Micron (MU) Earnings Preview: HBM, DRAM Pricing, and 3 Critical Numbers to Watch

 

Micron Technology reports fiscal fourth-quarter results after the US close on Wednesday, September 30, with the call at 2:30 p.m. Mountain time, or 4:30 p.m. Eastern. The company confirmed the date in an investor relations release on August 26.

 

What makes this print unusually difficult is the bar Micron set for itself three months ago. According to the fiscal third-quarter results filed with the Securities and Exchange Commission, fourth-quarter guidance calls for revenue of $50.0 billion plus or minus $1.0 billion, gross margin of approximately 86%, and non-GAAP earnings per share of $31.00 plus or minus $1.00. That margin would exceed the peak of any prior memory cycle.

 

So the question is not whether Micron beats. With HBM capacity already contracted and conventional DRAM contract price increases visibly decelerating, what the market needs to verify is how long this level of profitability can persist, and what language management uses when it guides the next quarter. A great deal of optimism is already in the price: shares closed at $1,080.53 on September 24, up roughly 279% year to date.

 

 

 

The guidance is the hardest bar in the room. The $50.0 billion midpoint implies roughly 21% sequential growth from $41.46 billion, but fiscal Q4 contains 14 weeks rather than 13. On a per-week basis, the increase is closer to 12%.

 

An 86% gross margin has no precedent. Fiscal Q3 delivered 84.6% on a GAAP basis and 84.9% non-GAAP. The guide implies further expansion from an already record level, even as management has acknowledged a meaningful moderation in the pace of price increases.

 

HBM has become a contract vehicle rather than a short-term earnings lever. Micron has signed sixteen take-or-pay strategic customer agreements covering roughly 20% of DRAM volume and about 30% of NAND volume, with minimum contracted revenue near $100 billion.

 

The conventional DRAM price curve is flattening. TrendForce expects server DRAM contract prices to rise 13% to 18% quarter on quarter in 3Q26, a sharp deceleration, while mobile DRAM increases have narrowed to 8% to 13%.

 

The disagreement is unusually violent. Citi raised its target to $1,300, Wells Fargo cut its target to $1,400, and Michael Burry added to a short position days before the print. Options are pricing an implied move of roughly 10%.

 

 

 

The fiscal third quarter produced a set of numbers that conventional cycle frameworks struggle to explain. Revenue of $41.456 billion compared with $23.860 billion in the prior quarter and $9.301 billion a year earlier, a sequential gain of about 74% and a year-over-year gain of roughly 346%. GAAP net income reached $28.243 billion and non-GAAP earnings per share came in at $25.11. Operating cash flow was $25.388 billion and adjusted free cash flow was $18.304 billion.

 

By segment, the Cloud Memory business generated $13.769 billion, Core Data Center $11.524 billion, Mobile and Client $11.521 billion, and Automotive and Embedded $4.634 billion. Data center revenue exceeded $25 billion in the quarter, an annualized run rate above $100 billion.

 

The problem is that this performance pinned expectations to $50 billion and an 86% margin. Consensus now sits slightly above the guidance midpoint, with aggregated estimates near $50.6 billion in revenue and about $31.27 in earnings per share. Citi analyst Atif Malik raised his target to $1,300 from $1,150 on September 23 and models fourth-quarter revenue of $51 billion with earnings of $31.45. Even the high end of company guidance barely clears the models that have already been revised upward.

 

 

A detail that changes the interpretation sits in Micron's fiscal third-quarter Form 10-Q: fiscal 2026 contains 53 weeks, with 13 weeks in the third quarter and 14 weeks in the fourth.

 

That means the headline 21% sequential increase is not a like-for-like comparison. Dividing $41.456 billion by 13 weeks gives roughly $3.19 billion per week. The $50.0 billion guidance midpoint across 14 weeks gives roughly $3.57 billion per week, a sequential increase of about 12% in weekly run rate. The top of the range implies roughly 14% and the bottom implies under 10%.

 

The adjustment does not change the dollars, but it changes the story. On a per-week basis, Micron's sequential momentum has already stepped down from the explosive pace of the second and third quarters toward something closer to an ordinary high-demand environment. Anyone comparing $41 billion with $50 billion after the print will overstate the acceleration.

 

 

 

Memory margins are a price function. Over four quarters, Micron's GAAP gross margin travelled from 37.7% in the third quarter of fiscal 2025 to 84.6% in the third quarter of fiscal 2026 without pausing. The driver was successive waves of DRAM contract price increases: TrendForce reported that conventional DRAM contract prices rose roughly 93% to 98% sequentially in 1Q26, lifting total DRAM industry revenue 81% quarter on quarter to $97 billion.

