ASML / RESEARCH · ↑ INDEX 2026-05-03
Silent Engineering Fund
May 03, 2026
ASML
Analysis Date: 2026-05-03
OVERWEIGHT

ASML is the only company on earth that can sell an EUV lithography tool, and even ASML cannot build one without Carl Zeiss SMT — a partnership that began in 1984, formalized in a 24.9% equity stake, and codified in a 20-F line that states ASML "would effectively cease to be able to conduct our business" without it. That is the moat, written by ASML's own counsel: a single-vendor monopoly that itself depends on a single supplier, with integrated R&D roadmaps and no second source on the planet. The stock has run +115% YoY into $1,427 and a $550B cap, and the natural reaction is that the easy money is gone. But TSMC just pushed its high-NA EUV qualification deeper into the decade, and the next ASP step-up now lands closer to 2029 than the consensus tape priced — a tension the body unpacks. Underneath: ~$126B of customer capex committed across TSMC Arizona, Samsung Texas, Intel Ohio, and TSMC Japan, every site spined by ASML tools. Meanwhile the China revenue mix compresses 36% → 29% → 19% over 18 months as export controls bind, with Korea, Taiwan, and US fabs absorbing the displaced demand. One vendor. One optics partner. One bottleneck for everyone reshoring silicon at once.

Sources: fundamentals_report, market_report, news_report, investment_debate_state.bull_history, final_trade_decision, risk_factors_item_1a (Zeiss 20-F dependency clause), customer reshoring capex aggregate, China mix trajectory.

Confidence: 7 facts | 1 inferred | 0 speculative

Gaps: None — all required fields present.

Portfolio Decision

Initiate or build to a 4% position in ASML at $1,427 via three scaled tranches (50% at $1,427, 30% at $1,375, 20% at $1,300), with a hard stop at $1,200 and mechanical trims above $1,575. The 12-month target is $1,575-$1,650 (+10-16% upside before dividends) anchored on FY 2026 EPS of €31-32 at 28-29x forward. Time horizon: 12-18 months for the core thesis, with Q2 2026 earnings on 2026-07-15 as the first major re-rating event.

ASML is the only EUV monopoly in the global semiconductor equipment market and the single most important supplier underwriting Moore's Law continuation, AI memory expansion, and customer-funded fab reshoring. The fundamental case is anchored in three concrete metrics: a €38.8B backlog representing ~115% of FY 2026 guide midpoint, 2025 net income growth of +27% to €9.6B, and a 52.8% gross margin / 34.6% operating margin / 52.2% ROE profile that is unique in capital equipment globally. The bull case for the next 18 months is a continuation of the 2025 reacceleration into a €36-40B 2026 revenue print, supported by Q4 2025's €13.2B booking quarter and ongoing memory/AI capex pull from SK Hynix, Samsung, and Micron. The bear case is real and bounded: Q2 2026 gross margin guide of 51-52% confirms 80bps of structural compression, TSMC's April 23 deferral of high-NA EUV adoption to ~2029 caps the next major ASP step-up, and China revenue at just 19% of Q1 2026 sales (down from 36.1% in FY 2024) reflects irreversible export-control compression. The Carl Zeiss SMT single-source optics risk and 38% top-2 customer concentration are explicitly disclosed risk factors warranting position sizing discipline (capped at 4-5%, not 7-8%).

The Overweight rating reflects three judgment calls. First, the bull's specificity bar was cleared on 3 of 5 items cleanly (dated catalysts, budget cycle, consolidation reference) and partially on 2 (memory math magnitude, kill-switch tightness) — a single full-equivalent miss that downgrades from Buy to Overweight per the research checklist rule. Second, the entry price is good but not cheap; +115% in 12 months has front-run multiple expansion, and the next 12 months are about earnings catching up to the 29.9x forward multiple, not further multiple rerating. Third, the portfolio manager's prior cross-ticker lesson from the AAPL +3.7% case (memory log) reinforces the discipline of taking quality-monopoly Hold-to-Overweight positions with patient sizing rather than swing-for-the-fences sizing.

