What ASML does, precisely: ASML is the sole manufacturer of Extreme Ultraviolet (EUV) lithography systems — the machines that physically print transistor patterns onto silicon wafers. Every advanced semiconductor below the 7nm node, including chips from TSMC, Samsung, and Intel, requires ASML's machines. Without ASML, the global AI infrastructure stack does not exist as currently constituted. They also make Deep Ultraviolet (DUV) lithography systems for less-advanced nodes, plus metrology and inspection equipment, and they service their installed base under long-term agreements.

The business model has three mutually reinforcing layers: sell the machine (capital equipment), sell upgrades to existing machines (field options), and service the machine throughout its 20+ year life. This creates a flywheel: the larger the installed base, the larger and more predictable the recurring service stream.

Revenue quality: ASML's backlog at year-end 2025 stood at €38.8B — representing roughly 1.2 years of forward revenue at the midpoint of 2026 guidance. This is not a subscription business, but backlog provides unusual visibility for a capital equipment company. Customers sign binding orders years in advance because machines cannot be sourced elsewhere. The service segment (Installed Base Management) is genuinely recurring and is growing faster than systems sales as the global installed base expands.

Pricing power: ASML has exercised pricing power consistently and without meaningful pushback. An EUV NXE system sold for roughly €100M five years ago; today comparable systems clear €180M+. The new High-NA EXE:5200B systems price at approximately €380M each. There is no alternative supplier, so customers pay. No major volume loss has resulted from price increases.

Customer concentration: ASML's revenue is highly concentrated: TSMC, Samsung, and Intel collectively represent the vast majority of EUV system sales. China (predominantly mature-node customers buying DUV) was 33% of total sales in 2025 but is guided to fall to approximately 20% in 2026 following export restrictions. Individual customer concentration is significant — TSMC alone likely accounts for 35–45% of EUV orders.

Scale: FY2025 total revenue was €32.7B ($36B+), up from €28.3B in 2024 (+15.5%), €26.9B in 2023, €21.2B in 2022, and €18.6B in 2021. Five-year revenue CAGR is approximately 12%. Market cap ~$613B. Employees: approximately 43,882.

Profitability: ASML's margins are exceptional and have been expanding over time, though the mix of EUV vs. DUV and service creates quarterly variability.

Cash flow quality: Operating cash flow consistently exceeds net income, reflecting the favorable working capital dynamics of ASML's business model. The company collects advance payments (down payments on large orders), which means it often gets paid before it delivers. FCF conversion rate is high: operating cash flow ran to approximately €12.1B vs. net income of €9.6B in FY2025, an ~126% conversion ratio. There are no significant accounting red flags or unusual adjustments.

Balance sheet: ASML carries roughly €4.7B in total debt against €13.3B+ in cash, resulting in a net cash position of approximately €8.6B. Debt-to-EBITDA is below 0.3×. The company has no meaningful leverage risk. Key off-balance sheet items to monitor: operating leases (not material at scale), pension obligations (standard Netherlands-domiciled corporate; manageable), and the €1.3B investment in Mistral AI made in 2025 (venture-stage, mark-to-market risk).

Capital intensity: ASML is a moderately capital-intensive assembler, not a manufacturer — it outsources most component production. Capex was approximately €1.6–1.8B in FY2025, roughly 5–6% of revenue. The company is investing $2.2B in facilities in 2026 (a 20% increase YoY) to expand EUV production capacity. Maintenance capex is a small fraction; the bulk is growth investment. This keeps FCF generation strong even during expansion phases.

Working capital dynamics: Favorable. ASML receives partial advance payments when customers place orders (often 20–30% of the machine value) and progress payments during manufacturing. This is a business that is partially pre-funded by its customers — similar in structure to a high-value custom manufacturer. Cash conversion cycle is short relative to peers.

ROIC: Depending on methodology, ROIC runs 30–74%. The wide range reflects different treatments of goodwill and intangibles. The most conservative widely-cited figure (~30%) is still comfortably above ASML's cost of capital, confirming genuine economic value creation. ROE runs approximately 52%, driven by asset-light assembly and favorable capital structure.