 

That curve is flattening. TrendForce noted in July that several US cloud providers have signed multi-year long-term agreements that restrict suppliers from raising prices for those customers, narrowing the expected third-quarter server DRAM increase to 13% to 18%, with quarterly gains continuing from the second half of 2026 through the second half of 2027 at a moderating pace. The mobile picture is clearer still: TrendForce's third-quarter mobile memory report puts the quarterly increase at roughly 8% to 13% and expects further convergence in 4Q26 under the combined pressure of soft end demand and elevated inventory.

 

Slower increases are not the same as declines. TrendForce has lifted its 4Q26 contract price outlook, expecting cloud demand to keep DRAM and NAND prices rising while weak consumer demand narrows gains elsewhere. For margins the distinction is decisive. Prices holding at elevated levels can defend the current margin, but only continued large increases can push it from 84.9% higher still.

 

 

The other half of the margin equation is cost. Micron's 1-gamma DRAM node and G9 NAND node are ramping and, per management commentary on the June call, are on track to become the highest-volume nodes in company history. Node migration is the structural support beneath the margin.

 

Pressure runs the other way through depreciation. Across the first nine months of fiscal 2026, expenditures for property, plant and equipment totalled $19.602 billion against depreciation and amortization of $6.862 billion. New capacity is being built simultaneously in Idaho, New York, Japan, Singapore and Taiwan, and almost none of it has entered the depreciation schedule. When Idaho ID1 produces first wafers in mid-calendar 2027 and Singapore advanced packaging begins contributing HBM capacity in the first half of that year, the fixed cost base steps up. Asked on the call about margins beyond 86%, management declined to guide past the fourth quarter and instead reiterated that tightness should persist beyond 2027.

 

 

 

On the product side, Micron disclosed that HBM4 built on 1-beta DRAM is in high-volume shipment for its lead customer's platform, with qualification samples shipped to multiple end customers, while HBM4E on 1-gamma is in development for volume production in calendar 2027. Management said on the call that HBM4 twelve-high is ramping about twice as fast as HBM3E, that more than $1 billion of HBM4 has already shipped, and that HBM3E and HBM4 capacity is fully booked through 2027 with customer requests extending into 2028 and beyond.

 

The competitive picture has not changed as quickly. IDC data cited in SK hynix's registration statement filed with the Securities and Exchange Commission put SK hynix at 56.4% of HBM revenue in the first quarter of 2026 and 29.1% of overall DRAM. Micron's stated goal is to bring its HBM share closer to its DRAM share, and progress on that path shapes how the market models the 2027 revenue mix.

 

There is a counterintuitive wrinkle worth holding onto. HBM is not necessarily the highest-margin product right now. TrendForce flagged late last year that as DDR5 contract prices climbed and HBM3E competition intensified, DDR5 profitability would surpass HBM3E beginning in 2026. That helps explain why Micron's highest-margin segments in the third quarter were Core Data Center and Mobile and Client, both at 87%, ahead of Cloud Memory at 83%.

 

 

The structural shift comes from the strategic customer agreements. The 10-Q confirms they are structured as take-or-pay agreements with binding commitments for specific volumes over multi-year terms. On the call, management described sixteen such agreements covering roughly 20% of DRAM volume and about 30% of NAND volume through calendar 2030, with fourteen of them carrying cumulative minimum revenue near $100 billion and approximately $22 billion in customer financial commitments, including about $18 billion of cash deposits. Management's framing was that even at floor prices, margins sit above any prior cycle peak.

 

Those agreements explain two unusual balance sheet lines. Receivables jumped to $31.025 billion at the end of the third quarter from $9.265 billion at fiscal year-end, while other noncurrent liabilities rose to $7.086 billion from $1.443 billion. The first reflects the scale of revenue growth and payment terms, the second relates to customer deposits. Both are worth checking on the day, because they describe contract execution more faithfully than the income statement does.

 

The sober reading is that long-term contracts lock in volume and a price floor, not today's prices. If spot and contract pricing falls back toward those floors in 2027, revenue becomes more certain while margin upside narrows. The value of these agreements is a shallower trough, not a longer peak.

 

 

 

The first is whether gross margin lands near 86%. It speaks to pricing power more directly than revenue does. A meaningful miss on margin would be read as a cycle signal even alongside a revenue beat.

 

The second is the fiscal first-quarter guide. Historically, Micron's share reaction has tracked forward guidance more closely than reported results. The market wants to know whether management is willing to extend the high-revenue, high-margin framing one quarter further, and what vocabulary it uses to describe the pricing environment while doing so.

