ASML: Moat Analysis

30 August 2026. First assessment. Written against ASML's Q2 2026 results of 15 July, the High-NA EUV ramp as it stands in mid-2026, and the China export-control regime cutting into 2026 sales.

This document asks one question: how strong is the moat, and is it still intact. It contains no price, no position and no target. ASML is the sole maker of the machines that print every advanced chip on earth, with record demand from artificial intelligence, and it also carries the most concentrated customer base and the sharpest geopolitical exposure of any company in this set. That combination is exactly why the moat has to be judged on its own, before any price is looked at.

0. Verdict

Field	Reading
Strength	Exceptional. A genuine technology monopoly with no substitute at the leading edge, built on thirty years and a supply chain no competitor or state has been able to replicate
Condition	Intact. The monopoly is complete, order intake is very strong, and the technology lead is widening with High-NA, not narrowing
Verdict	Exceptional / Intact / Level 3 demand. Declined at the third gate: the moat is exceptional and undamaged, but the demand rests on the node race continuing, which is durable but not guaranteed to return after a structural shift
Security	Moderate. No competitor can reach the mechanism, but the customer base is extremely concentrated and a government can shrink the addressable market by decree
Pricing authority	Unconstrained. There is no alternative below the leading node, and prices per system rise with each generation
Demand anchoring	Level 3. The demand ASML ultimately serves is the continuation of the node race, the drive to ever-smaller transistors. That has a real replacement cycle and is probably durable, but it is not anchored in a biological necessity or a permanent human drive; it rests on a specific technological trajectory that could shift (see What cannot be seen, and A2). Durable, but without the guarantee that level-1 or level-2 demand carries
Class A conditions	0 of 6 triggered. Export controls have narrowed the market but not the monopoly; that is a market-size effect, not a moat breach
Class B gauges	0 of 6 triggered. Sales, margin and order intake are at or above guidance, and the installed-base annuity is growing
Decision	Not eligible for a Layer 2 entry, and this is a case that shows the demand gate doing its own work. The moat is exceptional and intact, yet the name is still declined, on the third gate alone. The demand rests on the continuation of the node race, which is level-3: durable, with a real replacement cycle, but not anchored in biology or a permanent human drive, so a drawdown cannot be assumed to reverse the way a level-1 or level-2 name's would. A perfect moat on demand that could structurally plateau is not a drawdown that can be waited out on faith

In one line: the one company without which no advanced chip can be made, holding a monopoly that thirty years and ten billion dollars could not reproduce, whose real risks are not to the moat but to the size of the market it is allowed to serve.

Why the verdict reads this way. ASML's Q2 2026 numbers are exceptional: net sales of €9.3 billion above the top of guidance, a 54% gross margin, order intake described as very strong, and full-year guidance raised to €43 to €45 billion on AI-driven demand. On the numbers this is a monopoly compounding at the centre of the largest capital-spending wave in the industry's history. And unlike most names in this set, the numbers and the mechanism agree: the moat is not just producing good output, it is genuinely intact and widening, because High-NA extends the monopoly to the next nodes rather than exposing it. The reason condition is clean rather than qualified is that ASML's threats do not touch the mechanism. Export controls shrink the market ASML may sell into; customer concentration makes its revenue lumpy and dependent on a few buyers' capex. Neither of those breaks the monopoly. They are risks to the size and steadiness of the prize, which is why they sit in the register as market and revenue risks rather than as moat conditions.

How strength and condition are judged is in the annex.

1. What the company does

ASML makes the machines that print the patterns on advanced computer chips, and for the most advanced chips it is the only company that can.

Making a chip means drawing circuit patterns onto silicon, and the finest patterns require light of an extremely short wavelength focused with extraordinary precision. ASML's extreme ultraviolet, or EUV, machines do this at a level no other machine can reach, and every chip at the 7-nanometre node and below is made with them. Below the leading edge, ASML also dominates the prior generation of deep ultraviolet, or DUV, machines. Its business has two halves: selling the systems, which each cost from roughly €180 million for a standard EUV machine to over €380 million for the new High-NA machines, and servicing and upgrading the thousands of machines already installed, a growing annuity called Installed Base Management.

How the money is actually made

ASML sells a small number of extraordinarily expensive machines to a small number of customers, and then earns a growing stream of service and upgrade revenue on every machine it has ever installed.

A single EUV machine is one of the most complex objects ever manufactured, assembled from hundreds of thousands of parts, and there is no alternative source. A chipmaker that wants to build leading-edge chips must buy from ASML, at ASML's price, because no one else makes the tool. That is the source of the pricing power: each generation of machine costs more than the last, and customers pay because the alternative is not a cheaper machine but no leading-edge chips at all.

The second half, the installed base, matters more than it first appears. Every machine ASML sells generates years of service, spare parts and upgrade revenue, and this annuity grows as the installed fleet grows. In Q2 2026 it was €2.8 billion, ahead of guidance, and it is the steadier, higher-margin counterweight to the lumpy timing of system sales. The more machines in the field, the larger and more predictable this stream becomes.

