Lam Research makes the machines that carve and coat chips
Etch machines
carve patterns into the layers
Deposition machines
lay down layers a few atoms thick
Parts and service
keep installed machines running
A chip is hundreds of thin layers on a silicon wafer, each patterned. Lam sells the machines that lay down those layers and carve them, mostly to the few companies that make memory and the most advanced chips.
The company · 2 · What exactly does it sell?
Two-thirds new machines, one-third parts and service
Each $1 of sales, fiscal 2026
64%36%
New machinesParts, service, upgrades
New machines sell when chipmakers build or upgrade factories. Parts, service and upgrades for the machines already installed come every year and hold up better when building slows.
The company · 3 · How does it make money?
It keeps over a third of each sales dollar as profit
Of every $1 of sales in fiscal 2026
50.5 centsleft after the cost of making the machines
35 centsleft after research, offices and sales staff too
$23.2Btotal sales in the year
Research is the biggest cost after the factories: about 10 cents of each dollar, because every new chip generation needs new machines.
Is it a good business? · 4 · What gives it an edge?
Once its machine is approved, chipmakers keep it
One chip generation
how long an approved machine usually keeps its step
Top 10
in chipmakers' own survey of equipment suppliers' service
4 points
more of equipment spending its machines can serve, management says
Approving a machine for a factory step takes months of tests, so chipmakers rarely switch mid-generation. Lam leads the hardest carving: holes through hundreds of stacked memory layers.
Is it a good business? · 5 · What is the catch?
Few customers, China, and strong rivals
55%of fiscal 2026 sales from four customers
34%of fiscal 2026 sales from China, under US export rules
Applied Materialsand Tokyo Electron sell the same steps to the same buyers
Each new chip generation is a fresh contest for each step, and China is helping its own toolmakers grow.
Is it a good business? · 6 · What does its record show?
Sales and profit up fast, after a dip
Sales, $B
17.2
17.4
14.9
18.4
23.2
26.1
Jun 2022Jun 2023Jun 2024Jun 2025Jun 2026last 12 mo
Sales$14.9B → $26.1Bsales, fiscal 2024 to the past year
Profit29 → 37 centsoperating profit per $1, fiscal 2024 to the past year
Dip−14.5%sales in fiscal 2024, when memory makers cut back
Research and office costs grew slower than sales, so profit rose faster. That works in reverse when spending falls.
Is it a good business? · 7 · How much room is there to grow?
Chipmakers will spend $143.9B on equipment this year
$143.9Ball chip-factory equipment, 2026
$38.8Bof it for DRAM memory, up 39%
+21.8%spending forecast for 2027
Those are SEMI, the industry body's, forecasts. Nobody publishes how much goes to carving and coating, Lam's part, so its own market is not measured.
Is it a good business? · 8 · Who is in charge?
A long-serving team that hands back most of the cash
Since 2018Tim Archer, chief executive
Under 1%of the company owned by insiders
$25Bnew share repurchases approved in November
In fiscal 2026 Lam spent $3.85B buying back shares and $1.27B on dividends. Bosses' bonuses pay on the share of sales kept as profit.
The story right now · 9 · Why are chipmakers buying so many machines?
AI needs memory, and memory needs Lam's machines
+39%DRAM equipment spending, 2026
46%of Lam's machine sales to memory makers, September quarter
Over $200Bequipment spending management expects in 2027
AI servers need far more memory. Memory is now built as tall stacks of layers, and every extra layer needs more of the coating and carving Lam sells.
The story right now · 10 · Why does a record year not last?
Chipmakers buy in bursts, then pause
Chipmakers buildNew factories need machines.
Lam's sales jump+54% in a year.
Supply catches upSpending pauses.
Sales fall−14.5% in fiscal 2024.
So we value Lam on a normal year: 31 cents of each sales dollar as operating profit, its five-year average, not the 40 cents of the latest quarter.
The story right now · 11 · What changed lately?
