BofA raises Micron stock price target on AI memory demand strength

Micron reported fiscal Q3 2026 adjusted EPS of $25.11 versus $20.49 consensus and revenue of $41.46B versus $35.69B expected, while BofA lifted its price target to $1,550 from $1,500 and kept a Buy rating. The company also highlighted 16 strategic customer agreements, AI memory demand, and a strong outlook for 2026-2030, with BofA raising out-year EPS estimates by 4% to 7%. Free cash flow margins could approach 50% to 60%, and potential FY27 buybacks were estimated at $32B.

Analysis

The key second-order read-through is that memory is becoming less of a cyclical component input and more of a contractual infrastructure layer inside AI capex. If a growing share of customer demand is locked via take-or-pay structures with pricing bands, the industry’s historical boom-bust amplitude compresses, which should rerate the whole memory complex from trading multiple to mid-cycle cash-flow compounder. That matters not just for MU, but for any supplier or adjacent equipment vendor whose revenues historically depended on spot pricing volatility.

The near-term implication is a tighter squeeze on competitors with less scale or weaker balance sheets. When one producer can lock in long-duration AI supply agreements, customers are effectively pre-allocating scarce advanced memory capacity well ahead of deployment, which can starve smaller players of premium wafers and push them into lower-margin commodity business. Over the next 6-18 months, that creates a split market: quality names with contractual visibility gain multiple support, while laggards face rising inventory risk if they overbuild into a supposedly “strong” cycle.

The consensus risk is that investors may be extrapolating current pricing power too far into 2027-2030 without fully discounting supply response. The more the sector advertises structurally attractive returns, the more likely foundry capacity, capex, and technological substitution accelerate, especially if hyperscalers diversify memory architectures or optimize around lower-memory models. Also, the buyback narrative is real but probably more useful as a downside cushion than a primary catalyst; if AI demand merely normalizes from extreme growth, the stock can still de-rate hard from peak sentiment despite strong FCF.

Catalyst-wise, the next 1-2 quarters matter less for operating beats than for whether management confirms that these customer agreements are enforceable, scalable, and accretive across the product stack. If execution shows the contracts extend beyond a handful of marquee customers, the market should start capitalizing MU like a high-quality infrastructure cash generator rather than a semi-cycle name. If not, the current rally may have front-loaded too much of the long-term story.

Key Decisions for Investors

• Long MU on pullbacks over the next 2-6 weeks; target continuation if management reiterates contract visibility and buyback capacity. Risk/reward favors buying weakness rather than chasing strength after a sharp move, with downside protected by improving FCF and contractual revenue.
• Pair trade: long MU / short a weaker memory or semiconductor cyclical with less pricing power over 3-6 months. The goal is to isolate contract-backed AI memory exposure versus names still dependent on spot pricing and inventory cycles.
• Buy MU call spreads 3-9 months out rather than outright stock if volatility remains elevated. This captures further rerating while defining risk if the market decides the contract story is already priced in.
• If MU trades above the implied multiple embedded in the new target before further contract disclosure, trim 20-30% and rotate into better laggards in the AI infrastructure chain. The trade is best as a convexity bet on cash-flow visibility, not a momentum chase.
• Watch for confirmation that customers are pre-committing to multiple years of demand; if that evidence broadens, consider a basket long in memory-adjacent capital equipment. That would be the cleaner second-order beneficiary as capacity investment follows contract lock-in.