Why the monopoly is the whole story

ASML holds 100% of the EUV market and roughly 94% of lithography overall. This is not a leading share; it is the entire market at the leading edge. No NVIDIA GPU, no Apple processor, no AI accelerator exists without an ASML machine somewhere upstream.

The monopoly is why demand timing does not threaten the position. When chipmakers pause capital spending in a downcycle, ASML's sales fall, but no customer goes anywhere, because there is nowhere to go. The orders return when spending returns. And the current cycle is the opposite of a pause: AI demand is driving a surge in advanced-logic and high-bandwidth-memory capacity, both of which use more EUV layers per chip, which is why order intake is very strong and guidance is rising.

Where the money came from in Q2 2026

Total net sales were €9.3 billion, above the high end of guidance. Net system sales were €6.6 billion, of which €3.8 billion was EUV, including one High-NA system, and €2.8 billion was non-EUV. System sales split almost evenly between logic at 51% and memory at 49%. Installed Base Management was €2.8 billion, about €300 million above guidance. Gross margin was 54%, operating margin 37.1%, and net income €2.9 billion, or 31.3% of sales. Full-year guidance was raised to €43 to €45 billion, with EUV system sales expected to grow over 45% year on year. China is expected to fall to around 20% of sales in 2026 from 33% in 2025 under export controls.

2. The moat

ASML's moat is a technology monopoly with reinforcing layers around it, and the honest work is being precise about why it is one of the deepest in existence, and about where its risks actually sit, which is not in the moat.

Layer	Mechanism	Why it works
Foundational	The EUV technology monopoly	The only machines that can print advanced chips, with no substitute below 7nm
Foundational	The irreproducible supply chain	Zeiss optics, Cymer light sources, and thirty years of integration no rival or state can copy
Reinforcing	Customer roadmap lock-in	Chipmakers design their entire process around ASML's tools and timing
Reinforcing	The installed-base annuity	Every machine sold becomes years of high-margin service and upgrades
Reinforcing	The reinvestment moat	Over €4 billion a year of R&D keeps extending the lead faster than anyone can close it
Reinforcing	High-NA, the next-node extension	The new machines carry the monopoly forward to 2nm and below
Optional	DUV and mature-node systems	A real business, more contested, not where the leading-edge moat lives

Foundational: the EUV technology monopoly

This is the moat, and it is as close to absolute as a technology position gets. To make a chip at the leading edge, you need EUV lithography. To get EUV lithography, you buy from ASML, because ASML is the only company that makes it. There is no second source, no alternative technology at the leading node, and no way to build advanced chips without it.

The depth of this comes from what it took to build. ASML bet on EUV in the late 1990s and 2000s, when Nikon and Canon looked at the astronomical cost and technical difficulty of using 13.5-nanometre light and walked away. ASML persevered for two decades, with over nine billion dollars of R&D and strategic investment from its own biggest customers, and emerged as the only company that solved it. That decade-long gamble did not just win a market; it ended the lithography wars, leaving ASML alone at the leading edge.

Three things follow, and the third is the one that shapes this file.

The monopoly produces unconstrained pricing power, because there is no alternative below the leading node. Each generation of machine costs more, and customers pay, because the choice is not a cheaper tool but no advanced chips.

The monopoly is extraordinarily durable against competitors, because the barrier is not a patent that expires but an accumulation of capital, tacit knowledge and a unique supply chain that took thirty years to build. A state with unlimited resources and national priority, China's SMEE, remains ten to twenty years behind and cannot close the gap by spending.

And the monopoly's risks lie almost entirely outside the mechanism. No competitor can take the position, so the things that can hurt ASML are not moat threats at all: how many machines its handful of customers choose to buy, and how much of the world it is legally allowed to sell into. That is an unusual shape, and the register below is built around it.

Foundational: the irreproducible supply chain

The monopoly is not just ASML's own engineering; it is a supply chain no one else can assemble. The optics come from Zeiss, mirrors polished to a smoothness measured in fractions of a nanometre, the flattest surfaces humans manufacture. The light sources come from Cymer, which ASML bought. The integration of these into a working machine is protected by thousands of patents and, more importantly, by decades of accumulated know-how that is not written down anywhere and cannot be reverse-engineered from a finished machine.

This is why the monopoly is not a temporary technology lead that a well-funded rival could catch. A competitor would have to reproduce not just the machine but the entire supply chain and the tacit knowledge behind it, from a standing start, while ASML spends over four billion a year extending the lead. The barrier is cumulative and it widens, which is the opposite of a patent moat that erodes as protection expires.

The asymmetry that defines it, and where the risk actually sits

The strongest moats are the ones where the only party who can damage the mechanism is the company itself. ASML's moat is close to that on the competitive dimension: no rival and no state can reach the mechanism, and the company would have to damage its own technology lead to lose it, which nothing suggests.

