The 2027 Number Arrived Above $200 Billion and the Stock Fell 11.8%: China Carried September, Supply Capped December, and the Multiple Reset Faster Than the Earnings

Results vs. Consensus

Lam guides four lines each quarter. It beat all four midpoints this quarter but cleared none of the high ends, a step down from June, when three of the four exceeded the top of the range. The revenue and EPS beats against the Street were each about 1%.

"We were pleased with our continued strong execution in the September quarter, resulting in our fifth consecutive quarter of record revenue, as well as record gross margin, operating margin, and earnings per share." — Doug Bettinger, CFO

Guide vs. Actual

Actual vs. Street

Put the two quarters side by side and the shape of the print is clear. In July, the reported quarter beat modestly and the forward guide beat the Street by roughly $1.0B, which drove an 18% rally. This time the reported quarter beat by about the same margin and the forward guide landed on consensus. The forward surprise that moved the stock in July was absent, so attention turned to what was in the quarter, and in this quarter that was mostly China.

Quarter-over-Quarter Comparison

Year-over-Year Comparison

Income-statement, balance-sheet and cash-flow figures are as reported in the Item 2.02 Exhibit 99.1 furnished October 21, 2026. Free cash flow is operating cash flow less capital expenditures and intangible assets. Segment and regional dollar figures elsewhere in this report are implied from disclosed percentages applied to reported totals and are labeled as such; the company discloses those splits in percentage terms only. September 2025 non-GAAP EPS and mix percentages are as stated on the October 2025 call.

Revenue Assessment

September revenue of $8.219B is the fifth consecutive quarterly record, and the 22.3% sequential gain is the fastest in our coverage arc, after 9.4% in March and 15.1% in June. Systems revenue grew 34.9% to $5.733B while CSBG was flat at $2.487B. Systems therefore delivered $1,483M of the $1,497M increase, and CSBG fell to 30.3% of revenue from 36.8%.

The geographic split shows where the money came from. China rose to an implied $2,877M from $1,748M. Management said both the domestic and the multinational customer bases invested strongly, and that foundry strength included mature-node spending in China. Taiwan fell to an implied $1,480M from a record $1,815M, even as foundry posted record revenue. Korea held at 20% and grew roughly in line with the company. Outside China, revenue rose to an implied $5,343M from $4,974M, a 7.4% gain. That is a solid quarter, but it is not the 22% quarter the headline describes.

By segment, foundry contributed the most dollars, an implied $767M increase to about $2,637M. NAND added an implied $513M, DRAM $169M and logic/other $34M. Management characterized NAND's sequential growth as roughly 40%. The rounded mix percentages imply more than that, so we treat the management figure as the anchor and the implied dollars as approximate.

Calendar 2026 now looks like this. Including the December midpoint, Lam's revenue is tracking to roughly $29.4B, up about 43% from calendar 2025's $20.6B, against WFE growth of roughly 41% to the mid-$150 billion range. Management expects calendar 2026 systems revenue to grow "well over 40%." That is the third consecutive year of outgrowing WFE, which the CEO restated on the call.

Margin Assessment

Non-GAAP gross margin of 52.4% is another all-time record, 40bps above June and 40bps above the guide midpoint. The incremental gross margin on the $1,497M of added revenue was 53.9%. Management named two drivers, favorable customer and product mix, and added that it continues to "balance our cost and pricing postures." That is the second consecutive quarter in which management named pricing as a driver without quantifying it.

The mix attribution matters for how much weight the record deserves. In October 2025, when China was 43% of revenue, management said customer mix would be a gross margin headwind "as the China mix normalizes." This quarter China rose 900bps and gross margin rose 40bps. A reasonable reading is that the underlying margin was roughly flat in September and that China flattered it. The December guide is the better test: gross margin is guided up another 10bps to 52.5% "in spite of headwinds related to customer mix," with China expected to fall sharply. If Lam delivers that, the margin gains will have held up through an adverse mix swing, which is the evidence July's framework revision needs.

Non-GAAP operating margin of 40.2% is a record and the first time above 40%. Operating expenses rose 9.6% to $1,004.3M. Management attributed the increase mainly to incentive compensation tied to higher profitability, plus incremental resources for Akara, Halo, panel form factors and Dextro, with R&D at about two-thirds of the total. The incremental operating margin was 48.1%, above the 44.9% implied by the July guide. Headcount grew by about 1,900 to roughly 24,300, mostly in factory and field teams, which is the cost of shipping 35% more systems revenue in one quarter.

EPS and Cash Flow Assessment

Non-GAAP EPS of $2.20 rose 20.9% sequentially, slightly behind revenue growth of 22.3%, entirely because of tax. The non-GAAP tax rate stepped to 16.5% from 11%, exactly as management signaled in July. At June's rate the quarter would have produced roughly $2.35. GAAP EPS of $2.19 includes an $11.5M tax charge from a change in tax law, which non-GAAP excludes. The CFO repeated that the higher rate holds for December "and likely beyond," and asked that models be updated to reflect it. Ours already were.

Cash is where this quarter most clearly improved on July. Operating cash flow was $3.498B and free cash flow a record $3.240B, or 117.7% of GAAP net income, after a fiscal year in which conversion had fallen to 67%. Part of that is a working-capital release: the change in operating assets and liabilities was a $535M source of cash against an $860M use in June, as other current liabilities rose about $1.38B. Removing that release still leaves free cash flow of about $2.70B, roughly 98% of net income. Receivables rose only 6.0% against revenue up 22.3%, taking days sales outstanding to 63 from 72. Inventory rose $274M to $4.55B as Lam builds ahead of demand, yet turns improved to 3.4x.

