FEATURED

Micron Reports June 24

Four days. That’s how long until Micron Technology reports fiscal Q3 2026 earnings after the close on June 24. And right now, heading into that report, the conversation around this stock has shifted in a way that doesn’t happen often.

Analysts are no longer debating whether Micron can grow. They’re debating whether the traditional valuation framework for memory companies still applies at all.

Here’s the recent record, briefly: Micron’s Q2 fiscal 2026 revenue came in at $23.86 billion, up 196% year-over-year and 75% sequentially. Non-GAAP EPS hit $12.20, beating consensus by roughly 39%. Non-GAAP gross margin reached 75%. Free cash flow for the quarter was $6.9 billion. These are not incremental improvements. These are numbers that restructure how you think about the company.

For Q3, the report due June 24, Micron guided revenue of $33.5 billion plus or minus $750 million, with non-GAAP gross margins of approximately 81% and EPS of $19.15 plus or minus $0.40. Wall Street’s consensus EPS estimate has since climbed to around $19.72, up 68% from where it sat just three months ago. That revision trajectory is the real story, not just the number itself. Year-over-year earnings growth implied by consensus is roughly 930%. Stifel raised its price target to $1,500 from $550. Wedbush went to $1,300. Deutsche Bank and TD Cowen both moved to $1,500. RBC Capital and Rosenblatt landed at $1,200. Every one of those targets sits above where the stock is trading right now.

The stock closed at $1,043 on June 17, then jumped another 9% on June 18 as the target hikes landed, closing near $1,134. It has roughly tripled year-to-date, and the 52-week low was $103. That kind of move in a large-cap name is almost without precedent outside of a restructuring or acquisition. This one is driven by earnings.

What’s Actually Driving This

The short answer is High Bandwidth Memory, or HBM. It’s a type of advanced DRAM that stacks multiple memory dies vertically to deliver dramatically higher data transfer speeds than conventional memory. Nvidia’s Blackwell GPUs require it. AMD’s AI processors require it. Every major AI data center buildout requires it. Each Blackwell B200 GPU requires six stacks of HBM3E, totaling 192 GB per chip. The next-generation Blackwell Ultra is expected to push that even higher.

Micron’s entire 2026 HBM production is sold out under binding contracts, including HBM4. Not mostly sold out. Fully contracted. That supply dynamic is what separates this moment from prior memory cycles, where pricing swings were violent and unpredictable because contracts were short and supply could flood quickly. Multi-year agreements and persistent supply constraints are underpinning something closer to durable pricing power rather than cyclical pricing leverage.

Micron has begun volume shipment of HBM4, designed for Nvidia’s Vera Rubin platform, with bandwidth above 2.8 TB/s. The company is targeting over $25 billion in fiscal 2026 capital expenditures to expand manufacturing, with Q3 capex projected at roughly $7 billion. That includes a new megafab project in New York in partnership with Bechtel, and manufacturing expansions in Virginia and elsewhere. Slight tangent, but it matters: Apple CEO Tim Cook recently flagged that memory chip price increases have become unavoidable for the company. That kind of signal, from the CEO of the world’s most valuable business, carries more weight than most analyst notes.

The Structural Question

Here’s the part that most of the mainstream coverage isn’t spending enough time on. Memory has always been a commodity business. Pricing went up, supply flooded, margins collapsed, stocks crashed. Every cycle for decades followed that basic rhythm. The question now is whether AI infrastructure demand has broken that rhythm or just delayed it.

The bull case: AI-driven memory demand is not like PC demand or smartphone demand, which plateaued after initial adoption. Every new AI model generation requires more memory per cluster. Inference workloads, where AI models run in production, are growing even faster than training workloads. Micron projects the HBM market could reach $100 billion by 2028, growing at roughly 40% annually. The architecture of AI computing is bending toward more memory, not less.

The bear case: Samsung and SK Hynix are not standing still. SK Hynix is currently the dominant HBM supplier to Nvidia, with an estimated 43% market share in HBM for 2026. Samsung is competing aggressively. When Micron’s capex eventually adds supply to the market alongside expansions from the other two major players, the pricing cycle could turn faster than the multi-year contracts suggest. Memory has always reverted. Investors who ignore that history pay later.

The honest position is somewhere between those two views. The June 24 report will provide the clearest signal yet about which direction the cycle is leaning.

Technical and Positioning Context

Micron hit an all-time closing high of $1,134 on June 18. Options markets are pricing in elevated volatility ahead of the earnings date, with the implied move running roughly 17% in either direction based on current straddle pricing. At the current stock price, that’s a swing of close to $190 per share in either direction. Active traders should consider position sizing accordingly rather than treating this as a binary bet.

Analyst revenue estimates for fiscal Q3 span $33.7 billion to $40.9 billion, reflecting genuine uncertainty about how fast AI data center investment sustains at current pace. That dispersion is unusually wide even by semiconductor standards. Wedbush’s own Q3 revenue estimate sits at $38.5 billion against a consensus of roughly $34.8 billion. The stock is not cheap relative to trailing earnings. It is cheap relative to forward estimates that themselves carry wide ranges.

Three Scenarios for June 24

Bull Case: Micron reports Q3 revenue near or above $36 billion, gross margins confirm at or above 81%, and management guides Q4 in the $38 to $42 billion range with margin expansion continuing. The structural re-rating thesis accelerates. Stock tests the $1,300 to $1,500 analyst target range in the sessions following the report.

Base Case: Q3 revenue comes in at $34 to $35 billion, in line with or slightly above guidance. Gross margins confirm around 80% to 81%. Q4 guidance is solid but not dramatically above current consensus. Stock drifts 5% to 8% higher post-earnings, institutions continue to accumulate, and the picture heading into calendar Q3 memory pricing data remains constructive.

Bear Case: Revenue meets guidance but Q4 commentary introduces caution, whether around HBM4 ramp timing, competition from SK Hynix, or the capex burden relative to free cash flow generation. The stock pulls back 15% to 20% in a post-earnings flush, reverting toward the $900 to $950 range. This scenario does not invalidate the multi-year outlook. It shifts the entry point.

Who This Matters To

Growth investors get the clearest signal from the magnitude of earnings growth and the long-term contracted revenue visibility. Options traders face one of the more active implied-move events in the semiconductor space this year, with the roughly 17% straddle worth monitoring carefully. Macro-focused investors should note that Micron is a direct read-through on AI infrastructure spending. If the company’s results and guidance confirm that hyperscaler capex is sustaining, that’s a broad market signal, not just a single-stock development. And for investors focused on cash generation: Q2 free cash flow came in at $6.9 billion, with materially higher free cash flow expected in Q3 per management’s own commentary.

June 24 is four days away. The numbers going in are the strongest in Micron’s history. Whether they get stronger, or whether guidance introduces the first note of caution, is the question the market is spending this week positioning around.