ASML Holding NV ADR · Moat analysis

Report previewASML.USReport date Last updated
← Back to reports
Report contents
01

Industry landscape

Where the company competes: an EUV monopoly and a DUV oligopoly

Market structure: concentrated at the top, with high entry barriers. ASML builds lithography systems, the machines that print circuit patterns onto silicon wafers by projecting light through a patterned mask. Its customers are the fabs that make leading-edge logic and memory chips: TSMC, Samsung, Intel, SK Hynix and Micron, Reuters reported in July 2026. ASML sells no chips itself. It sells the tool that the leading edge of the chip industry currently cannot build without, which is a narrower and more defensible position than the label "semiconductor equipment" implies.

The market is global lithography equipment, and it splits into two submarkets that behave differently. In DUV, where tools make chips for industrial and consumer electronics, ASML held above 80% share as of that Reuters report, ahead of Nikon and Canon of Japan. In EUV, the shorter-wavelength tools used to print most AI processors and leading-edge memory, ASML is the only supplier. Chinese and US rivals are trying to develop alternatives, but none has shipped a tool at the leading edge, and the newest High NA generation, which Intel and others are testing, exists in only a handful of units.

SegmentASML shareCredible rivalsTypical price per tool
EUV (including High NA)about 100%Chinese and US development programs; none shipping at the leading edgeabout $200M; High NA $350M–$400M
DUVabove 80%Nikon, Canonabout $60M

Shares and prices as reported by Reuters, July 2026.

The value chain is narrow at both ends. Upstream, ASML leans on a small roster of European suppliers: Zeiss makes the mirrors that steer EUV light inside a vacuum, ground more precisely than space-telescope optics, and Trumpf builds the lasers that strike tin droplets 50,000 times a second to produce light at 13.5 nanometers. Downstream, the buyer list is short — the leading-edge logic and DRAM fabs, perhaps five or six companies in total. The company also draws revenue from the installed base of tools already in customers' fabs, which it reports separately as Installed Base Management, so part of the business is recurring and tied to machines sold years earlier.

Geography matters less than it once did. A tool qualified in Hsinchu works the same in Icheon or Hillsboro, so engineering decides the sale, not proximity, and ASML assembles EUV machines in the Netherlands before flying them out in roughly 40 containers. Export controls have layered a political boundary on top of a technical one. US restrictions on shipments to China, plus the emergence of Chinese suppliers in the middle of the DUV range, mean the addressable market is not uniform across regions, and ASML does not disclose share by tool type.

Concentration is the first thing to note. The practical test for entry barriers is whether the meaningful competitors fit on one hand. In EUV they do not fit on one finger, and in DUV three suppliers cover effectively the whole market. Either way, the leaders are countable on one hand, which is the usual signal that entry is hard. The barrier here is physical and accumulated rather than commercial. Printing at 13.5 nanometers requires magnetically levitated stages moving wafers faster than 3 meters per second, with accelerations above 15 times the force of gravity, and a single-source optics chain that cannot be duplicated on a startup timetable. Qualification inside a customer's fab takes years, so an entrant needs both a working tool and a chipmaker willing to bet a multibillion-dollar fab on it. That combination is what has kept the field at three names for years.

Buyer concentration is the counterweight, and it is real. When five or six customers generate most system revenue, they negotiate hard and time orders to their own capex cycles. That does not erase the advantage, but it shapes how the returns arrive: volume depends on how many tools customers choose to install, not on unilateral price increases. Reuters reported that ASML shipped 48 EUV systems in 2025 and forecasts 65 in 2026 and 85 in 2027, which makes this a capacity and execution story as much as a demand story. Zeiss is a sole source for the critical optics, so supply of a key component may bind before orders do.

What would change the assessment: a Chinese supplier moving into immersion DUV at scale, or a credible EUV alternative reaching a production fab. Neither has happened yet. What the structure does not show is how much pricing power ASML retains against buyers who can stretch or pull forward orders, and the current evidence only establishes that the customer list is short and the supplier list is shorter.

02

Competitive advantage

Does the company have a moat? Yes, a strong one built on EUV

Moat rating: Strong. ASML passes both tests that decide whether an entry barrier is real. The market is concentrated and stays concentrated, and the leader earns far more than its cost of capital year after year. In extreme-ultraviolet lithography (EUV), the tools that print the smallest features on leading-edge logic and memory chips, ASML has no competitor. Reuters reported in July 2026 that its share of the older deep-ultraviolet (DUV) generation was above 80%, ahead of Nikon and Canon, and that neither rival has shipped an EUV tool into a production fab.