Capital return stepped back up. Repurchases were $1,025M including excise tax, and included a newly entered accelerated share repurchase that continues into December. Dividends were $325M. Together that is about 41.7% of the quarter's free cash flow on the cash-flow-statement basis. Management cited approximately 41% for the quarter and 59% for calendar 2026 to date, against a long-term target of 85%. The dividend declared rose to $0.33 per share from $0.26, an increase of 26.9%. Cash and equivalents ended at $7.44B against $3.74B of debt, and $3B remains on the authorization.

Segment Performance

China's implied +64.6% is marked red as a concentration signal, not a growth miss.

NAND: Still Growing Fast, With a New Argument Beyond Layer Counts

NAND rose to 26% of systems revenue from 23%, an implied $1.49B. The CFO described roughly 40% sequential growth as conversions to 256-layer-and-above devices continued, with initial spending on 384-layer devices. In July we expected NAND to decelerate from June's doubling while still growing. It did both, but the deceleration was smaller than we modeled. The growth came from both CSBG upgrades and new equipment purchased for the conversions, and management called both meaningful contributors.

The more important NAND disclosure was a new demand argument. Management described AI splitting storage into tiers: QLC for cost-effective capacity, TLC as a high-bandwidth tier for context memory and KV cache, and an emerging SLC-class tier for low-latency agentic workloads. SLC stores roughly a third of the bits per device, so management quantified that each percentage point of NAND bits shifting to SLC-class devices requires about 2% more installed wafer capacity for the same bit demand.

"We believe adoption of an SLC storage tier could become a meaningful driver of NAND wafer capacity." — Tim Archer, CEO

Assessment: This is the most interesting structural claim in the print, because it gives NAND wafer capacity a growth driver that does not depend on bit-demand growth. It answers, at least in part, July's third-party forecast of NAND oversupply in 2027: oversupply measured in bits matters less if each bit needs more wafer area. The caveat is that management would not size SLC adoption and said its customers should speak to it. We treat SLC as unmodeled upside and keep NAND at the bottom of the 2027 growth ordering, where management placed it. We now model NAND systems revenue at roughly $5.8B to $6.5B in FY27, up from $4.0B to $4.5B, which assumes NAND holds near September's level and grows modestly from there.

DRAM: Record Dollars, Smaller Share, and a Roadmap Question Answered

DRAM reached record revenue at 20% of systems, an implied $1.15B, up about 17% sequentially even as its share fell from 23%. Spending stayed focused on technology upgrades across the 1α, 1β and 1γ nodes for DDR5, LPDDR5 and, above all, high-bandwidth memory. Management again named DRAM the fastest-growing segment for 2027.

The quarter also brought a roadmap question to the surface. On October 21, the day of the print, Micron said it would prioritize 3D DRAM over the 4F² architecture, and Lam's shares fell 1.0% while equipment peers rose. The CEO's answer was that etch and deposition intensity rise in either path. A longer wait for the next architecture pulls in advanced patterning, where Aether dry resist and Akara conductor etch are gaining. A direct jump from today's 6F² to 3D DRAM raises etch and deposition intensity "even faster, even sooner."

"Aether solutions are already ramping in HBM high volume manufacturing with Low-NA EUV, and we see additional insertion opportunities ahead across both DRAM and foundry logic customers." — Tim Archer, CEO

Assessment: The 3D DRAM answer is logically sound. A vertically stacked DRAM is structurally closer to 3D NAND, where Lam's high-aspect-ratio etch and deposition franchise is strongest, so the market's reaction looked backward to us. The quarter-level evidence also supports the pillar: record dollars, Aether ramping in HBM production, and DRAM plus foundry/logic accounting for about 80% of Lam's systems growth in calendar 2026 to date. We model DRAM at roughly $5.3B to $6.0B in FY27 systems revenue, above our July range of $4.5B to $5.5B.

Foundry: Record Revenue From the Leading Edge and From China

Foundry rose to 46% of systems from 44%, an implied $2.64B and a record. Management cited leading-edge investment including advanced packaging, plus mature-node spending in China. Taiwan's share fell to 18% from 27% in the same quarter, so the leading-edge component likely grew less than the segment total implies, and China mature-node spending likely did a large share of the work.

The product disclosures were stronger than the mix. At a leading customer, Lam won tool-of-record positions for all four of its backside power products: Vector deposition, Sabre electroplating, Flex dielectric etch and the Halo molybdenum tool. Asked how far Lam has come toward the doubling of foundry/logic SAM per wafer promised at the 2025 Investor Day, the CEO said "maybe halfway." Management also expects mature-node spending outside China to be strong in 2027, after a 2026 in which it did not stand out.

Assessment: Foundry's record owes more to China this quarter than the label suggests. The leading-edge business is still on the trajectory the thesis needs: backside power wins across four products, gate-all-around share gains, and advanced packaging growth above 70%. The 2027 setup also adds a leg we did not have, mature-node capacity outside China driven by tightness in analog, power and microcontrollers. We model foundry at roughly $10.5B to $11.5B in FY27 systems revenue, up from $8.5B to $9.5B.

China: Back to 35%, Guided Sharply Lower

China rose to 35% of revenue from 26%, an implied $2.88B, up about 65% sequentially and about 26% year-over-year. The composition is the opposite of what we flagged as encouraging in July. In June, multinationals in China grew and domestic customers declined, and the CFO said that trend might continue through the rest of the year. In September, both groups were strong. Asked about it on the call, the CFO said the result "wasn't a surprise" and was "exactly as expected."

"In the December quarter, we expect China will represent a significantly lower percentage of our total revenue compared with the September quarter." — Doug Bettinger, CFO

Management expects the December decline to come from domestic customers, with multinationals in China staying reasonably strong. One piece of arithmetic is worth tracking. In October 2025, management expected China to fall below 30% of calendar 2026 revenue. Through September, China has contributed an implied $6.61B of $20.78B in calendar 2026. To finish below 30%, China would need to be about 25.6% or less of December revenue at the guide midpoint.