Sales up 54% in a year, and a record share kept as profit
$8.22Bsales in the September quarter
40 centsoperating profit per $1, a record
$8.60Bexpected for the December quarter
−16%share fall on 5 November, on AI spending fears
Strong quarter
Supply limits held the December forecast back, and sales to China are expected to fall. Twice in a month the shares fell over a tenth after good news.
The story right now · 12 · What is everyone arguing about?
A long boom, or a peak?
Many investors think
Either a long AI boom worth paying for, or a peak to sell
We think
The boom is real, but today's profit level is a peak
We would be wrong if Lam keeps over 35 cents of each dollar as operating profit through a year when sales fall.
The price and our view · 13 · What is a share worth?
The price sits a little above our value
$199what we think a share is worth · the last close is of that
$199our value
Attractive below $159
Neutral
Unattractive above $238
$93.86gloomy: growth halves, less kept per $1
$199our estimate: experts' path, profit back to normal
$279sunny: growth beats the forecast by a quarter
Value per share in three futures. None is a prediction of the share price.
We estimate what a share is worth from the cash the business should earn over many years, at its five-year average profit per dollar. Our rule: buy only when the price is at least a fifth below our estimate.
The price and our view · 14 · How could we be wrong?
The cycle turning is the biggest risk
The cycle turnsBiggest
Sales fell 14.5% in 2024.
China fallsShowing now
Expected down this quarter.
Four big customersWatch
Over half of sales.
Its share of each stepUnknown
Only management measures it.
US export rules limit what Lam may sell in China, and new rules could cut off more customers.
The price and our view · 15 · How will we know?
January tests the December quarter
Sales of $8.2B or moreGood sign
2027 spending forecast above $200BGood sign
Under 40 cents of profit per $1Bad sign
JAN 202727
Second-quarter fiscal 2027 results, 27 January. Companies report every three months; each report is a test of the story.
Our view also changes on price alone: a fifth below our value we turn positive, a fifth above it negative.
The price and our view · 16 · So, what do we think?
A strong toolmaker, priced a little above a normal year
$199our value
Attractive below $159
Neutral
Unattractive above $238
NeutralMedium conviction
Lam leads the machines memory makers need, in a real boom. But today's profit is a peak, and on a normal year the price sits a little above our value of about $199 a share, so we wait.
↑ ↓ to move · C for contents
What it does
More about the company
Who it is
A chip is built by laying down hundreds of thin layers on a silicon wafer and carving patterns into them. Lam makes the machines for those two steps, deposition and etch, plus tools that clean the wafer between them. Its buyers are the few companies that make chips: memory makers like Micron, Samsung and SK hynix, and TSMC. A third of its sales are spare parts, service and upgrades for the machines it has already installed.
How it makes money
Lam earns money two ways. About two-thirds of sales are new machines, bought when a chipmaker builds a factory, adds capacity or moves to a new chip generation; this half rises and falls with chipmakers' spending. The other third is parts, service and upgrades for its installed machines, which grows as the fleet grows and holds up better in a downturn. It sells to few buyers: four customers took over half of fiscal 2026 sales.
Growth and profit
What growth costs
Growth has made Lam more profitable: research and overheads grew more slowly than sales, so they take a smaller share of each dollar than two years ago. What grows with the boom is people and factories: it added about 1,900 staff in a single quarter. In a downturn that works in reverse.
Its record
The numbers
Is it growing?Fast
Sales grew 33% over the twelve months to Sep 2026, faster than the 26% in the year to Jun 2026.
Is it profitable?Strong
Operating margin is 37%, up from 29% in the year to Jun 2024.
Does growth pay off?Pays off
Running costs take 14¢ of each sales dollar, about where it was (14¢) in the year to Jun 2022.
Does it earn good returns?Strong
Each dollar put into the business earns 52% a year after tax, up from 43% in the year to Jun 2022.
Is it making cash?Strong
It turned 25% of sales into free cash, $6.54B over the last twelve months.
Is the balance sheet safe?Net cash
It holds $7.44B of cash against $3.74B of debt, $3.70B more cash than it owes.