But ASML has an unusual feature that most exceptional moats do not: its deepest risks sit entirely outside the moat, in two places. The first is the customer. ASML sells to a handful of buyers, and the top two, TSMC and Samsung, are around 38% of revenue, with TSMC alone near 24%. That concentration is not a moat weakness, because those customers cannot switch, but it is a revenue risk, because ASML's sales depend on a few companies' capital-spending decisions, and the actor who decides how many machines to buy sits at the customer, not at ASML. The second is the government. Export controls can shrink the market ASML is allowed to serve, cutting China from 33% to 20% of sales, and that is a decision taken by regulators, entirely outside both ASML and its customers.

Security therefore reads moderate rather than high, but for a reason specific to ASML: not because the mechanism is exposed, it is not, but because the size and steadiness of the prize depend on actors outside the company, the customers and the governments. The moat is intact and unreachable. What is exposed is how much of the world it is allowed and asked to serve.

Reinforcing: customer roadmap lock-in

Chipmakers do not just buy ASML machines; they design their entire manufacturing process and product roadmap around ASML's tools and delivery timing. A fab's node schedule is tied to when it can get ASML machines, and the machines are integrated into workflows that took years to build. This deepens the monopoly, because even in the hypothetical case of an alternative appearing, a customer could not simply switch: their whole process is built around ASML. The High-NA milestone at Intel, qualifying the new machine on production layers, is an example of this lock-in deepening, as the customer's most advanced node becomes dependent on ASML's newest tool.

Reinforcing: the installed-base annuity

Every machine ASML has ever sold generates ongoing service, spare-parts and upgrade revenue, and this stream grows with the installed fleet. At €2.8 billion in the quarter and above guidance, it is the steadier, higher-margin counterweight to lumpy system sales, and it is almost entirely locked in, because a customer running an ASML machine has no one else to service it. This turns the one-time sale into a decades-long relationship and makes the revenue base more durable than system-sales timing alone would suggest.

Reinforcing: the reinvestment moat

ASML spends over four billion a year on R&D, funded by monopoly profits, and that spending extends the lead faster than any competitor could close it. This is the mechanism that makes the moat widen rather than erode: the profits from being the only supplier fund the research that keeps ASML the only supplier. It is a self-reinforcing loop that a smaller competitor, without the monopoly profits, structurally cannot match, and it is why thirty years in, the gap to the nearest rival is measured in decades rather than years.

Reinforcing: High-NA, the next-node extension

The new High-NA EUV machines, at over €380 million each, extend the monopoly to the 2nm node and below. This matters for the moat because it answers the one way a technology monopoly could erode, by the technology itself reaching a limit. Instead of hitting a wall, ASML has the next-generation tool ready and qualifying at customers, which carries the monopoly forward rather than leaving an opening. High-NA is the proof that the lead is widening: it is a harder machine that concentrates the customer base even further, because only TSMC, Samsung and Intel can afford and adopt it.

The honest qualification is that High-NA adoption is not guaranteed to be fast. If the cost-benefit of High-NA does not satisfy customers, they may stretch standard EUV with double-patterning instead, slowing the adoption of ASML's most expensive machines. That is a risk to the pace of High-NA revenue, not to the monopoly, since the double-patterning alternative also runs on ASML machines.

Optional: DUV and mature-node systems

ASML's deep ultraviolet and mature-node business is real and large, but it is more contested: Nikon and Canon still make legacy systems for mature nodes, and Canon is pushing nanoimprint lithography as a lower-cost alternative for some memory applications. This is placed as optional to the leading-edge moat, because the monopoly and the pricing power live in EUV, not in the more competitive mature-node tools. It is recorded so that any share pressure in mature-node DUV is never read as evidence about the EUV monopoly, which is a separate and far stronger position.

Evidence of strength: the competitive record

The strength of the moat is shown by what happened to everyone who tried to compete, and the record is one-sided. Nikon and Canon, once ASML's peers in lithography, looked at EUV and walked away, and now make only legacy systems for mature nodes. China's SMEE, backed by a state treating semiconductor independence as a national priority with effectively unlimited funding, remains ten to twenty years behind and cannot close the gap. Canon's nanoimprint and various research alternatives are, on the most optimistic estimates, five to ten years from commercial viability and target niche applications rather than the advanced logic where the monopoly lives.

The honest counterweight is that a monopoly this complete invites two things: substitution research and political intervention. The substitution research has been under way for years and has produced nothing that threatens the leading edge, which is strong evidence for the moat. The political intervention is real and is the actual risk, but it works by restricting ASML's market, not by creating a competitor, which is why it sits in the register as a market risk rather than a moat breach.

Evidence of strength: pricing power

ASML's pricing power is as clean as it exists. Each generation of machine costs more than the last, from roughly €180 million for standard EUV to over €380 million for High-NA, and customers pay because there is no alternative below the leading node. There is no regulator setting the price, no competitor undercutting it, and no customer with the leverage to refuse, because refusing means not building advanced chips. The pricing power is unconstrained in the strict sense: the only limit is the customer's own economics, not any alternative.