Assessment: This reverses part of the progress we credited in July, so we move the China concentration bear point back to emerging. The domestic customers that grew are the ones most exposed to export-control escalation, and their spending appears to have lifted both revenue and gross margin in September. The offsetting point is that management guided the decline explicitly and still guided gross margin higher. If December China lands in the mid-20s and gross margin holds at 52.5%, this quarter will read as a lumpy timing quarter. We model China at 26% to 30% of FY27 revenue, up from 22% to 26%.

Taiwan and Korea: A Leading-Edge Pause After a Record June

Taiwan fell to an implied $1.48B from a record $1.82B in June, a decline of about 18.5%. Korea held at 20% of revenue, an implied $1.64B, up in line with the company and consistent with record DRAM spending. Management gave no specific explanation for the Taiwan decline. Our best guess is tool-delivery timing into new fabs plus the supply tightness that management said emerged during the quarter.

Assessment: One quarter of lower Taiwan revenue after a record does not challenge the leading-edge thesis, especially with backside power and gate-all-around wins disclosed in the same call. It does mean the September quarter's growth came less from the leading edge than the headline suggests. We will look for Taiwan to rebound as China falls in December.

CSBG: Flat Sequentially, Raised for the Year

The Customer Support Business Group posted a fourth consecutive record at $2.487B, up 0.6% sequentially and 40.0% year-over-year. That matches July's guidance of a "similar profile." Spares remain the largest component, Reliant was strong on mature-node investment, and upgrades remained strong with the NAND conversion cycle. Management now expects calendar 2026 CSBG revenue to grow more than 30%, more than twice the long-term target set at the last Investor Day. For calendar 2027, it gave a first outlook of more than 15% growth.

"The value we are delivering with Equipment Intelligence and Dextro Cobots recently helped us secure performance-based service contracts with multiple memory manufacturers." — Tim Archer, CEO

Assessment: A flat quarter after a 17% jump in June is consolidation, not deceleration, and the calendar-year guides are both above our July assumptions. Performance-based service contracts are the most strategically useful CSBG disclosure in our coverage arc. They price service on customer outcomes rather than on parts, which is how the Equipment Intelligence investment earns a return. CSBG growth of 15% or more on a base above $9.4B adds at least $1.4B of revenue in 2027. We model CSBG at $10.0B to $10.8B in FY27, up from $9.5B to $10.5B.

Key Operating Indicators

Key Topics & Management Commentary

Overall Management Tone: Management was confident on demand and more operational than in July. It volunteered a 2027 number a quarter early after refusing one four times last quarter, and framed every constraint as supply rather than demand. The call was less convincing on the near term: management declined to size the supply shortfall, the China swing, the unconstrained 2027 figure or the accelerated repurchase. Its explanation that China's domestic strength was "exactly as expected" sat awkwardly against what it had said in July.

1. Calendar 2027 WFE: More Than $200 Billion, Delivered Early and Called Constrained

"Our preliminary 2027 WFE outlook is for more than $200 billion of spending, with percentage growth led by DRAM, followed by leading-edge foundry logic and NAND." — Tim Archer, CEO

This is the largest disclosure of the print. It resolves the single largest open item from July, when management refused to quantify 2027 four times while calling the setup "extraordinary." The CFO noted that Lam does not normally give this much specificity this early and said a "sharper pencil" would come on the January call. Against the mid-$150 billion 2026 outlook, more than $200 billion implies roughly 29% growth.

Two qualifiers come with it. Management described the number as constrained by clean-room availability and by supply-chain limits on certain components, and said unconstrained demand would be higher without giving a figure. Management also said it has customer demand visibility through 2028, with 10 new tier-1 fabs coming online during 2027 and layering in over the year.

Assessment: More than $200 billion sits above the $170B to $185B range we modeled in July, so our 2027 WFE assumption moves to $200B to $215B. A constrained number also behaves differently in a downturn from a demand-limited one. If demand weakens at the margin, a constrained forecast has a cushion of unmet demand to absorb it before the spending figure falls. That is the best answer management has given to the 2028 question the market keeps asking, although it is a claim about visibility, not evidence of 2028 orders.

2. Calendar 2026 WFE Raised a Third Time, to the Mid-$150 Billion Range

"Basically, what we saw happening across all aspects of our customer base is everybody squeezed a little more out of the clean rooms that they had and/or pulled projects in a little bit quicker than maybe they originally intended." — Doug Bettinger, CFO

The 2026 path has gone from the $135 billion range in January to $140 billion with upside bias in April, the low $150 billion range in July and now the mid-$150 billion range. In July the CEO cautioned that further in-year upside would be hard to find. It was found anyway, though the increment is smaller than the July raise.

Assessment: A third raise inside one calendar year, with the year's last quarter already guided, is more confirmation than new information. The mechanism matters more than the size: customers kept finding capacity in existing clean rooms, which supports the argument that the 2027 constraint will bend as well. We set our calendar 2026 WFE estimate at $154B to $157B.

3. China Swings Back to 35%, and It Carried the Quarter

"China contributed 35% of revenue, up from the June quarter level of 26%. We saw strong investments here by both domestic and multinational customers." — Doug Bettinger, CFO

China's implied increase of about $1.13B was roughly three-quarters of the company's $1.50B sequential gain. Management linked foundry's record to mature-node spending in China, and linked the gross margin gain to customer mix. The December guide calls for China to fall significantly, driven by domestic customers.

Assessment: The quarter's surprise was not that China was strong, but how much of the record it explains. Excluding China, September revenue grew 7.4%, and December revenue should grow roughly 16% to 21% if China falls to the 25% to 28% range. Both figures describe a business that is still growing fast, more evenly than the headline sequence of +22% then +5% suggests. The market traded the headline sequence.