Are shares being diluted?Diluting
The share count rose 793% in 4.3 years, and pay in shares costs 1.5% of sales.
Why it wins
The evidence
Approving a machine for a step in a chip factory takes a chipmaker months of testing, so once Lam's tool is chosen for a step it usually stays for that whole generation of chips. Lam is strongest in the hardest etching, cutting holes through hundreds of stacked layers in memory chips.
Customers rate it well: in an outside survey of chipmakers it ranked in the top ten equipment suppliers for service. That keeps steps it has won, but each new generation is a fresh contest.
Who could beat it
The rivals in full
Applied Materials, the largest equipment maker, sells the same deposition and etch steps to the same customers, and Tokyo Electron competes in etch and cleaning. Samsung buys some cleaning tools from its own affiliate. In China, government support is helping local makers of tools and spare parts, while US export rules limit what Lam may sell there.
a chipmaker qualifies Applied's tool for a step at the next node instead of Lam's
Tokyo Electron etch, deposition and clean8035.T
a chipmaker picks Tokyo Electron for etch or clean steps
ASM International atomic layer depositionASM.AS
a chipmaker uses ASM for thin-film deposition steps Lam also sells
Semes, Samsung's equipment affiliate005930.KS
Samsung buys clean tools from its own affiliate
Third-party spare-parts suppliers and Chinese domestic parts makersnot listed
a chipmaker buys spares for its Lam tools from someone else
How big it could get
The market and who has signed
The market
SEMI, the industry association, expects chipmakers to spend $143.9B on factory equipment in 2026, 23% more than in 2025, and another 22% more in 2027. Memory is growing fastest: equipment for DRAM, the working memory AI servers need, is forecast up 39% in 2026.
Nobody publishes how much of that is etch and deposition, Lam's part of it, so the size of Lam's own market is not measured.
Who runs it
Leaders, pay and the cash
Who runs it
Tim Archer has been chief executive since 2018, after 18 years at Novellus, which Lam bought in 2012. Doug Bettinger has been finance chief since 2013. Directors and officers own under 1% of the shares. Their bonus pays on profit margin and market share, and most share awards pay on how the stock does against other chip companies.
What they did with the money
Lam hands most of its cash back. In fiscal 2026 it spent $3.85B buying back shares and $1.27B on dividends, and repaid $750M of debt. In November it approved a further $25B of buybacks, and it aims to return 85% of free cash flow over time.
The last report
The latest report
Lam's quarterly sales rose 54% on a year earlier to a record $8.22B
What it did to the thesis Strong quarter
It showed the memory spending boom real and Lam's margin still climbing, which supports the business case. It did not show the margin lasting, which is what the price needs: a record quarter in a boom cannot say what a normal year earns.
The report
In the three months to 27 September 2026, the first quarter of fiscal 2027, sales rose 22% from the quarter before to $8.22B and operating profit reached 40% of sales, both records. Memory makers bought nearly half of the machines, up from a third a year earlier.
Management expects about $8.60B of sales in the December quarter, held back by supply limits, with China sales falling. The shares fell 12% the next day on fears the spending boom is near its peak.
What's happening now
The background
The shares are about half their June high. They fell 12% the day after a record quarter and 16% on 5 November, on worries that AI chip spending is peaking, then rose 8% after the new buyback.
Management expects equipment spending above $200B in 2027. The next test is December quarter results on 27 January 2027, with the first firm 2027 outlook.
Where we disagree
Both sides in full
The market
A long AI and memory equipment boom with Lam gaining share, worth paying for; or a peak, worth selling at the first sign.
We believe
The boom is real and Lam is strong in it, but today's margin is a peak; on a normal year the price is about fair.
The argument
The market swings between two views of the same boom: analysts expect years more growth, and the shares drop on any hint of a peak. We agree the boom is real. Our question is what a normal year earns.
Analysts' forecasts imply a long-run margin above any year Lam has filed. Against us: management aims for a higher gross margin, and if today's margin holds, the business is worth more than the price.
