Report contents
At a glance
- Business model: Lam Research sells plasma etch and thin-film deposition chambers, plus spares and service, to a short list of memory and leading-edge logic fabs. Revenue rose 26% to $23.23 billion in FY2026 as gate-all-around and taller 3D NAND added process steps; the company does not disclose content per fab, and dual-sourcing by Applied Materials and Tokyo Electron caps pricing.
- Financial quality: FY2026 net income of $7.27 billion converted to $5.86 billion of operating cash flow, 0.81x and the weakest since FY2022, as receivables outgrew revenue. Returns remain far above the cost of capital (65% ROE, roughly 45% on invested capital), and $7.44 billion of cash against $3.8 billion of debt at September 2026 leaves room to absorb a downturn.
Business model
Business model: each node adds steps Lam gets paid for
Lam Research builds the equipment that deposits films onto a wafer and etches them away. Plasma etch and thin-film deposition are the core products, with wafer cleaning alongside, sold to the fabs that pattern logic and memory chips. The customer list doing the paying is short: TSMC, Samsung, SK Hynix, Micron, Intel and China's memory makers, each buying out of a multiyear capital budget rather than a consumables line item. Lam's FY2026 10-K, filed August 7, 2026, reports revenue of $23.23 billion, up 26%, and a gross margin of 50.5%, up from 48.7% in FY2025. The June 2026 quarter was the fourth consecutive record.
Two reportable segments carry that revenue. Semiconductor Systems sells the tools; Customer Support Business Group sells spares, service and upgrades against tools already running in production. The business type is an oligopoly manufacturer, and the stage is an upcycle inside a mature industry. Those two facts explain most of what follows: the profit pool is deep in a boom, the order book is somebody else's capital budget, and the customer can stop calling.
Demand comes from process intensity, not just wafer starts
The end markets are memory and leading-edge logic, and they move on different clocks. Both are adding process steps per wafer. Gate-all-around transistors wrap the gate on more sides and need extra deposition and etch passes; taller 3D NAND stacks mean more layers to drill. Each added step is a chance to sell another chamber, so revenue per fab can rise even if the world starts the same number of wafers.
That mechanism is the reason equipment demand has outgrown chip units, but the evidence is indirect. Growth of 26% says the mechanism was working in FY2026. Lam does not disclose steps per wafer or content per fab, so the split between complexity gains and customers simply building capacity cannot be read off the filing. The tradeoff that does show up is volume against capital: a 26% ramp required inventory, cleanroom space and hiring ahead of revenue, and gross margin improved only 1.8 points on that growth. Incremental revenue is profitable, but a meaningful share of each dollar goes to bought-in components and direct labor.
Pricing power is real, with a ceiling set by dual-sourcing
Etch and deposition are a three-way market: Lam, Applied Materials and Tokyo Electron, with KLA adjacent in process control. Fabs deliberately qualify at least two suppliers on critical steps so they can shift volume when price or delivery disappoints. That caps how much of an added step's value Lam keeps even where it holds the largest share. The 50.5% gross margin is the best available read on the balance struck so far. A price concession lands in margin before it lands in revenue, so share gains that arrive alongside a softening margin are worth less than the headline suggests.
The recurring business is steadier than it is large
Customer Support Business Group revenue tracks the installed base and how hard customers run it. Spares consumption rises with wafer starts at fabs already built, so it bends less than systems orders when capital spending turns. It remains the smaller segment, and Lam publishes no quarterly chamber count, so the installed base has to be inferred from segment revenue. Total revenue fell in FY2024, which shows the support line cushions a downturn without offsetting one. Anyone treating it as an anchor through the next cycle is over-reading its size.
Orders, geography and the cash lag
Most revenue is billed to fabs in Korea, Taiwan, China and Japan, with the US and Europe smaller. China is the complication: shipments to Chinese memory and foundry customers depend on US export licenses that have been rewritten more than once, so part of the geographic mix is set by policy. Payment comes per tool, recognized when the customer accepts it at the fab, after shipment and installation. That timing is why cash trails bookings through a ramp; FY2026 operating cash flow was 0.81x net income, the weakest conversion since FY2022. The working-capital detail belongs with financial quality, but the lag explains why order momentum and cash rarely arrive in the same quarter.