 

The third is capital expenditure. Fiscal 2026 spending has risen toward roughly $27 billion, with management pointing to a further step up in fiscal 2027 and more than half of the year-over-year increase tied to construction. Capex is simultaneously the leading indicator of future supply and a direct deduction from free cash flow. The faster it rises, the more pressure it places on the supply-demand balance beyond 2028.

 

 

Positioning into the print is crowded. Options on the October 2 weekly expiration imply a single-day move of roughly 10%. That matches recent behaviour: across the last three quarterly reactions, results beat each time while the stock's response ranged from a single-digit decline to a mid-teens gain. A beat has not reliably predicted direction.

 

The split runs through the sell side as well. In the same week Citi raised its target, Wells Fargo cut its target to $1,400 from $1,525 while raising earnings estimates. The sharper opposition comes from Michael Burry, who according to Stocktwits said in a September 22 Substack post that he added to his Micron short "in some size," arguing memory producers had run up to ridiculous prices. He cited Acer's comment that DDR4 now has more sellers than buyers and pointed to improving yields at Chinese supplier CXMT as a source of future price pressure.

 

For traders, the practical consequence is that the risk around the print is two-sided. Exposure to Micron no longer flows only through equity markets either: MEXC disclosed in its monthly operating update that individual stock contracts rose to 30% of its traditional finance futures volume in July after three consecutive months of gains, with memory names among the fastest-growing contracts.

 

 

 

 

If reported gross margin meets or exceeds 86% while first-quarter guidance steps up on both revenue and margin, the bullish thesis gets its key validation: AI demand has lengthened the top of the memory cycle rather than merely delaying its end. In that outcome, the revenue floor embedded in the strategic customer agreements gets repriced as valuation support instead of a growth ceiling. The supporting evidence to watch is whether HBM4 qualification broadens beyond the lead customer and whether additional agreements are announced.

 

 

The bear case does not need a miss. It only needs a shift in the language around price. If management retreats from the framing that tightness persists beyond 2027 toward something vaguer, or if first-quarter margin guidance comes in below 86%, the market is likely to treat that as the first marker of an inflection.

 

External variables are real too. Capacity expansion by Chinese suppliers, inventory building in consumer end markets, and oversupply in mature products such as DDR4 would all hit conventional DRAM pricing first before reaching the blended margin. None of this is likely to reshape the second half of 2026, but all of it shapes how the market prices 2027.

 

 

Beyond the print itself, several checkpoints follow. TrendForce's fourth-quarter contract price data will show whether the deceleration matches the forecast. Quarterly results from SK hynix and Samsung will provide an independent read on HBM and conventional DRAM. Mid-October industry equipment events typically surface signals about 2027 capital spending intentions. For Micron specifically, the clearest tell remains whether the count of strategic customer agreements keeps rising, since that is management's most direct statement of confidence in the length of this cycle.

 

 

James Mitchell's view is that the most common misreading of this print concerns which metric carries the information. Attention is fixed on whether revenue clears $50 billion, but in an industry where price is set by the size of a supply gap, revenue is the number most easily distorted by short-term factors. The extra week makes the point: on a per-week basis the sequential increase is about 12%, not the 21% the headline suggests. Gross margin carries the signal, because it measures pricing power, and pricing power is the function of where the cycle actually stands.

 

The second misreading concerns the nature of the long-term contracts. The strategic customer agreements are often described as locking in $100 billion of revenue, but what they lock in is volume and a price floor, not current prices. That distinction matters for risk management. The agreements raise the floor under future cash flows while surrendering the price elasticity of that portion of output. They make Micron's cash flow look more like a contracted capital-intensive manufacturer and less like a pure cyclical, but they also mean that when prices fall, contracted customers do not renegotiate upward, and Micron cannot fully reprice that capacity in the next upswing.

 

The third point is about positioning rather than fundamentals. Technically, the stock trades near the upper end of a 52-week range that runs from $154.65 to $1,255.00, after a year-to-date gain approaching 280%. At that location, the roughly 10% implied single-day move does not look expensive. Results have beaten expectations in each of the last three quarters while the share reaction has spanned from a decline to a substantial gain, which is what happens when a high level of expectation is already embedded in price and the direction of marginal news matters more than the absolute figure. Anyone taking a directional position ahead of the print should size it against the exposure implied by that volatility, not against confidence in the outcome.