What it's worth
How we value it
Valued on the analysts' revenue path for three years, each later year closing 30% of the gap to 4%, on GAAP operating profit with stock pay a cost, the margin moved to its five-year average because today's is a cycle peak, and the tax management guides.
What the price assumes
Sales growth in the first year, then slowing as ours does, at the same profit per sale.
The price assumed
34.6%
Our estimate
42%
Measured at the close of
13 Nov 2026
The three cases
Bear case$93.86
Growth runs at half the analysts' path, 20.8%, 9.4% and 6.6%, as a WFE pause arrives early, the margin settles a fifth below the five-year average at 25.1%, and new capital earns only the 8% rate: the downturn test.
Our value$199
Revenue follows the analysts' path (C37), 41.6%, 18.8% and 13.1%, then fades toward 4%, closing 30% of the gap each year; the margin moves from today's 37.0% to its five-year average of 31.4% (C31), because today's margin is a cycle peak (C41); tax is the 16.5% management guides (C21); new capital earns 12%.
Bull case$279
Growth runs a quarter above the analysts' path, 52%, 23.5% and 16.4%, as WFE passes $200 billion with Lam gaining share, the margin settles a tenth above the average at 34.5%, and new capital earns 15%.
What could go wrong
Every risk
The risks
The cycle turns
Chipmakers cut spending after a boom; in fiscal 2024 Lam's sales fell 15% and its margin dropped.
Few, big customers
Four customers took over half of fiscal 2026 sales, so one pausing a factory moves a year.
China and export rules
China is a third of sales, US rules restrict what Lam can sell there, and local rivals are growing.
Rivals at each generation
Applied Materials and Tokyo Electron sell the same steps, and each new chip generation is a new contest.
What we still do not know
How much of etch and deposition Lam wins each year: only management's word says it is gaining.
How much its newer products, like its dry photoresist, sell today.
Which customer is which among the four that took over half of sales.
What changes our view
Every route to a new view
Every route to a new view
→ Attractive
The share price falls to about 80% of what we think the business is worth.
→ Unattractive
The share price rises to about 120% of what we think the business is worth.
→ Attractive
Operating profit stays above 35% of sales through a whole year in which sales fall.
The next test
It supports the case if
Revenue of $8.20B or more, the bottom of the guide, operating margin of 40% or more, and a 2027 WFE outlook of $200 billion or more.
It pressures the case if
Revenue below $8.20B, operating margin below 40%, or a 2027 WFE outlook cut below $200 billion.
We would also be wrong if
Lam's tool sales grow slower than the whole equipment market for a year.
A big customer moves an etch or deposition step to a rival at a new chip generation.
New export rules cut Lam off from Chinese customers worth over a tenth of sales.
The verdict
Scores and how we weighed it
The scores
Growth
3 of 4
Business quality
3 of 4
Financial position
4 of 4
Execution record
4 of 4
Per share capture
3 of 4
Potential
3 of 4
Risk profile
2 of 4
Valuation
3 of 4
Evidence quality
3 of 4
How we weighed it
ValueMixed
The price sits a little above our normal-year value, inside the band, with the bear case far below.
BusinessPositive
Chipmakers keep a qualified tool for the life of a node, and it leads memory etch.
GrowthMixed
Sales are surging, but the growth follows chipmakers' spending, which has fallen before.
RisksNegative
A few customers take most sales, China is a large share under export rules, and the cycle turns.
RivalsMixed
Priced above its two filed rivals, at a higher margin.
The groupMixed
Chip equipment names move together on spending fears; most read Neutral or worse on our values.
Latest newsPositive
The latest report was a record and the outlook rose, with a large new buyback.
In plain words
Our estimate of value
We forecast how much cash the business can earn over the next ten years and beyond, then ask what that stream is worth today to an investor who wants an 8% yearly return. We do it three times: a gloomy future, an expected one and a sunny one. We call a share Attractive only when its price is at least 20% below the expected estimate, and Unattractive when it is 20% above.