The one qualification is that pricing power over the leading-edge machine is not the same as unlimited pricing over the whole business. In mature-node DUV, where Nikon and Canon compete, pricing is more disciplined. But in EUV, where the monopoly and most of the profit sit, the pricing power is total.

Evidence of strength: the financial fingerprint

The numbers are the fingerprint of a technology monopoly at full strength. A 54% gross margin and a 37% operating margin on machines this complex reflect a company that sets its own price. The installed-base annuity of €2.8 billion a quarter, growing and above guidance, shows the durability underneath the lumpy system sales. Order intake described as very strong, and full-year guidance raised to €43 to €45 billion, show demand running ahead of expectations on the AI wave. Free cash flow funds a €1.1 billion buyback in the quarter under a multi-year program, alongside a growing dividend.

One caution governs the read, and it is different from most names in this set. The numbers here genuinely do reflect the mechanism, because the monopoly is intact and the demand is real, so there is less gap between output and mechanism than usual. The caution is instead about lumpiness and market size: system-sales timing is uneven, the China restriction removes a chunk of the addressable market, and customer concentration means a single buyer's capex pause would show up sharply. The financial risk is to the size and steadiness of the revenue, not to the monopoly that produces it.

Alternative explanations

A moat claim is only worth anything if the competing explanations fit the data worse.

It is just riding the AI capex wave and will fall when it turns. The AI wave is real and is lifting demand, but it explains the current growth rate, not the moat. When the cycle turns, ASML's sales will fall and no customer will leave, because there is no alternative supplier; the monopoly is orthogonal to the cycle. Confusing the cycle with the moat would be the error, and the register separates them.

The monopoly is temporary and competitors will catch up. Thirty years of evidence says otherwise. Nikon and Canon exited, SMEE cannot close a decades-wide gap with unlimited state funding, and the reinvestment moat widens the lead faster than anyone closes it. A technology monopoly built on cumulative tacit knowledge and a unique supply chain is the opposite of a patent that expires.

Export controls are breaking the moat. This is the most important explanation to get right, and it is wrong in a specific way. Export controls are real and they hurt, but they shrink the market ASML may sell into; they do not create a competitor or give any customer an alternative. A smaller monopoly is still a monopoly. The controls are a market-size risk, correctly placed in the register as such, not a moat breach.

Customer concentration means the customers have the power. The concentration is real and is a revenue risk, but it does not transfer pricing power to the customers, because they cannot switch. TSMC needs ASML more than ASML needs any single order, since TSMC has no other source of leading-edge tools. The concentration makes ASML's revenue lumpy and dependent on a few capex budgets; it does not make the customers able to dictate terms.

The preferred explanation is that ASML's moat is a technology monopoly at the leading edge, built on thirty years and an irreproducible supply chain, widening through reinvestment and High-NA, with the unusual feature that its real risks, customer concentration and geopolitics, sit outside the moat mechanism entirely. That account fits the complete market share, the unconstrained pricing, the failed competitors, the intact demand, and the specific nature of the export-control and concentration risks all at once.

3. What could break it

3a. Who can break it

The register follows from one question: who takes the decision that damages this moat, and would their action be visible in the numbers?

Mechanism	Actor	Visible in the numbers?
The EUV monopoly	A competitor or state reaching the leading edge	Yes, but with a decade of warning, if ever
The addressable market	Governments, through export controls	Yes, immediately, as a market-size cut
Demand for machines	The handful of customers, through capex	Yes, quickly, as lumpy system sales
The next-node extension	A shift in chip architecture away from smaller nodes	Only late, as node-race demand fades
Mature-node share	Nikon, Canon, nanoimprint	Yes, but immaterial to the EUV moat

The unusual feature of this table is that the moat mechanism, the top row, is the one least likely to move and would give a decade of warning if it ever did, while the rows that can move quickly, export controls and customer capex, are not moat threats at all but risks to the size and timing of revenue. This is the reverse of most companies, where the fast-moving risks are the dangerous ones. For ASML, the dangerous mechanism is nearly immovable, and the movable risks are not to the mechanism. A soft quarter of system sales, or a China cut, is not evidence the moat is eroding; it is evidence that demand or the addressable market moved, with the monopoly fully intact.

The one genuine long-term moat risk, the fourth row, is the subtle one: not a competitor, but a change in how chips are made that reduces the importance of ever-smaller nodes. That is the only path by which ASML's monopoly could become less valuable without anyone catching up, and it is the one to hold in view over the long run.

3b. Class A: mechanism conditions

These describe events, each with an actor. A trigger here is a structural change to the moat, not a cyclical dip in chip capex. The reasoning behind the class split is in the annex.