 

The variable most worth tracking afterwards is not the quarter itself but the relationship between gross margin and the rate of conventional DRAM contract price increases. If contract price gains converge toward single digits while margin holds above 80%, product mix and cost declines are genuinely doing the work and the cycle narrative needs rewriting. If margin falls in step with the price curve, then what is special about this cycle is its amplitude rather than its nature.

 

Across assets, the significance of this run extends past semiconductors. Memory has moved from a commoditized component to a strategic bottleneck in AI compute, and the visible expression of that shift is long-dated take-or-pay contracting displacing spot negotiation. Energy, shipping and the previous cloud infrastructure build all went through versions of the same transition, and the usual result was a narrower cycle rather than no cycle. For cross-asset and crypto traders, it also explains why price moves in the AI hardware chain now feed through compute costs and data center capital spending into the pricing of risk assets more broadly.

 

 

 

Micron has confirmed it will report after the US market close on Wednesday, September 30, 2026, with the earnings call at 2:30 p.m. Mountain time, equivalent to 4:30 p.m. Eastern. The call will be webcast from the company's investor relations site with a replay available for approximately one year. The date was formally confirmed in an investor relations release on August 26.

 

 

Micron's own guidance calls for revenue of $50.0 billion plus or minus $1.0 billion, non-GAAP earnings per share of $31.00 plus or minus $1.00, and gross margin of approximately 86%. Sell-side consensus sits slightly above the midpoint at roughly $50.6 billion in revenue and about $31.27 in earnings per share. Citi's model is more bullish at $51 billion and $31.45. A year earlier, the company earned $3.03 per share.

 

 

Because it would exceed the peak of any previous memory cycle. Fiscal Q3 delivered 84.6% on a GAAP basis and 84.9% non-GAAP, so the guide implies expansion from an already record level. With conventional DRAM contract price increases already decelerating, whether that margin is delivered says more about where Micron sits in the cycle than the revenue line does.

 

 

Fiscal 2026 is a 53-week year, with 13 weeks in the third quarter and 14 in the fourth. In dollar terms, the $50.0 billion midpoint is about 21% above $41.456 billion. On a weekly run-rate basis, the third quarter produced roughly $3.19 billion per week against a guided $3.57 billion, an increase closer to 12%. Ignoring that difference overstates the underlying momentum.

 

 

The company does not break out HBM revenue, but it has disclosed that more than $1 billion of HBM4 has shipped, that twelve-high HBM4 is ramping roughly twice as fast as HBM3E, and that HBM3E and HBM4 capacity is fully booked through 2027. Data center revenue exceeded $25 billion in the third quarter. On share, IDC data shows SK hynix still led HBM revenue at 56.4% in the first quarter of 2026.

 

 

Not entirely. The agreements are take-or-pay, and the sixteen signed so far cover roughly 20% of DRAM volume and about 30% of NAND volume with minimum contracted revenue near $100 billion, which genuinely improves revenue visibility. But they lock in volume and a price floor rather than today's elevated prices. When pricing falls, the floor provides protection while margin volatility remains.

 

 

Michael Burry disclosed on September 22 that he had added to a short position, and his case rests on the cycle rather than the quarter. He cited Acer's comment that DDR4 now has more sellers than buyers and pointed to capacity and yield improvements at Chinese supplier CXMT as future price pressure, arguing memory producers had detached from their cyclical character. That is one publicly stated view; sell-side consensus remains weighted toward buy ratings.

 

 

Options on the October 2 weekly expiration imply a single-day move of roughly 10%. Across the last three quarterly reports Micron beat expectations each time, yet the share reaction ranged from a single-digit decline to a mid-teens gain. That suggests no stable relationship between beating estimates and direction, with next-quarter guidance and management's language on pricing usually the more decisive input.

 

 

The information above is provided for general market information and analysis only and does not constitute investment advice, financial advice, legal advice, tax advice or a recommendation to trade. Prices of equities, crypto assets and related derivatives can move sharply, leverage amplifies both gains and losses, and past performance, technical indicators and market data do not guarantee future results. The guidance, analyst forecasts, contract price projections and market data referenced here may change at any time, and official company disclosures and the latest releases from the relevant institutions should be treated as authoritative. Earnings outcomes are inherently uncertain, and any position built on expectations can lose value if reported figures or forward guidance differ from what the market assumed. Readers should conduct their own research and make decisions based on their own financial circumstances, investment objectives and risk tolerance, consulting a qualified professional where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from the use of this information.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise: Technical Analysis, Market Trends & Cycles, Trading Strategies, Bitcoin & Altcoin Analysis, Risk Management.

 

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)