Concentration and persistence together are what separate a structural barrier from a good run. A company can lift returns for a while through better execution, and rivals then copy what works, so high profits on their own prove little. What counts is profit that survives while the structure holds. ASML's return on equity landed between 47% and 63% in each of the past five fiscal years, and the number of firms able to sell a leading-edge lithography tool has not increased in a decade. That is the pattern entry barriers produce, and it is why the rating rests on both tests rather than on either one alone.

Market position: Strong. The governing submarket is EUV, and within it ASML is the only supplier, which makes the ranking question moot. A share of 100% is not a figure that improves. In DUV the company leads an oligopoly: three suppliers, ASML with the largest slice, Nikon and Canon behind. The arena therefore has three tiers by tool generation: one supplier at the top, three in the middle, and a widening group at the low end, where the machines are less advanced and margins thinner. ASML does not break out share by tool type, so the bottom tier cannot be tracked from company disclosure.

Market position stability: Strong. A five-year change of less than 2 percentage points is the rough threshold for very strong barriers. ASML's EUV share has changed by nothing at all over that window, and no rival has qualified a competing tool. The DUV position above 80% has not been overturned either. The caveat is that this stability is measured at the leading edge, where the barrier is physical; it says less about the middle of the DUV range, where the company competes with Japanese incumbents and new Chinese entrants.

Profitability: Strong. Five years of results put ASML far above the 15%-25% average after-tax return on invested capital that marks clear evidence of an advantage.

Fiscal yearRevenue (EUR bn)Gross margin (%)Net margin (%)ROE (%)ROA (%)
202118.6152.731.649.020.5
202221.1750.526.652.516.3
202327.5651.328.463.419.9
202428.2651.326.847.417.1
202532.6752.829.450.519.4

Figures as reported in the company's financial statements.

Return on equity averaged 52.6% across those years and return on assets 18.6%. ROE overstates the case, because buybacks keep the equity base small: at the end of fiscal 2025, total equity was EUR 19.6 billion against EUR 50.6 billion of assets. ROA is the fairer read, and roughly 19% for a capital-goods manufacturer is still a wide spread over any plausible cost of capital. DBS Group Research, in a note dated February 2026, estimates return on invested capital near 40%, a broker figure rather than a reported one, landing in the same range. A like-for-like comparison with Nikon and Canon is not possible, since both report lithography inside broader equipment segments and publish no lithography-only profit.

Gross margin has stayed between 50.5% and 52.8% for five years, the strongest pricing evidence in the record: no competitor has forced a sustained retreat on price. Net margin slipped in fiscal 2024, when revenue rose 2.5% and net income fell 3.4%, a reminder that the business tracks fab capital spending. ASML ended fiscal 2025 with EUR 12.9 billion of cash and generated EUR 10.6 billion of free cash flow, so the returns are cash-backed.

Moat rating: Strong. Two findings carry the rating. ASML is the sole supplier in the submarket that drives its economics, with a DUV franchise behind it, and it converts that position into returns several times its cost of capital without margin erosion over five years. The qualification is about degree, not direction: buyer concentration, export controls and a cyclical order book shape how much of the value ASML keeps, and none has dented it so far. What would move the rating down is evidence that its largest customers can source a leading-edge tool elsewhere, or that the leading edge stops advancing and the advantage shifts from capability to price.

Where the advantage comes from and how long it can last: knowledge, qualification and scale protect a monopoly

Three things do the work: the knowledge and supplier base required to build an EUV tool at all, the captivity that follows from qualifying one inside a fab, and the scale that lets ASML amortize fixed development across the whole leading-edge customer base. The first two are structural. Scale reinforces them, but with a customer list of five or six fabs, scale alone would not be enough.

Supply-side: the knowledge to build the tool is the barrier

Supply-side advantage: Strong. An entrant would have to build a light source, an optics column and a wafer stage that place patterns with atomic-scale accuracy, then persuade a chipmaker to stake a fab on the result. ASML's EUV light comes from a laser striking tin droplets tens of thousands of times a second, and the mirrors that steer it are made by Zeiss, with drive lasers from Trumpf. Those EUV-specific components have one buyer. A rival cannot order them from a catalog, and it cannot order them from suppliers whose entire leading-edge business runs through the incumbent.