CEO — Christophe Fouquet: Appointed President and CEO on April 24, 2024, succeeding the long-serving Peter Wennink. Fouquet joined ASML in 2008, rose through Applications and EUV business line leadership, and was EVP of EUV (the company's most strategic business unit) from 2018 until becoming CEO. This is the profile of a deep insider — a technology-first operator who built the EUV franchise from the inside, not a financial engineer brought in from outside. He holds a physics degree from Institut Polytechnique de Grenoble and previously worked at KLA Tencor and Applied Materials, giving him industry-wide perspective.

Two years into the role, Fouquet has executed a deliberate strategic pivot from "hyper-innovation" (building EUV from scratch) to "operational scale" (ramping High-NA production, expanding manufacturing capacity, growing the service base). This is the correct strategic posture for the current moment. Guidance given since his appointment has been met or exceeded: FY2025 guidance of ~15% revenue growth was achieved exactly; Q1 2026 came in above the high end of guidance, triggering a full-year guidance raise.

Skin in the game: Fouquet directly owns approximately 0.002–0.008% of shares outstanding, worth roughly €10–46M depending on the source and timing. This is modest in percentage terms but meaningful in absolute dollars. Annual total compensation is approximately €5.4–7M (18% salary, 82% performance-linked), below comparable US CEOs at this market cap. Compensation structure heavily weights long-term incentives. No significant insider selling has been reported.

Key lieutenants: Roger Dassen (CFO) is a long-tenured executive with a reputation for precision and conservatism in guidance — a good trait. Frédéric Schneider-Maunoury (COO) oversees operations. Wayne Allan rounds out the Board of Management. Average management tenure is approximately 5.5 years — seasoned without being entrenched.

Board: The Supervisory Board is separate from the Board of Management under Netherlands two-tier governance. The Chairman is not the CEO (good). The board is predominantly independent. The Dutch governance model provides structural protection against short-termism.

Capital allocation: Excellent track record. FY2025 returned €8.5B to shareholders via dividends and buybacks. A new €12B share buyback program was announced in January 2026, to be executed by December 2028. The €1.3B Mistral AI investment is the one notable departure from pure capital return — a strategic bet on AI-driven lithography optimization that may or may not generate financial returns. Share count has been declining meaningfully (−1.5% in the last year).

Does a moat exist? Yes. ASML possesses one of the widest and most durable competitive moats in global industry — not just in technology. The claim is not hyperbole; it is supported rigorously by the financial evidence.

Moat type — multiple reinforcing sources:

1. Absolute intangible asset moat (patents + tacit knowledge): EUV lithography requires a 13.5nm wavelength light source, mirrors polished to atomic flatness, and a supply chain ASML spent two decades and over $9 billion constructing with Carl Zeiss SMT (optics) and Cymer (light sources, now owned by ASML). A single EUV machine contains over 100,000 parts, 3,000 kilometers of wiring, and mirrors so precise that if scaled to the size of Germany, the largest surface irregularity would be less than one millimeter. ASML has over 12,000 patents. The tacit knowledge embedded in its workforce and supplier relationships cannot be transferred by reading documents.

2. Switching cost moat: Chipmakers integrate ASML's machines into fab processes that take years to qualify and tune. The process expertise built around NXE tools cannot be transferred to a hypothetical alternative machine without massive yield loss and re-qualification cost. TSMC or Samsung cannot switch even if an alternative existed.

3. Efficient scale: The total addressable market for EUV lithography systems is large but not unlimited — perhaps 60–90 units/year at peak. The investment required to enter (10+ years, $10B+) is prohibitive for any return profile given the market size. Nikon and Canon, the only historical competitors in lithography, have ceded the leading edge entirely; Canon announced an alternative (nanoimprint) for commodity logic, but it cannot address sub-7nm AI chip production.

Moat strength: Exceptional. A well-funded competitor beginning today could not produce a competitive EUV system within 10 years, and likely not within 15–20 years. China has attempted to develop domestic alternatives (SMEE); the most recent reports suggest they remain approximately 15–20 years behind ASML's current capability.