4. The December Guide: $8.60 Billion, Held Down by Supply

"We are expecting revenue of $8.6 billion ± $400 million. I would mention that this revenue guide would have been a bit higher if not for some near-term supply chain constraints that we are actively managing." — Doug Bettinger, CFO

The guide implies 4.6% sequential growth and about 61% year-over-year growth. Gross margin of 52.5% and operating margin of 40.5% are both 10bps to 30bps above September. Implied operating expenses of about $1,032M rise only 2.8%, so the guide's incremental operating margin is about 47.9%. EPS of $2.33 implies a tax rate near 16.5%. Asked whether the guide contained anything one-off, the CFO said nothing unique, one-off or episodic, and said shipments that slip from December move into March.

Assessment: The guide met consensus of roughly $8.54B and missed the higher buy-side expectations the stock had priced. The quality inside it is better than the top line: margins expand despite an adverse customer mix, opex is nearly flat, and the revenue that slips is deferred, not lost. The wide range of plus or minus $400M, now about 4.7% of revenue, gives the market room to imagine a lower outcome, and on October 22 it did. We model December at the midpoint and add the slipped revenue to March.

5. Supply-Chain Constraints: New, Specific to Lam's Own Suppliers, and Framed as Temporary

"For those of us who were around during the COVID shortages, this is nothing like that. We learned a lot of lessons during that period." — Tim Archer, CEO

This is the first time in our coverage arc that management has said Lam's own supply chain, rather than customer clean-room space, limited a guide. The CEO identified where it bites: in product areas where etch and deposition intensity is rising fastest, such as advanced packaging, some suppliers see Lam's demand growing much faster than Lam itself. Lam held events with its global suppliers in the past two months, said its suppliers are well capitalized, and said it would consider co-investment if needed. It expects to be "caught up very shortly." The balance sheet shows the strain: other current liabilities rose about $1.38B, which a questioner tied to payables, and inventory grew $274M.

Assessment: A supply limit in a demand-strong market is the better kind of problem, and Lam's turns improved while it built inventory, so this is not a working-capital failure. It still moves our lead-time bear point to emerging. Supply limits shift revenue between quarters, and in a stock priced on sequential momentum, a shift between quarters becomes a price event. The test is March: if the slipped December shipments show up there, the constraint will have been a timing issue.

6. Operating Margin Above 40% for the First Time; Gross Margin "Plus or Minus" Where It Is

"Operating margin exceeded 40% for the first time in our history, and we are guiding for this to expand slightly." — Doug Bettinger, CFO

September operating margin of 40.2% compares with a long-term target of the mid-40s set only in July. On gross margin, the CFO said the near term would stay "plus or minus the level that we are at," with the mid-50s as a multi-year destination. He again named the levers: new products such as Akara and Aether, manufacturing close to customers ("a little bit of juice left to squeeze there"), and pricing.

Assessment: The operating margin path toward the mid-40s is now visibly under way, at 40.5% guided for December. Gross margin is where the July framework still needs proof, and the near-term comment says to expect little movement for now. Our FY27 gross margin assumption stays at 52% to 53%, and we lift FY27 operating margin to 40.5% to 41.5%.

7. NAND Storage Tiers: SLC as a Wafer-Capacity Driver

"Converting from TLC to SLC class devices reduces bits per device by roughly 3 times. Said differently, for each percentage point mix shift of NAND bits to SLC class devices, approximately 2% more installed wafer capacity will be needed to support the same bit demand." — Tim Archer, CEO

Management paired the SLC argument with an AI-adoption data point: average weekly token usage is reported to be up roughly 250-fold in 18 months. It also tied the tiering to agentic workloads that need low-latency storage. The CEO said customers have started discussing SLC-class and high-bandwidth-flash products publicly, and that Lam has the industry's largest NAND installed base and covers every critical deposition and etch step.

Assessment: The framework is a useful counter to bit-based NAND bear cases, and the 1-to-2 ratio is specific enough to model once customers show adoption. Until then, it remains an option on NAND capacity rather than a forecast, and management declined to say what share of bits might move to SLC.

8. Aether Raised by $1 Billion; Akara Productivity Claims

"As a result, we now expect cumulative Aether-related revenue to reach approximately $2.5 billion by the end of the decade, roughly $1 billion above our prior expectations." — Tim Archer, CEO

Aether dry resist has shown sub-10nm resolution with High-NA EUV at about 10% lower exposure dose, and it is in high-volume HBM production with Low-NA EUV. Management said the raise reflects both new customer adoptions and higher volumes at existing production positions. On Akara, management claimed up to 50% reduction in wafer-to-wafer and lot-to-lot variability and up to 50% more wafer output per fab footprint, with multiple chipmakers adopting it.

Assessment: Raising a cumulative revenue target for one product line by about two-thirds, to roughly $2.5B from an implied $1.5B, is a meaningful signal of adoption. The output-per-footprint claim for Akara also matters strategically: in a clean-room-constrained industry, a tool that produces more wafers per square foot competes on capacity, not just on process performance. Both support the etch and deposition leadership pillar.

9. Backside Power: Tool-of-Record for All Four Products at a Leading Customer

"We are gaining share in this expanding backside power SAM opportunity and recently secured tool of record wins for all four products at a leading customer." — Tim Archer, CEO

Backside power delivery separates power and signal routing, which calls for thicker backside metal and more deposition, plating and etch steps. Lam named Vector for deposition, Sabre for plating, Flex for dielectric etch and Halo for molybdenum connections between the backside and the front side. Customers are ramping backside power at 2nm and below.

Assessment: A clean sweep of four tool-of-record positions in a new architecture is the kind of share evidence we track for the leading-edge pillar. It is also a foundry/logic win, the segment where Lam has historically been weaker than in memory. The unknown is which customer and how fast its backside ramp proceeds, and management disclosed neither.