What would overturn the main judgment
The judgment is that Lam's revenue outgrows wafer starts because each node adds steps it can sell into. If gate-all-around and 3D NAND transitions stop adding deposition and etch intensity, or if Lam loses sockets at those transitions to Applied Materials or Tokyo Electron, the content story collapses into a plain volume cycle and FY2026 becomes a peak. Tool selection at each new node is the check, along with whether systems revenue keeps outrunning industry equipment spending over a full year.
Questions worth following
- Can Lam defend its share of etch and deposition spending as customers move to new device architectures?
Share won at each node tends to carry forward, because qualified process tools are reused across generations. Losing a step to a competitor lowers systems revenue now and shrinks the service base attached to that installed tool later.
- Will Lam's installed-base service and spares demand hold up if fab utilization drops?
Support revenue is tied to tools already in fabs, so it cushions a downturn only if customers keep maintaining equipment at lower utilization. The answer determines how much of Lam's revenue behaves like an annuity rather than a pure equipment cycle.
- Can Lam hold tool pricing as fabs qualify second suppliers?
Fabs routinely qualify at least two vendors on critical etch and deposition steps so they can shift volume. If that discipline tightens, added process steps would lift unit volumes without lifting price, and the 50.5% gross margin would look like cycle mix instead of durable pricing power.
Financial quality
Financial quality: strong returns, one soft spot in cash
Lam Research earned a 65% return on equity in fiscal 2026 and, on an estimate built from its reported figures, about 45% on invested capital with cash left in the capital base and near 68% with cash netted out. Those are exceptional numbers for a capital equipment maker, and they are not borrowed. Interest-bearing debt stood at $3.8 billion at both June and September 2026, only 14–16% of assets, and interest expense ran $160 million for the whole of FY2026. The leverage behind the ROE is mostly operational, from payables and accrued liabilities, plus a book equity that buybacks keep modest. The blemish sits in cash: FY2026 operating cash flow covered 0.81 of net income, the weakest since FY2022.
A small capital base does most of the work
Capex was $0.97 billion in FY2026, or 4.2% of revenue, and $0.26 billion in the September 2026 quarter. Growth has not demanded proportional plant. Asset turnover recovered to 1.04 in FY2026 from 0.79 two years earlier, and return on assets reached 32.4%, the best of the five years shown. The cycle test is informative: in FY2024 revenue fell 14.5% and ROE still came in at 45.7%, well above any reasonable hurdle. Equity compounded from $6.3 billion at June 2022 to $13.9 billion at September 2026 after four years of buybacks, so the high ROE does not rest on a shrinking denominator alone.
| USD bn unless noted | FY2022 | FY2023 | FY2024 | FY2025 | FY2026 | Q1 FY2027 (Sep 2026) | Assessment |
|---|---|---|---|---|---|---|---|
| Revenue | 17.23 | 17.43 | 14.91 | 18.44 | 23.23 | 8.22 | Strong, cyclical |
| Gross margin % | 45.7 | 44.6 | 47.3 | 48.7 | 50.5 | 52.3 | Strong |
| Operating margin % | 31.2 | 29.7 | 28.6 | 32.0 | 35.3 | 39.9 | Strong |
| Net margin % | 26.7 | 25.9 | 25.7 | 29.1 | 31.3 | 33.5 | Exceptional |
| ROE % | 74.8 | 62.3 | 45.7 | 58.2 | 65.1 | 20.9 | Exceptional; buybacks shrink equity |
| ROA % | 27.8 | 25.1 | 20.4 | 26.7 | 32.4 | 11.0 | Exceptional |
| Operating cash flow | 3.10 | 5.18 | 4.65 | 6.17 | 5.86 | 3.50 | Adequate |
| Free cash flow | 2.55 | 4.68 | 4.26 | 5.41 | 4.89 | 3.24 | Strong |
| OCF / net income (x) | 0.67 | 1.15 | 1.21 | 1.15 | 0.81 | 1.27 | Adequate |
| Capex / revenue % | 3.2 | 2.9 | 2.6 | 4.1 | 4.2 | 3.2 | Asset-light |
| Liabilities / assets % | 63.5 | 56.3 | 54.5 | 53.8 | 47.0 | 47.2 | Adequate; debt only 14–16% |
| Current ratio (x) | 2.69 | 3.16 | 2.97 | 2.21 | 2.63 | 2.48 | Strong |
| Days sales outstanding (approx.) | 78 | 75 | 65 | 58 | 68 | 63 | Adequate |
Quarterly ROE, ROA and cash flow figures are single-quarter amounts, not annualized. Days sales outstanding is derived from the reported receivables turnover, which uses average balances.