Moat trend: Widening. High-NA EUV (the EXE series) extends ASML's monopoly further into the future. Customers who have adopted EUV are now adopting High-NA — extending the lock-in. The Installed Base Management segment growing at 26% annually demonstrates the moat deepening with the installed base size.

Evidence in the numbers: 52.8% gross margins, 30%+ ROIC consistently, and the ability to raise average selling prices from €100M to €380M over a decade with zero customer attrition are the financial fingerprints of a genuine monopoly moat. No other semiconductor equipment company achieves comparable metrics.

TAM and growth: The global semiconductor equipment market is estimated at approximately $100–120B today and is projected to approach $150–280B by 2028–2030, driven primarily by AI infrastructure buildout. ASML's own 2030 revenue guidance of €44–60B implies the total wafer fabrication equipment market expands substantially — management would not commit to those targets without credible demand signals from customers who are ordering years in advance.

Secular tailwinds — exceptionally strong: Artificial intelligence is the most powerful secular driver ASML has ever faced. Every leading AI chip (NVIDIA H100/B200 series, AMD MI300, Google TPU, custom silicon from Microsoft, Meta, Amazon, Apple) is manufactured using ASML EUV systems. As Big Tech's AI infrastructure spending accelerates past $600B annually, every dollar flows partially through ASML. Advanced logic and DRAM for AI are both simultaneously ramping, creating dual cycles that rarely coincide. Advanced DRAM (HBM) production is now a material and growing portion of EUV demand.

Secular headwinds — present but manageable: Export controls on China eliminate ASML's access to a market that was 33% of 2025 revenue. The headwind is real and significant. However, EUV was never sold to China (export restrictions on EUV predate recent escalations), meaning the lost revenue is predominantly lower-margin DUV. The mix shift away from China toward AI-driven EUV actually improves ASML's margin profile at the gross line even as reported China revenue falls.

Competitive intensity: In EUV, ASML has zero competition. In DUV, Nikon retains some market share in older-generation tools but cannot compete at leading edge. In metrology/inspection, KLA Corporation (KLAC) is the dominant player and a genuine competitor, though the two companies primarily serve different steps in the fab process. Applied Materials (AMAT) and Lam Research compete in deposition/etch — orthogonal to lithography.

Cyclicality: ASML is cyclical at the DUV level. EUV demand has been more structural due to the limited number of chipmakers that can absorb the volumes. During 2020, orders and revenue held up well as COVID accelerated digitalization. During 2008–2009, ASML revenue fell approximately 30% — a sharp cyclical decline that predates the EUV era. Today, with 25%+ of revenue in recurring services, and EUV demand structurally anchored to a multi-year AI capex cycle, the business is somewhat more resilient to downturns than a decade ago. However, any significant semiconductor capex pause would still hit system sales meaningfully.

ASML is not cheap. It is priced as a monopoly compounder at a historically high multiple. The honest valuation question is not "is it cheap" but "is the price justified by the quality and trajectory of the business?"

DCF sanity check: Using conservative assumptions — 2026E revenue of €38B (mid-guidance), 10% revenue CAGR through 2030 (below company guidance), margins flat to slightly expanding toward 54%, 10% discount rate, 3% terminal growth — intrinsic value comes in roughly in the €1,100–1,300/share range in EUR (approximately $1,200–1,400 USD at current exchange). At $1,584, the stock is pricing in significant execution of the bull case. GuruFocus's GF Value model pegs fair value at ~$1,121, flagging the stock as significantly overvalued by ~41%. This is not a value investor's entry point.

Why the stock is at its current price: ASML has rallied approximately 132% from its 52-week low of $683 (mid-2025). The low reflected the October 2024 "accidental" early earnings release that exposed weaker-than-expected bookings and triggered a ~35% single-day crash. Recovery has been driven by: (1) Q1 2026 earnings beat and guidance raise to €36–40B, (2) re-acceleration of AI capex by hyperscalers, (3) improving EUV/High-NA booking trajectory, and (4) broad re-rating of AI infrastructure stocks.