#	Condition	Actor	Observable event	Reading now
A1	A competitor or state reaches leading-edge lithography	A rival, or a state program	A working EUV-class tool qualified at an advanced node from anyone but ASML	Clean. No competitor within a decade; SMEE far behind, Nikon and Canon exited
A2	Chip architecture shifts away from the node race, reducing EUV's centrality	Chip designers, the industry	A durable move to a paradigm, such as photonics or advanced packaging, that bypasses ever-smaller nodes	Clean, a slow watch. Advanced packaging complements rather than replaces; no bypass in sight
A3	High-NA fails to be adopted, capping the next-node extension	Customers	TSMC and Samsung declining High-NA in favour of stretching standard EUV	Clean. Intel qualifying High-NA on production layers; adoption pace is the only question
A4	Export controls tighten to the point of structurally shrinking the market	Governments	Controls extending beyond China to materially cut the addressable market further	Live and worsening at the margin. China cut to ~20%; a market-size risk, not a moat breach
A5	The irreproducible supply chain is disrupted	Zeiss, Cymer, or a supply shock	Loss of a unique supplier or a supply-chain break with no substitute	Clean. Supply chain integrated and owned or locked in
A6	Customer concentration turns into a demand collapse	The top customers	A simultaneous, sustained capex pause by TSMC and Samsung	Clean. Order intake very strong; AI capex at record levels

On A4, the condition that is live but is not a moat breach. A4 is the condition most in motion, and it is important to be precise about what it is. Export controls have already cut China from 33% to 20% of sales, and could tighten further. But this shrinks the market ASML is allowed to serve; it does not touch the monopoly. A smaller monopoly is still a monopoly, and the customers ASML loses to export controls do not go to a competitor, because there is none, they simply cannot buy at all. So A4 is written as a market-size condition, not a moat condition: it hurts revenue and it is the live geopolitical risk, but it cannot break the mechanism, and it is placed here to be watched as a market risk rather than mistaken for erosion.

On A2, the only true long-term moat risk. A2 is the subtle one and the one that matters most over a long horizon. ASML's monopoly is valuable because advanced chips need ever-smaller nodes, and ever-smaller nodes need EUV. If chipmaking shifts to a paradigm where node size matters less, advanced packaging, chiplets, photonic computing, or something not yet visible, the monopoly could become less valuable without any competitor catching up. Nothing on the horizon does this today; advanced packaging currently complements EUV rather than replacing it. But it is the one path by which the moat erodes without an actor reaching the mechanism, and it is carried as the deepest long-term watch.

Calibration. A1 is well understood and clean, with thirty years of failed competition behind it, and it would give a decade of warning. A4 is live and measurable, but it is explicitly a market-size condition rather than a moat breach. A2 has no precedent and no timeline, so it is reasoned rather than observed. A3, A5 and A6 are clean and well understood. The file does not claim precision on A2, which is genuinely unknowable in its timing.

3c. Class B: gauges

These are measurements. A number has two causes, the mechanism and the capex cycle, and the number alone cannot tell you which moved. A gauge never rejects the moat on its own; it obliges the attribution test in 3e.

#	Gauge	What it isolates	Expected direction if the moat erodes	Reading now
B1	EUV system average selling price	Pricing power at the monopoly	Falls, which would signal an alternative appearing	Clean. Rising with each generation, High-NA at a large premium
B2	Order intake and backlog	Forward demand and monopoly durability	Falls durably, beyond a cyclical dip	Clean. Order intake very strong through H1 2026
B3	Installed Base Management growth	The annuity's durability	Stalls or falls as the fleet stops growing	Clean. €2.8bn, above guidance
B4	Gross margin	Overall pricing and mix strength	Falls as pricing power weakens	Clean. 54%, above guidance, guided to rise
B5	China share and total addressable market	The export-control market risk	Falls as controls tighten	Watch. China to ~20%; a market-size read, not a moat read
B6	High-NA adoption rate	The next-node extension	Slow or stalled adoption caps future EUV revenue	Early. Intel qualifying; TSMC and Samsung pace the key variable

Why the China cut is not a moat signal in B5. China falling from 33% to 20% of sales is a real revenue headwind and it will show in the numbers, but it is close to meaningless as a read on the moat, because it is a market-size effect imposed by governments, not a sign that anything about ASML's monopoly weakened. B5 is in the register to track the geopolitical risk to revenue, explicitly separated from the moat, so that a China-driven sales dip is never misread as erosion. The monopoly over the market ASML is allowed to serve is exactly as complete after the cut as before it.

B6 is the forward gauge that matters most. High-NA adoption is where the next decade of EUV revenue growth sits, and it is the one gauge with real uncertainty: not whether High-NA works, Intel has qualified it, but how fast TSMC and Samsung adopt it versus stretching standard EUV. That paces future revenue, though not the monopoly, since both paths run on ASML machines.

3d. Comparator sets

Both sets are fixed here, in advance, so a later reading cannot pick the comparison that suits the conclusion.

Macro peers, which share the semiconductor-capex cycle and separate a spending slowdown from a mechanism problem: the broader semiconductor capital-equipment group, Applied Materials, KLA, Lam Research, and Tokyo Electron. If ASML's system sales soften while these soften too, it is the capex cycle; if ASML alone moved, it would be company-specific, which for a monopoly would be surprising and worth investigating.