The cost structure widens the gap. Developing a lithography generation is a fixed expense paid years before the first sale, and ASML spreads it across essentially the whole leading-edge fab base. An entrant amortizes the same work over zero customers while funding a qualification process that takes years inside a customer's fab. Capital is not the binding constraint: capex was EUR 1.6 billion in fiscal 2025, per DBS Group Research, small next to EUR 32.7 billion of revenue. What money cannot buy is the field data from an installed base that grew while a rival was still drawing its first design, or the engineering organization that produced it.

The durability question is whether that knowledge decays. Knowledge advantages weaken when a technology stops moving, because imitators get time; ASML's protection depends partly on the leading edge continuing to advance, with High NA now entering customer fabs. This is not a cost advantage in the usual sense: ASML's tools are the most expensive in the industry and its gross margin sits near 52%. The relevant claim is narrower and stronger — no one else can build one.

Demand-side: qualification creates slow, partial captivity

Demand-side advantage: Fairly Strong. Captivity at a fab is a matter of qualification, not affection. An EUV scanner sits at the center of a process flow; the fab tunes recipes and yields around that machine over years, and swapping in another supplier means repeating the learning curve with a multibillion-dollar line at risk. That is a genuine switching cost, and it explains why a share of 100% has not required discounting to defend.

Two mechanics reinforce it. The tool is a modest share of the fab's total cost, roughly $200 million against a facility costing many billions, so price sensitivity on the tool is lower than the size of the purchase suggests. The installed base also generates its own revenue: ASML reports service and upgrades on machines already sold as a separate line, so a tool delivered in 2020 still earns in 2026.

The margin record shows captivity translating into pricing. Gross margin held between 50.5% and 52.8% for five years and net margin between 26.6% and 31.6%, so no competitor's discount has forced a durable step down. The rating stops at Fairly Strong because the captive buyers number five or six companies, all of them larger and more sophisticated purchasers than a typical customer base, and they negotiate accordingly. They dual-source DUV, time orders to their own capex cycles, and press the supplier to fund the next generation. The company's 2030 target of a 56%-60% gross margin, as reported by DBS Group Research, is a goal, not an achieved price level, and the fiscal 2024 dip (revenue up 2.5%, net income down 3.4%) shows the order book still moves with fab spending. There is no network effect in this business; the lock-in is qualification-based, slow to build and slow to lose.

Durability risk: the substitute is a different way to make chips

The threat to ASML is not a rival lithography supplier. It is anything that reduces the value of printing the smallest features, and three substitutes matter.

Advanced packaging, chiplets and 3D stacking let designers extract more performance from existing nodes, which cuts the number of EUV layers a device needs. If that becomes the cheaper route for a meaningful share of the market, unit demand for the most expensive tools could flatten even while overall chip demand grows. The evidence does not yet establish that this is happening; the reported unit path runs the other way, with 48 EUV systems shipped in 2025 and 65 and 85 expected in 2026 and 2027, per Reuters.

DUV multipatterning stays good enough for most chips, which is why DUV remains the larger segment by units; EUV pricing power does not extend down the range in the same way.

Chinese toolmakers are the entry that is actually occurring, in the middle of the DUV range. If one reaches immersion-class DUV at scale, ASML's China system sales shrink. Export controls cut both ways: they bar EUV shipments to China, capping the customer base, and they give Chinese fabs and equipment makers a reason to fund domestic alternatives.

Cross-price sensitivity is asymmetric. If ASML raises EUV prices, customers have no alternative supplier to switch to, but they can respond at the design stage by using fewer EUV layers or waiting a node. That is the shape of the protection: strong against rival vendors, weaker against the customer's own engineering choices.

So far the evidence only shows the position holding. Reported first-half fiscal 2026 revenue of EUR 18.1 billion was up about 17% from EUR 15.4 billion a year earlier, and no competitor has qualified a leading-edge tool. What would change the assessment is a Chinese immersion DUV supplier reaching scale, High NA failing to win adoption, or a shift in chip design that lowers EUV layer counts across the industry.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Continue the comparison

Continue reading the full report

  1. Comparison and summaryHow ASML Holding NV ADR compares with its closest rivals, and our moat rating

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 ASML Holding NV ADR