10. Advanced and Panel-Level Packaging: Above 70% Reaffirmed, Panel Share to Exceed Wafer Level

"Given our customer engagement, installed base growth, and technology leadership, we expect to achieve panel-level packaging market share that surpasses the strong position we already hold in current wafer-level packaging." — Tim Archer, CEO

Lam has now shipped panel plating systems to more than 20 customers, across panel sizes from 300x300mm to 600x600mm, with the Kallisto platform positioned for glass-core substrates. In wafer-level packaging, Syndion won four new TSV positions and Striker ALD won multiple TSV liner applications. The CFO said the above-70% growth for 2026 still holds, and that he would be "stunned" if advanced packaging did not grow nicely in 2027.

Assessment: Claiming panel share above an already strong wafer-level position is an aggressive statement, and the CEO also said panel is still at a very early stage. We continue to treat panel as a free option beyond 2028. The persistent gap is the revenue base. After four quarters of growth rates without a dollar figure, management also deferred a sizing request for EMIB-T and CoWoS as an action item.

11. CSBG Outlook: More Than 30% in 2026, More Than 15% in 2027

"We expect CSBG momentum to continue in calendar year 2027, with revenue growing over 15% year-over-year." — Tim Archer, CEO

Asked to rank CSBG's growth drivers, the CEO ranked them by excitement and named Equipment Intelligence, cobot-driven maintenance and performance-based contracts first. By financial weight, spares remain the largest component. Management gave no numbers by component and did not address CSBG operating margins.

Assessment: A first-time 2027 CSBG outlook of more than 15% is above the high-single to low-double-digit framework from the 2025 Investor Day. It gives the model a recurring revenue floor that rises with the installed base regardless of new-tool timing. That makes it the clearest evidence that this cycle's installed base will earn revenue after the shipments slow.

12. Capital Return: An Accelerated Repurchase and a 27% Dividend Increase

"During the quarter, we entered into an accelerated share repurchase structure that we'll continue to execute in the December quarter." — Doug Bettinger, CFO

The quarter's repurchases of about $1.0B compare with $247M in June. Management declined to disclose the size or end date of the accelerated repurchase, beyond saying it continues through December alongside other activity. The 85% of free cash flow target is unchanged, with return at 59% of free cash flow for calendar 2026 to date. Management also noted $750M of maturing debt was repaid during the year.

Assessment: July's buyback cut looked like balance-sheet building, and this quarter confirms it. With record free cash flow and $7.4B of cash, Lam resumed repurchases into a falling stock price and raised the dividend 27%. The $3B remaining on the authorization is less than one quarter of current free cash flow, so a new authorization is likely within the next few quarters. We now model FY27 repurchases of $3.5B to $4.5B, up from $3.0B to $4.0B.

Guidance & Outlook

Implied FY27 trajectory: September's $8.22B plus December's $8.60B midpoint puts the first half of FY27 at $16.82B. Calendar 2027 WFE growth of roughly 29%, plus Lam's stated expectation of outgrowing WFE again and the slipped December shipments landing in March, supports a second half of roughly $18.2B to $19.7B. Our FY27 revenue estimate moves to $35.0B to $36.5B from $33B to $35B.

Street positioning: The December guide sits about 0.7% above a consensus near $8.54B on revenue and in line on EPS, so the guide itself should move Street estimates little. Estimate changes should come mainly from the second half of FY27 and from calendar 2027.

Guidance style: Across the six quarters we have covered, Lam has landed every guided line above its midpoint. EPS exceeded the top of its range in four of the five quarters before this one; this quarter it landed $0.10 below the high end, and its $0.05 beat against the midpoint is the smallest in our coverage arc. We still read the December gross margin guide as carrying upside skew, because it already absorbs a China-mix headwind, and the revenue midpoint as constrained by supply rather than padded.

Analyst Q&A Highlights

How Much the Supply Constraint Cost December, and Whether It Comes Back

The opening question went straight to the line in the prepared remarks that the stock traded on. It cited the payables build as visible evidence of the strain and suggested a $300M to $400M impact. Management declined to size it but committed that the shipments would move into March, not disappear.

Q: "You did talk about the near-term constraints. I guess you can see that from payables. I think payables were up like $1.3 billion. Can you just give us a sense? Is it fair to say this is costing you $300 million-$400 million? Is that a reasonable number to assume that that is costing you in the guidance for December, and do you get that back in March? Do you actually get that back? Thanks."
— Timothy Arcuri, UBS

A: "Yeah, Tim, that is a good question. I am not going to put a specific number on it. In the script, I said it would have been a little bit higher. So it would have been a little bit higher. And yeah, Tim, to the extent that it is something we would have shipped in December, because of timing, we will ship that in the March quarter."
— Doug Bettinger, CFO

Assessment: The refusal to size it left the market to supply its own number, and a "little bit higher" guide on a $400M range is easy to read as more than a little. The commitment to ship in March is the useful part and it is checkable next quarter. Since the guide met consensus even with the constraint, the unconstrained December would have been a beat.

3D DRAM Versus 4F², and Why the Stock Fell on It

A question raised a memory maker's statement earlier that day that it would prioritize 3D DRAM over 4F², and noted Lam's stock had traded down on it. Management answered that etch and deposition intensity rise in either architecture path, and that a slower migration has its own benefit through advanced patterning.

Q: "They were saying that they're going to prioritize 3D DRAM rather than 4F², and your stock traded down, which was a bit odd to me. It seems that people don't really understand what your position is there. So Tim, I wonder, maybe you can talk about that. Can you talk about how you do for in a 3D world versus a 4F² world and what you're seeing in terms of your customers' DRAM roadmaps."
— Timothy Arcuri, UBS

A: "I would say in the short term, etch and dep intensity are rising, and if really we're making a leap right from 6F² into 3D DRAM, then etch and dep intensity is going to rise even faster, even sooner than we would have thought, and that's a great thing."
— Tim Archer, CEO

Assessment: The answer is persuasive on architecture: a stacked DRAM looks more like 3D NAND, Lam's strongest franchise. What it does not address is timing. A roadmap that skips 4F² could leave a gap in DRAM equipment spending between generations, and management did not say whether customers' 1γ-and-beyond plans are changing. We count the exchange as a net positive for the DRAM pillar, not a neutral one.