Mechanism peers, which test the monopoly rather than the cycle: Nikon and Canon as the exited lithography rivals and the nanoimprint alternative, and SMEE as the state-backed effort. The instructive point is that these are not real competitors at the leading edge, which is the evidence for the moat, and the register watches them not because they threaten ASML today but because A1, a competitor reaching the leading edge, would first show up as progress at one of them. The one that is not a company, the A2 architecture shift, has no peer to watch and is tracked through the industry's technology direction instead.

3e. The attribution test

Run this every time a Class B gauge moves. All three must come back clean for the cause to be recorded as cyclical or market-driven rather than a moat problem.

1. Is there a nameable external cause with a date? For soft system sales, is it a capex-cycle pause or an export-control cut, or is it something touching the monopoly itself?

2. Do the macro peers move with it? If the semiconductor-equipment group softens together, it is the capex cycle. If ASML alone moves, investigate, because for a monopoly that is unusual.

3. Is the mechanism side unchanged? Is ASML still the only EUV supplier, is pricing still rising, is order intake still strong beneath the timing noise, or has something actually changed at the monopoly?

   
   

The escalation rule is not a fixed count of quarters. A cyclical or market attribution holds only while the named cause is present and verifiable. If sales fall and the cause is neither the capex cycle nor export controls but something at the monopoly, or if a competitor's leading-edge progress appears in the mechanism peers, the matter escalates to a Class A judgement regardless of the calendar.

One caution specific to ASML. The usual danger, a moat eroding while the numbers stay clean, is less acute here, because the moat is nearly immovable and would give long warning. The specific danger for ASML is the reverse: mistaking a market-size or cyclical dip for moat erosion. A China cut or a capex pause will make the numbers look worse while the monopoly is completely intact, and the discipline is to not read those as moat damage. The one thing that would be moat damage, A1 or A2, will not come from the quarterly sales line at all; it will come from the mechanism peers or the industry's technology direction, which is why those are watched directly.

How this document is revised. On any evidence of a competitor or state reaching leading-edge lithography, whatever the calendar. On any material change to export controls or the addressable market. On any shift in chip architecture that bears on the centrality of the node race. On the quarterly results. And on anything unforeseen where the question of whether to revise even arises. For this moat, the technology direction of the whole industry matters as much as ASML's own numbers, because the one true long-term risk lives there.

4. What cannot be seen

Two things carry real weight and have no clean, timely signal. Listing them stops "conditions mostly clean" from being read as "nothing to watch".

A paradigm shift away from the node race. ASML's monopoly is valuable because making better chips has meant making smaller transistors, and smaller transistors need EUV. If the industry finds a different path to better computing, photonic computing, radically different architectures, or a shift where advanced packaging and chiplets do more of the work than node shrinks, then ASML could remain the unchallenged master of a technology that matters less. This would erode the value of the moat without any competitor ever catching up, which is why it is the one risk that does not show up as a competitor and cannot be watched through the mechanism peers. It has no timeline and no clean signal, and it is the deepest long-term uncertainty in the file.

The geopolitical ceiling on the market. Export controls today cut China. But the deeper uncertainty is how far the politicisation of advanced chips goes: whether the addressable market keeps shrinking as more of the world is walled off, whether a customer country becomes a flashpoint, above all Taiwan, where the largest customer sits. This is not a moat risk, the monopoly survives any map, but it is a risk to the size and stability of the prize that no gauge captures in advance, because it is set by geopolitics rather than by the business.

Two structural limits are worth stating plainly. The architecture-shift risk, A2, has no precedent and no timeline, so it is reasoned rather than observed. And this is a moat whose mechanism is nearly unbreakable but whose value depends on a small number of customers and a permissive geopolitical environment, which is a different risk shape from a moat that is itself fragile.

5. Assumptions

#	Assumption	Status
1	Advanced chips keep requiring ever-smaller nodes, and nodes keep requiring EUV	High confidence near term, the deepest long-run uncertainty
2	No competitor or state reaches leading-edge lithography this decade	High confidence. Thirty years of evidence, decades-wide gap
3	High-NA is adopted by the leading foundries, extending the monopoly	High confidence on eventual adoption, lower on pace
4	The customer base keeps investing through the cycle	Moderate to high confidence, but concentrated in a few buyers
5	Export controls shrink but do not eliminate the addressable market	Moderate confidence. The direction of travel is tightening
6	No paradigm shift displaces the importance of the node race	Moderate confidence over the long run, unmeasurable in timing

6. Basis of this assessment

This is the first Layer 1 written on ASML, so there is no prior verdict to move from. It records the starting position that future revisions will read against.

