supply commitments; conversely, GPU incumbents without guaranteed HBM allocation face product cadence and ASP risk even if compute demand remains healthy. Tight HBM pushes yield optimization and packaging services up the value chain, creating margin opportunities for foundry/OSAT partners and potential arbitrage for companies that can convert die-to-module faster.

Key risks are classic cycle reversals and structural shifts: a large cloud capex pause, architectural moves to reduce HBM dependence, or an aggressive capacity response from other memory suppliers would quickly unwind current pricing. Regulatory/export shocks or a sudden move to more inference-focused workloads (which use less HBM per TFLOP) could also flip tailwinds into headwinds within 1-4 quarters.

Consensus is underestimating duration risk of supply discipline: tight markets that drive outsized returns also incentivize capex acceleration, but that lag gives a clear, tradeable runway. If you believe the current tightness persists through the next 12 months, there is a path for asymmetric returns; if not, downside from a cyclical overshoot is large and fast.

Key Decisions for Investors

• Long MU equity (12-month horizon): overweight position sized to capture re-rating from structural HBM/DRAM tightness. Target +35-50% upside; hard stop -18% to control a cyclical selloff. Hedge 20% of notional with put protection 6-9 months out to limit tail risk.
• Long SMCI (3-6 month tactical): buy call spread to play accelerated server wins and allocation premium to systems integrators. Target +40-60% if order flows continue; max loss limited to premium paid — use spreads to cap cost.
• Long NVDA calls (9-12 months) as convex play on AI demand; finance by selling near-term calls against a portion of existing NVDA exposure if already long. Risk: supply/packaging constraints could cap upside in near-term, so keep position size moderate.
• Relative/hedge: pair long MU / short broad semicap or lagging memory peers (where available) to isolate memory pricing exposure. Aim for 2:1 long/short notch so you capture pure memory tightness while limiting market beta; rebalance monthly and tighten stops after any sudden capex-related news.