Whether China's Domestic Strength Was a Surprise

A question asked management to quantify December's China decline and noted that September had surprised to the upside with both customer groups. Management declined to quantify it and said September had gone exactly as expected.

Q: "I think in your prepared comments you mentioned China would be significantly lower in the December versus September. Kind of wondering, can you quantify that? How will the mix look like? Because seems like September quarter was a surprise on the upside with both multinationals and domestics coming in."
— Krish Sankar, TD Cowen

A: "Yeah, Krish. To us, it wasn't a surprise. It was exactly as expected. I mean, the domestic customers in September were strong, but so were the global multinationals. I expect that in December we'll be meaningfully down, driven by the domestic customers in China. I expect in the December quarter, the multinationals in China likely continue to be reasonably strong, but the percent of revenue in China will be meaningfully down in the December quarter."
— Doug Bettinger, CFO

Assessment: "Exactly as expected" is hard to square with July's comment that domestic China declining while multinationals grew might continue through the rest of the year, and with a July guide that did not flag a China surge. Perhaps management knew and chose not to say, or perhaps the timing moved. Either way, Lam's China disclosure has become less predictive, and we give the December "meaningfully down" guidance less weight until it prints.

What 2027 Would Look Like Without Constraints

Having obtained a number, the questioning turned to what it excludes. The question asked for unconstrained 2027 demand, meaning WFE if clean rooms and suppliers were not limits. Management confirmed it would be higher and declined to say by how much.

Q: "For the first one, I understand that this more than $200 billion WFE next year is a constrained number. Do you guys have a point of view on what that number would be if it was not constrained, if clean rooms were available, if supply chain was not as tight as it is? I guess what I am asking is, what do you think the demand for WFEs in 2027 versus the supply number that you have given?"
— Stacy Rasgon, Bernstein Research

A: "Listen, things are going to continue to be constrained, we think, in the next year by the things that we have described. I am not going to put a number on what the unconstrained demand would be. It would be higher. I am not going to give you a precise number. Like I said, we will fine tune what we think the number is and give you a little more specificity on the next quarter call also."
— Doug Bettinger, CFO

Assessment: The gap between constrained and unconstrained demand is the cushion that makes a 2028 air pocket less likely, and management confirmed that it exists without sizing it. We read "it would be higher" as a statement that 2027 is limited by supply, which supports the duration argument. The January call is now the next checkpoint for both the 2027 number and its cushion.

What Moves the 2027 Outlook Higher or Lower

A question asked management to separate customer spending plans from clean-room availability as drivers of the 2027 range. The answer described a balancing act across fabs, equipment supply and hiring, and the CEO added that demand visibility already extends through 2028.

Q: "I was wondering, as you look into 2027, can you maybe frame for us some of the moving pieces that could adjust the WFE outlook higher or lower? How much of it really is your customers' individual spending plans for the different technology or product areas, versus just the general availability of clean room space overall?"
— James Schneider, Goldman Sachs

A: "When we look in the next year, we see 10 new tier I fabs coming online between the beginning of the year and the end of the year, and they will layer in over time. That's obviously a contributor. This year has been primarily clean room constrained. In that, again, you see, like we said, some emerging supply chain constraints that we're confident we're going to be able to work through effectively, but the whole industry is going to have to work through very similar things."
— Doug Bettinger, CFO

Assessment: The answer moves the binding constraint for 2027 from clean rooms alone to clean rooms plus equipment supply plus labor. That makes the number more likely to be pushed between quarters, and less likely to be cut for lack of demand. Ten tier-1 fab openings are the hard evidence underneath the 2027 outlook.

Gross Margin Once China Normalizes

A question asked directly whether gross margin can hold once the China mix tailwind fades, and how pricing and cost efficiency factor in. Management guided to stability near current levels with a multi-year path to the mid-50s.

Q: "I guess on a normalized basis, as we look out to the next few quarters, once China normalizes, how should we think about gross margins? In the past, you did talk about some pricing actions, and of course, you have the cost and efficiency improvements as well. Just curious as to how we should think about gross margins as we go into next year."
— Srini Pajjuri, RBC Capital Markets

A: "Yeah, listen, in the near term, Srini, I think you are going to be plus or minus the level that we are at. We did suggest over the next several years, we see an opportunity to expand gross margin in the mid-50s. You kind of triangulate yourself between these two points. But in the near term, it will bounce around generally where it is."
— Doug Bettinger, CFO

Assessment: "Bounce around generally where it is" is a commitment to hold roughly 52% through a China mix decline, which, if delivered, is better than a mix-adjusted reading of September would suggest. It also tells us not to expect mid-50s progress in calendar 2027's first half. The levers named, new products, close-to-customer manufacturing and pricing, are the same as July's, and pricing is still unquantified.

Why CSBG Was Flat When NAND Upgrades Were Strong

A question pointed out an apparent inconsistency: NAND upgrades sit in CSBG and were described as strong, yet CSBG was flat while NAND grew sharply. The answer explained that conversions also require new systems, and a follow-up obtained the statement that both pieces were meaningful contributors.

Q: "It sounded like upgrades were strong. I do not know if that meant that they grew sequence or not. I assume they did because the NAND revenues grew a bunch, but the CSBG revenues were kind of flattish. I thought the upgrades were in CSBG. If that is the case, I guess what I am asking is, did upgrades drive the NAND growth? If they did, what else was going on in CSBG to absorb the growth such that the overall was flat?"
— Stacy Rasgon, Bernstein Research

A: "Yeah. Listen, Stacy, upgrades are a part of NAND and also CSBG. NAND also, when you go through these conversions, you need to buy new equipment also. Obviously, there was a contribution in demand spending from new equipment also."
— Doug Bettinger, CFO

Assessment: The exchange tells us the NAND cycle has moved from pure upgrades toward new tools. That is a better mix for systems revenue and suggests the conversions are more capital-intensive than in the first wave. It also suggests that upgrades within CSBG did not grow enough to lift the segment, which is consistent with a CSBG growth rate that moderates from here to the 15%-plus management guided for 2027.