The moat is judged a technology monopoly at the leading edge of chip manufacturing: ASML is the sole maker of EUV lithography, holding 100% of that market and roughly 94% of lithography overall, built on thirty years and over nine billion dollars of R&D and an irreproducible supply chain of Zeiss optics and Cymer light sources. Strength reads exceptional: there is no substitute below the 7nm node, no competitor within a decade despite unlimited state funding at SMEE, and the lead widens rather than erodes because monopoly profits fund over four billion a year of R&D and the High-NA machine carries the monopoly to the next nodes. This is one of the deepest moats in existence.

Condition reads intact, cleanly. The monopoly is complete, order intake is very strong on the AI wave, guidance was raised, the installed-base annuity is growing, and High-NA is qualifying at customers, so the mechanism is widening, not narrowing. No Class A condition is triggered. The one condition in motion, A4, export controls, is explicitly a market-size risk rather than a moat breach: it shrinks the market ASML may serve, cutting China to around 20% of sales, but it does not create a competitor or give any customer an alternative, so a smaller monopoly is still a monopoly.

The unusual feature of this name, stated plainly, is that its real risks sit outside the moat. Customer concentration, the top two buyers at 38% of revenue, makes the revenue lumpy and dependent on a few capex budgets, but it does not transfer pricing power to customers who cannot switch. Export controls and, deeper, the geopolitical status of Taiwan, threaten the size and stability of the market, not the monopoly. And the one true long-term moat risk, A2, is not a competitor at all but a possible shift in chipmaking away from the node race, which would devalue the monopoly without anyone catching up. None of these is visible as moat erosion in the numbers, which is why the discipline for this name is the reverse of most: not to mistake a market-size or cyclical dip for erosion of a monopoly that remains complete.

The verdict is Exceptional / Intact / Level 3 demand. It is one of the deepest and most durable moats in this or any set, intact and widening, whose distinctive feature is that the questions worth watching, customer concentration, geopolitics, and a far-off architecture shift, are risks to how much of the world the monopoly is asked and allowed to serve, not to the monopoly itself.

Future revisions are dated and appended below.

Annex: how this assessment is made

These are the rules the document is written under, kept separate so the file above stays about ASML and the rules cannot quietly change to suit a conclusion.

Judging strength

Strength is settled before condition, because the fields behind the condition axis only measure whether a moat is undamaged, not whether there was much of a moat to begin with. Three questions, all about how hard the moat is to attack rather than how well the company is trading. Is there a substitute the market could actually move to? Could a competitor replicate the position? Has it been attacked, and what happened?

ASML's answers are as clean as any in this set. There is no substitute below the leading node; a chipmaker that wants advanced chips has no alternative to EUV and no alternative to ASML. A competitor cannot replicate the position; thirty years, a unique supply chain, and tacit knowledge that cannot be reverse-engineered mean even a state with unlimited funding remains decades behind. And it has been attacked and won completely: Nikon and Canon tried and exited, SMEE tries and cannot close the gap, and substitution research remains years from the leading edge. That is exceptional strength on all three counts, and the reinvestment loop means it widens rather than holds.

Exceptional means all three answers come back clean and the strength is self-holding, funded by its own profits and deepening over time. ASML qualifies as clearly as any company can. The reason security nonetheless reads moderate rather than high is not a weakness in the moat but the concentration of customers and the geopolitical ceiling on the market, which are conditions on the value of the monopoly, not on its existence, and the verdict is built to keep that distinction visible.

The Layer 2 gate requires three things together: exceptional strength, intact condition, and demand anchored at level 1 or level 2. ASML passes the first two, and passes them more cleanly than almost any name, yet it is declined at the third: its demand is level 3. This shows the demand gate is not redundant with the moat gates. A moat can be flawless and undamaged and still fail to qualify, because whether the market returns after a fall is a separate question from how strong or intact the moat is, and for ASML that separate question is the binding one.

The verdict

The verdict is two judgements held apart, because collapsing them into a single grade hides the thing that matters most. One axis is strength: how deep was the moat to begin with, scored exceptional, strong, or ordinary. The other is condition: is it still whole, scored intact, impaired, or broken.

Keeping them separate is deliberate. ASML shows a third use of the split beyond the usual ones: a moat can be exceptional and intact on both axes while still carrying serious risks, if those risks sit outside the moat mechanism. ASML's customer concentration and geopolitical exposure are real and large, but they are risks to the size and steadiness of the revenue, not to the monopoly, so they do not lower either axis. They are captured as market and revenue risks in the register and in the security reading, rather than being allowed to muddy a condition score that is genuinely clean. Forcing them into the condition axis would misrepresent an intact monopoly as a damaged one.

Three questions feed the condition axis, answered in a fixed order so it cannot be talked into a softer reading afterwards. Can anyone outside the company reach the mechanism? That is security. Has anything altered the mechanism itself? That is condition, intact, impaired or broken. Does anyone else have a say in the price? That is pricing authority, unconstrained, constrained or contested.

Below all of it sits a rule the verdict cannot override. If the foundational layer is broken, the name is rejected whatever the axes would otherwise say. For ASML the foundational layer, the EUV monopoly, is intact and widening. Impaired would require a credible competitor reaching the leading edge, or a durable architecture shift devaluing the node race; broken would require ASML to no longer be the necessary supplier of advanced lithography. Neither is close.