Margins look structural, with a cyclical top
Gross margin climbed from 45.7% in FY2022 to 50.5% in FY2026 and 52.3% in the September 2026 quarter, which Morningstar called a record on a non-GAAP basis in its October 22 note. The sharper test came in FY2024, when revenue fell 14.5% and gross margin still gained 2.7 points. A margin that expands through a revenue decline is hard to explain with volume absorption, which points to mix and to more deposition and etch steps per wafer. Operating margin reached 39.9% in the September quarter. Tax flattered FY2026: a 12% effective rate lifted net margin to 31.3%. That rate rose to roughly 17% in the September quarter and net margin still reached 33.5%, so operating leverage is now doing the work.
The cash gap narrowed, for one quarter
Receivables rose 58% to $5.34 billion in FY2026 against 26% revenue growth, and days sales outstanding went from 58 to 68. Inventory was not the culprit, holding flat at $4.28 billion while turns improved to 2.68. In the September 2026 quarter operating cash flow jumped to $3.50 billion on $2.75 billion of net income, while receivables grew 6% on 22% higher revenue. That fits working-capital timing around a steep ramp. Two more quarters at or above 1.0 would settle it.
| Quarter ended | Revenue | Net income | Total assets | Total liabilities | Receivables | OCF | FCF | Gross margin % | Net margin % | Assessment | Trend |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Sep 2025 | 5.32 | 1.57 | 21.90 | 11.71 | 3.63 | 1.78 | 1.59 | 50.4 | 29.5 | Strong | Stable |
| Dec 2025 | 5.35 | 1.59 | 21.39 | 11.25 | 3.49 | 1.48 | 1.29 | 49.6 | 29.8 | Strong | Cash lagging earnings |
| Mar 2026 | 5.84 | 1.82 | 20.79 | 10.21 | 4.13 | 1.14 | 0.81 | 49.8 | 31.2 | Strong margins, weak cash | Cash gap widens |
| Jun 2026 | 6.72 | 2.28 | 23.53 | 11.06 | 5.34 | 1.46 | 1.27 | 51.7 | 33.9 | Strong | Receivables build |
| Sep 2026 | 8.22 | 2.75 | 26.35 | 12.45 | 5.66 | 3.50 | 3.24 | 52.3 | 33.5 | Exceptional | Cash conversion recovers |
All figures are USD bn and reflect single quarters; Lam's fiscal year ends in late June, so the September 2026 period opens FY2027.
Returns to shareholders now exceed free cash flow
Financing outflows were $5.72 billion in FY2026 against $4.89 billion of free cash flow. The prior three years were covered. Cash fell to $4.75 billion at March 2026 before rebounding to $7.44 billion at September, and a current ratio of 2.48 with a quick ratio of 1.86 shows inventory is not propping up liquidity. The constraint is price discipline on the buyback, since each retired share costs more at today's multiple. The quarterly dividend rose to $0.33 from $0.26.
What to check
- The split between billed and unbilled receivables in the FY2026 10-K. Sixty-eight days is not alarming for tools sold on acceptance terms, but a shift toward unbilled balances would mean FY2026 earnings ran ahead of the cash they produced.
- Operating cash flow to net income in the December 2026 and March 2027 quarters. Guidance of $8.6 billion ± $400 million implies another sequential revenue increase, and rising revenue normally consumes working capital first.
- Buyback volume relative to free cash flow in the next order slowdown. FY2026 was the first year since FY2022 that distributions outran free cash flow; repeating that in a down year would carry a different meaning than repeating it in an up year.
Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.
Continue reading the full report
- Key risksWhat could break the investment case
- Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
- SummaryOur overall judgement and what to watch next
Free to sign up · 20 research points · unlocking this report uses 10
20 points cover 2 full reports. No card needed.
Research reflects its generation date. Preview the first two modules.
More on Lam Research Corp
Lam Research Corp
Business, moat and key questions
Open company page