Owner earnings estimate: Net income €9.6B + D&A ~€0.8B − maintenance capex ~€0.6B = owner earnings ~€9.8B. At €32.7B enterprise value in EUR equivalent, that implies a price-to-owner-earnings of approximately 30×. More reasonable than headline P/E but still not cheap.

Value trap risk: Low. This is not a cheap stock hiding a deteriorating business. This is an exceptional business priced accordingly. The risk is paying a full price for a good business and experiencing years of multiple compression, not fundamental collapse.

Dividends: ASML pays a semi-annual dividend, currently yielding approximately 0.46–0.61% (USD). The yield is modest but growing — dividends have grown at a double-digit CAGR over five years. FCF payout ratio is low (approximately 20–25%), leaving ample room for continuation and growth. Dividend is well-covered and sustainable.

Share buybacks: ASML returned €8.5B to shareholders in FY2025 through dividends and buybacks combined. A new €12B buyback program was announced in January 2026, to be executed by December 2028. The share count has declined approximately 1.5–2% annually — not aggressive dilution-fighting, but consistent and meaningful over time. Buyback efficiency is reasonable; the stock was purchased at prices well below current levels during the mid-2025 trough.

M&A track record: ASML's most significant acquisition was Cymer (light source technology, ~$2.5B in 2013), which proved strategically critical — it brought EUV light source development in-house and accelerated commercialization. The €1.3B Mistral AI investment in 2025 is a strategic partnership aimed at deploying AI in lithography process optimization and operations — this is venture-scale risk at a small fraction of capital, not empire-building.

R&D investment: ASML invested approximately €4.7B in R&D in FY2025, representing roughly 14% of revenue — among the highest ratios in industrial technology. This is the engine of the moat. ASML's willingness to sustain and grow R&D spending even in revenue-soft quarters is a hallmark of management's long-term orientation.

Overall capital allocation grade: A. ASML is a shareholder-friendly capital allocator that invests heavily to sustain its moat and returns surplus cash through disciplined buybacks and a growing dividend, without dilutive M&A or excessive executive compensation.

Stated strategic priorities: (1) Ramp High-NA EUV (EXE series) to volume production — target of 20 High-NA tools by 2027/2028; (2) Expand EUV manufacturing capacity to at least 60 units in 2026 and 80+ by 2027; (3) Grow Installed Base Management (service/upgrades) from €8.2B toward €12B+ by 2030; (4) Develop next-generation Hyper-NA EUV (>0.7 NA) targeting the 2030+ horizon; (5) Deploy AI (including the Mistral partnership) to optimize holistic lithography and internal operations.

Early evidence of progress: Q1 2026 revenue came in at €8.8B, beating the high end of prior guidance. The guidance raise to €36–40B for FY2026 suggests customer demand is confirming the strategy. Revenue recognition on the first EXE:5200B system (High-NA) occurred in Q4 2025, after site acceptance testing — a critical milestone. EUV system sales grew 39% in FY2025. The Installed Base grew at 26%, demonstrating the flywheel is working.

Notable strategic decision — stopping quarterly bookings publication: Starting Q1 2026, ASML ceased publishing quarterly net bookings. Management frames this as reducing short-term noise (bookings are lumpy and not representative of underlying demand). However, this reduces transparency for investors and obscures demand signals. It is a legitimate governance concern, even if the rationale is defensible.

Management credibility on guidance: Strong. Under both Wennink and Fouquet, ASML has generally met or modestly beaten annual guidance, with the notable exception of the 2024 bookings scare. The pattern of setting a wide guidance range (currently €36–40B for 2026) and typically printing near or above the midpoint is consistent. Management has credibility.

Potential catalysts (next 12–24 months): (1) High-NA EUV volume ramp confirmation at TSMC and Samsung; (2) Additional guidance raises if AI capex remains elevated; (3) Resolution or clarification of China export control situation (either a further restriction is priced in, or partial easing provides upside); (4) Q2 2026 results on July 15, 2026 — the first quarter where China mix decline is fully visible; (5) Any update to 2030 revenue scenario targets.