Judging demand anchoring

Strength and condition together answer whether the company keeps its share of the market. They do not answer whether the market returns after a fall, and that is a separate question that decides whether a drawdown can be waited out at all. A moat can be fully intact while the demand beneath it shrinks permanently, so demand anchoring is judged on its own axis, on two tests: biology, is the underlying demand rooted in a permanent human or physical necessity or drive, and precedent, is there a history of similar demand collapsing and not returning.

The levels run from one to four. Level 1 is anchored in a physical or biological necessity that never stops, waste, water, illness, basic food. Level 2 is anchored in a permanent human drive that can fall cyclically but always returns, status, transacting, connection. Level 3 is anchored in an established infrastructure or habit with a replacement cycle, probably durable but with no biological guarantee, and exposed to a structural shift. Level 4 is anchored in a moment, a build-out or a specific medium that may never return to its peak. Only level 1 and level 2 are eligible for a Layer 2 entry, because only there does a drawdown reliably reverse.

ASML reads level 3, and it clearly separates the demand gate from the moat gates. The demand ASML serves is not a biological necessity and not a permanent human drive; it is the continuation of the semiconductor node race, the industry's push to smaller and smaller transistors, on which the value of a leading-edge lithography monopoly depends. That demand is durable in the way established infrastructure is durable: it has a real replacement cycle, machines wear out and each generation needs new tools, and it has run for decades. But it is not guaranteed the way level-1 or level-2 demand is, because it rests on a specific technological trajectory rather than on a permanent feature of people or biology. If the industry finds a different path to better computing, one where node size matters less, the demand under the monopoly could plateau permanently, and a drawdown then would not reliably reverse. That is what distinguishes level 3 from level 2: not that the demand is fragile, it is not, but that its permanence rests on a trajectory continuing rather than on a necessity that cannot stop.

It is worth marking the boundary with level 4 too, because ASML is not there. The chip demand is not a one-time build-out like the dotcom fibre glut; it has a genuine, recurring replacement cycle, which is why it is level 3 and not level 4. But a replacement cycle on a technological trajectory is a weaker guarantee than a biological or human-drive anchor, and under this framework only level 1 and level 2 clear the gate. So ASML, with a moat as deep as any in the set and a condition that is fully intact, is declined on the demand axis alone. That is the gate working as intended: an exceptional, undamaged moat is not a buy if its market may not return after a fall.

Why the conditions are split in two

A Class A condition describes something someone did, or a structural event with an actor. A Class B gauge is a number, and a number has two causes, so it cannot on its own tell you which moved.

For ASML the split does an unusual job: it separates the risks that look dangerous in the numbers from the risk that is actually dangerous to the moat. Export controls and capex pauses will move the Class B gauges sharply while the monopoly is completely intact, so a file scored on the gauges could wrongly read a market-size cut as erosion. The Class A conditions hold the line: the moat is only impaired if a competitor reaches the leading edge (A1) or the node race is bypassed (A2), neither of which shows up as a soft sales quarter. A Class A trigger is a structural verdict; a Class B move only ever obliges investigation, and for ASML that investigation usually concludes "market size or cycle, monopoly intact."

When a cyclical or market explanation expires

"It is the capex cycle" or "it is export controls" will be available whenever sales soften, and will usually be true, because both genuinely move ASML's revenue without touching the moat. The rule is that the attribution holds only while the named cause is present and verifiable. If sales fall and it is neither the cycle, the peers would move too, nor export controls, a datable government action, then the cause may be the mechanism and the matter escalates. For ASML, uniquely, the escalation trigger is more likely to come from outside the financials, a competitor's progress or an architecture shift, than from the sales line, which is why the mechanism peers and the industry's technology direction are watched directly.

Revision

The document is revised whenever something might have changed, and for this moat that means the industry's technology direction and the geopolitical environment carry as much weight as ASML's own numbers. Any sign of a competitor reaching the leading edge, any material change to export controls, or any shift in chip architecture pulls a revision forward regardless of when the last one was. Class A is walked in full every time, because the whole discipline of this name is to distinguish a market-size or cyclical dip, which leaves the moat intact, from the rare event that would actually touch the monopoly. The assignment of a condition to Class A or B is fixed before a trigger, never during the revision that reports one.

A note on what this document is and is not. This is written for the author's own investment process and published in that form. It reflects the analysis, assumptions and judgment of the author at the date of writing and nothing more. Nothing here is investment advice, a recommendation tailored to any reader, or an offer to buy or sell any security. Valuation figures, scenarios and price levels are illustrative analyst assumptions unless explicitly sourced to company reports or other primary data. Investing involves risk, including loss of capital, and past performance is no guarantee of future results. Any reader considering an investment should do their own work and consult a qualified adviser who knows their situation. The author may hold, or may come to hold, positions in the securities discussed.