How Much of the NAND Market Moves to SLC

A question took management's 1-to-2 SLC ratio and asked for the adoption figure needed to turn it into a forecast. Management declined and deferred to its customers.

Q: "You mentioned every 1% move to SLC requires about 2% increase in wafer capacity. As we look at next year with the move to SLC and potentially some HBF, any thoughts on how you think what percent of the market moves to SLC? Is it like 5% or how would you see that?"
— Vijay Rakesh, Mizuho

A: "I think it is too early for us to make any comment on that, and it would be probably something our customers should actually be commenting on. What you can rest assured is that in that sense, the intensity of Lam equipment in manufacturing those devices and those wafers would go up."
— Tim Archer, CEO

Assessment: A framework without an adoption estimate is a sensitivity, not a forecast. At the 5% shift the question suggested, the ratio implies about 10% more NAND wafer capacity for the same bits, which would be material. We will watch for customers to give that number, since management will not.

What They're NOT Saying

1. The size of the December supply shortfall: Declined. A questioner proposed $300M to $400M; management said only "a little bit higher." The market had to guess, and it sold the stock.
2. Which components are constrained: Management would not identify where the supply tightness sits, beyond suppliers to fast-growing product areas such as advanced packaging.
3. Unconstrained 2027 demand: Confirmed to be higher than more than $200 billion, then left unquantified until January.
4. How far China falls in December: "Significantly" and "meaningfully" lower, no figure. Management also did not revisit its October 2025 expectation that China would be below 30% of calendar 2026 revenue, which now requires December China at about 25.6% or less.
5. Why domestic China grew when July suggested it would decline: Called "exactly as expected" without reconciling it to the July comment.
6. The pricing contribution to gross margin: Named for the second consecutive quarter ("balance our cost and pricing postures") and not quantified, while customer mix did most of the visible work.
7. Why Taiwan revenue fell 18.5%: A $336M implied decline from a record June quarter was stated without explanation and drew no question.
8. The advanced packaging revenue base: A fourth quarter of growth rates with no dollar base. A request to size EMIB-T and CoWoS opportunities was deferred as an action item.
9. The second NAND upgrade wave: Management accepted the action item to size it in July. This quarter brought the SLC framework instead, not the sizing.
10. SLC adoption: "Too early," deferred to customers.
11. The accelerated repurchase: Size, start and end date all withheld.
12. The new US logic customer: July's confirmed "dialogue" was not updated. A question on newly announced logic fabs drew only that Lam knew of the projects in advance.
13. The NAND growth rate: Management cited roughly 40% sequential growth, while the rounded mix percentages imply roughly 46% to 59%, about 52% at the point estimate. The gap was not explained.

Market Reaction

• Pre-print setup (October 21 close, $260.81): up 52.4% year-to-date and 79.8% over twelve months, but down 13.7% over the trailing 30 days from $302.28 on September 21. The stock had peaked at $347.49 on October 2 and fell 24.9% into the print, including ten consecutive lower closes from October 5 to October 16. It entered the print 39.8% below its 52-week closing high of $433.33. The S&P 500 was up 13.3% year-to-date.
• Print day (October 21): the stock fell 1.0% in regular trading while Applied Materials, KLA and the semiconductor ETF rose between 1.0% and 2.1%, after a large memory maker said it would prioritize 3D DRAM over 4F². Shares then fell roughly 6% to 7% after hours following the release.
• Reaction session (October 22): opened at $234.50, a 10.1% gap down, traded a range of $227.68 to $242.10, and closed at $230.15, down 11.8% or $30.66. Volume of 22.8M shares compared with a 9.2M thirty-day average, 2.5x normal. The S&P 500 fell 1.2%.
• Peer reaction: Applied Materials fell 8.2%, KLA 9.5%, ASML 6.9% and Micron 6.3%. The semiconductor ETF fell 6.8% and the Nasdaq-100 ETF 2.9%. Meta, which reported the same evening with free cash flow turning negative, fell 7.0%.
• Net position: at $230.15 the stock is 46.9% below its 52-week closing high, 33.8% below the October 2 high, and 22.7% below the close after the July print. It is now 8.8% below where it traded the day before the July print, despite a forward earnings estimate about 38% higher.

The most useful way to read the session is to split it. The equipment group fell 7% to 10% together, which is more than a reaction to one company's December guide. Lam's print supplied the trigger: the supply-constraint line and a decelerating sequential guide were read across to every equipment name. Meta's same-evening print supplied the second force, by reviving the question of how hyperscalers fund AI capex once free cash flow runs out. Lam fell 2 to 5 points more than its closest equipment peers, and that excess is the company-specific part: a guide that met consensus but not expectations, a China-heavy quarter, and a range wide enough to imagine $8.2B.

The setup also explains the size. The stock had already fallen for most of October in a sector-wide de-rating, so the print had to reverse that trend to work. A guide in line with consensus could not do that. This is the same dynamic as July with the sign reversed. In July, a de-rated stock met a large forward beat and rallied 18%. In October, a stock that had rallied to a fresh post-print high and then started falling met an in-line guide, and fell 12%.

The arithmetic underneath matters most. The day before the print, at $260.81 on our prior FY27 EPS estimate of roughly $9.35, the shares traded at 27.9x. After the print, at $230.15 on a revised estimate of roughly $9.75, they trade at 23.6x. The multiple compressed by about 15% in one session while the estimate rose about 4%. That is below both readings we recorded around the July print, 35.8x before it and 31.8x after, in a quarter that delivered the 2027 number we had been waiting for.

Street Perspective

Debate: Is December's +4.6% a peak-growth signal, or a China-and-supply artifact?