Is AI a threat? No. AI cannot design or manufacture the machines that make AI chips. ASML's physical manufacturing process — precision optics, laser plasma physics, sub-atomic-scale engineering — is not subject to AI disruption. The company employs thousands of physicists and engineers whose work is at the frontier of what is physically possible; AI tools augment but cannot replace this capability on any foreseeable timescale.

Is AI a revenue opportunity? Emphatically yes — the most powerful revenue opportunity ASML has ever encountered. Every AI chip manufactured at leading-edge nodes (NVIDIA H100, B200, Blackwell, and successors; AMD MI series; custom TPUs, Trainium, Gaudi) requires ASML EUV systems. Each time compute demand for AI doubles, chipmakers need more fab capacity, which means more ASML machines. The AI buildout is the most powerful secular demand driver ASML has seen, exceeding the smartphone era in capital intensity.

Is AI a tool being deployed internally? Yes. The €1.3B Mistral AI investment in 2025 is a direct bet on using AI to improve "holistic lithography" — the optimization of lithography recipes, process control, and yield improvement across the entire fab process. ASML's computational lithography software (a growing business) is an AI-enabled product offering. Specific financial impact of internal AI deployment is not yet material but is directionally positive for margins over time.

R&D investment vs. peers: At ~14% of revenue, ASML's R&D intensity is among the highest in the semiconductor equipment space, ahead of Applied Materials (~9%) and Lam Research (~11%). ASML is unambiguously a technology leader in its domain — not a follower or laggard.

Data assets: ASML accumulates enormous amounts of process data from the millions of wafer exposures its installed base performs globally. This data — combined with the Holistic Lithography software suite — creates a growing software/data business that is currently undermonetized relative to its potential. As AI improves process optimization, ASML's data network effect may become an additional layer of competitive advantage.

Insider ownership: Approximately 0.002–0.008% directly owned by executives — modest in percentage terms. No significant insider selling was identified in the past 12 months. Dutch governance practices mean executive compensation is more restrained than US peers; the incentive structure is nonetheless long-term oriented through deferred share awards.

Institutional ownership: ASML is heavily held by large institutional investors globally. Key holders include Vanguard, BlackRock, and major European pension funds. Ownership skews toward long-duration, fundamental investors rather than short-term traders — consistent with ASML's history as a compounder. Institutional ownership appears to have been increasing alongside the price recovery in 2025–2026.

Short interest: Remarkably low. Short interest is approximately 0.18% of the float — effectively no meaningful bearish position is being held against this stock. This reflects both the strength of the fundamental thesis and the practical difficulty of shorting a true monopoly. Low short interest also means there is no short-squeeze dynamic to inflate the price artificially.

Analyst consensus: Strongly bullish. Of 38–39 analysts covering the stock (NASDAQ listing), 38 recommend Buy and 1 recommends Sell. The average 12-month price target is approximately $1,461–1,662, implying roughly flat to +5% upside from current prices. The spread from highest ($1,994) to lowest ($900) is extremely wide — reflecting genuine disagreement about China exposure and the trajectory of the AI cycle. The sell-side consensus has been persistently too conservative on ASML's growth trajectory over five years.

Activist involvement: None identified. ASML's governance structure (Netherlands two-tier board, substantial institutional ownership, Dutch government sensitivity to strategic assets) makes activist campaigns impractical and unlikely.

Asymmetry assessment: From the current price of $1,584, the bull case offers ~60–75% upside and the bear case represents ~35–43% downside. The expected-value weighted upside is approximately equal to or slightly below the downside — the 2:1 or better upside/downside ratio that characterizes an attractive entry is not clearly present at current prices. Critically, this assessment improves significantly at lower entry prices. At $1,200–1,350 (approximately 15–25% below current prices), the asymmetry improves to approximately 2.5:1 upside/downside — a genuinely favorable setup.