Bull view: The guide is supply-constrained, not demand-constrained, and the slipped shipments move into March. Strip out China's swing and revenue outside China accelerates from 7.4% sequential growth in September to roughly 16% to 21% in December if China falls to the 25% to 28% range. Gross margin rises despite an adverse mix, which is not what a peak looks like.

Bear view: Sequential growth falling from 22% to under 5% is exactly what the top of a cycle looks like, whatever the explanation. Supply constraints are a polite way of saying the cycle is outrunning the industry's ability to deliver, which historically precedes double-ordering and a correction. The guide range is wide enough that the low end, $8.2B, is flat sequentially.

Our take: The bull reading is better supported by the disclosed numbers, but it relies on management's China guidance, which was less predictive this quarter. The decisive data point will be March: a quarter that absorbs the slipped December shipments and grows outside China would settle this debate. We side with the bulls on direction and stay alert to the double-ordering risk the bear case names.

Debate: Does "more than $200 billion" make 2027 the peak?

Bull view: The number is constrained, management says unconstrained demand is higher, and customers have visibility through 2028. Ten tier-1 fabs opening across 2027 represent capacity that will need tools for years. A constrained forecast has a cushion that a demand-led forecast does not.

Bear view: A 29% increase in WFE following a 41% increase means the industry will have nearly doubled spending in two years. That is the setup for a digestion year, and the hyperscalers funding the demand are starting to show strain, with at least one now burning free cash flow and adding equity-funding options. Management's "visibility through 2028" is a demand-side claim that cannot be checked.

Our take: The bear case is right that 2028 cannot be evidenced from here, and it is the reason we keep position sizing moderate. But the market is pricing that risk aggressively: at 19.6x our FY28 estimate, the shares already discount a meaningful slowdown after 2027. A constrained 2027 with a demand cushion is more likely to roll into a slower 2028 than to fall off a cliff.

Debate: How much of the September margin record was China?

Bull view: Gross margin is guided higher for December while China falls sharply, so the record is not dependent on China. Operating margin crossed 40% on opex discipline that does not depend on any region, and new products like Akara and Aether carry structurally higher margins.

Bear view: Management credited customer mix, and customer mix at Lam has meant China. China rose 900bps and gross margin rose only 40bps, so on a mix-adjusted basis margins may have stalled. The December guide is a guide, not a result, and pricing remains unquantified at a time when supply constraints are giving every equipment company pricing power that will not last.

Our take: The bears have the better reading of September and the bulls have the better reading of December. We treat 52% as the sustainable gross margin floor for FY27 and the mid-50s as unproven, which is what our model already assumed. A December print at 52.5% with China in the mid-20s would be the strongest evidence yet that the July framework is achievable.

Debate: Has the multiple reset gone far enough?

Bull view: The shares trade at 23.6x FY27 and 19.6x FY28 estimates that rose this quarter, after a 33.8% decline from the October high. Free cash flow is at a record, the balance sheet holds net cash, and capital return is accelerating. Published price targets after the print still sit well above the current price, even after cuts.

Bear view: Peak-cycle earnings deserve a trough multiple, and semiconductor equipment stocks have historically de-rated hardest when growth peaks, not when it ends. Several targets were cut by 10% or more after the print, and the sector-wide selling suggests institutional de-risking that may not be finished.

Our take: A 23.6x forward multiple is not a trough multiple, so the bear case has room to be right on price in the near term. But it is well below the 31.8x we recorded after the July print, and it was set in a quarter where estimates rose and two of our re-rating triggers were met. We cut our fair value multiple to reflect the sector reset, and the upside to the new range is larger than it was after July.

Model Update Needed

Valuation: At the October 22 close of $230.15 and roughly 1.253B diluted shares, market capitalization is about $288B. With $7.4B of cash against $3.7B of debt, enterprise value is near $285B. On our revised FY27 non-GAAP EPS midpoint of $9.75, the shares trade at 23.6x, against 27.9x the day before the print on our prior estimate and 31.8x after the July print. On the FY28 midpoint of $11.75, they trade at 19.6x.

Our 12-month fair value range: $290 to $340, cut from $320 to $370, representing 26.0% to 47.7% above the October 22 close. The range equals 24.7x to 28.9x our FY28 non-GAAP EPS midpoint, down from 28.2x to 32.6x in July. We lower the multiple because the whole equipment group de-rated, and because the debate has moved to 2028, where our evidence is thinner than for 2027. Earnings estimates rose, and the new range runs from the lowest published target after the print to just below their median. The four supports are: (a) a quantified 2027 WFE outlook above $200 billion that management calls constrained, (b) free cash flow conversion restored to above 100% with capital return accelerating, (c) estimates rising while the multiple fell about 15% in one session, and (d) margin guidance that rises in December despite an adverse China mix.

Thesis Scorecard Post-Earnings

Overall: Thesis strengthened, with a lower-quality quarter inside it. The single largest open item, a quantified 2027, closed above our estimate, and last quarter's newest bear point, cash conversion, reversed within one quarter. Against that, China's concentration risk returned in the September quarter's composition, and Lam's own supply chain became a constraint for the first time. Neither changes the multi-year argument, but both make individual quarters lumpier, and lumpiness is what the stock is being punished for.

Action: Maintaining Outperform and raising conviction, with position sizing moving to 4% to 6% from 3% to 5%. In July we said we would raise conviction on (a) a quantified calendar 2027 WFE framework, (b) free cash flow conversion recovering toward 80% or more, or (c) a named commitment on the new US logic customer. The first two were met in full this quarter; the third was not. None of our three downgrade triggers fired. The sequential guide still grows, gross margin rose rather than falling below 51%, and NAND growth looks additive rather than pulled forward. We would turn more cautious on (a) a March quarter that does not recover the slipped December shipments, (b) December China above 30% with gross margin below 52%, or (c) a January 2027 WFE update below $200 billion.