The kill-switches are clear: (1) if Q2 2026 bookings come in below €5B AND 2026 guide is unchanged or trimmed, the rating drops to Hold; (2) if Q3 2026 China revenue prints below 12%, geopolitical compression has another leg and the position is trimmed by 25%; (3) if a major Carl Zeiss SMT operational disruption occurs, the position is exited regardless of thesis intactness; (4) if the stock breaks $1,200 on any catalyst, the multi-year thesis requires re-examination and the position is closed.

Quantitative Lane

Customer Concentration

Top customers by % of 2025 net sales (from 2025 Annual Report):

Customer / Country% of 2025 salesEUR% of 2024 salesYoY change
Largest customer (TSMC by inference)23.9%€7.797B16.6%+7.3pp
Top 2 customers38.0%€12.41Bn/an/a
Taiwan total25.5%€8.33B15.4%+10.1pp
South Korea total25.0%€8.17B22.7%+2.3pp
China total29.1%€9.51B36.1%-7.0pp

Customer concentration is rising structurally: TSMC's share grew from 16.6% to 23.9% in one year (+€3.1B absolute revenue) as Arizona and N2 ramps absorbed ASML capacity. Top 2 = 38% is an explicit risk factor disclosed in the 20-F. Geographic concentration: 50.5% of revenue from Taiwan + Korea, both home to leading-edge foundries and memory makers.

Backlog & Book-to-Bill

Backlog and book-to-bill (from Q4 2025 investor presentation, Jan 28, 2026):

PeriodNet Sales (€M)Net Bookings (€M)Book-to-Bill
Q4 20249,2637,0880.77x
Q1 20257,7423,9360.51x
Q2 20257,6925,5410.72x
Q3 20257,5165,3990.72x
Q4 20259,71813,1581.35x
FY 202532,66728,0350.86x
FY 202428,26318,8990.67x

End of FY 2025 backlog: €38,797M (€38.8B), up from €35,938M at end of FY 2024.

Q4 2025 bookings of €13.2B (€7.4B EUV alone) was the second-largest booking quarter in ASML history and reset the trajectory. The annual 0.86x book-to-bill ratio for FY 2025 looks weak in isolation but is misleading — three modest quarters were followed by a binary Q4 spike. The €38.8B backlog covers ~115% of the FY 2026 guide midpoint of €38B, meaning the year is essentially booked. Q1 2026 bookings figure not disclosed in press release; CEO commentary indicated "order intake continues to be very strong."

Forward visibility: EUV capacity booked through 2027 per management commentary.

Research Manager

Investment Plan — Overweight

Rationale:

The bull case is fundamentally correct on direction and structurally well-supported: ASML is a 50%+ gross margin monopoly with a €38.8B backlog, a 2025 P&L that re-accelerated to +16% revenue / +27% net income, customer-funded reshoring tailwinds, and 12-quarter visibility from booked orders. The bear case is correct on slope and timing: margin guide compression is real (80bps in 2026), TSMC's high-NA deferral pushes the next-leg story by 2-3 years, China at 19% of Q1 2026 may step down further to mid-teens, and the 29.9x forward P/E has front-run a substantial portion of the recovery.

The fight is over whether the next 12 months earn investors more by adding to the position (bull) or trimming it (bear). My read: the magnitude of the bull's revenue/backlog visibility is too high to short — €38.8B booked through 2027, customer-funded reshoring underwriting another 2-3 years of demand, and best-in-class ROE that grows with each buyback tranche. But the magnitude of the bear's points is too high to be aggressively long — at $1,427 vs. a $675 52-week low, you've already paid for the easy multi-bagger; the next 12 months are about earnings growth catching up to multiple, not multiple expansion. Overweight is the correct rating: stay long, add on weakness toward $1,300-$1,375, but don't be the marginal buyer at $1,500+.

Strategic Actions:

  1. Maintain core long position at current weight; multi-year hold with structural compounding.
  2. Add on weakness: Scale buying between $1,375 and $1,300. Below $1,300 is high-conviction add. Below $1,200 is back-up-the-truck.
  3. Trim on strength: Above $1,500, trim 15-20% of position; above $1,650, trim aggressively.
  4. Q2 2026 print (July 15) is the primary re-rating event. Bookings >€8B + guide raised again = upgrade to Buy. Bookings <€5B + guide unchanged = downgrade to Hold.
  5. Position sizing: 4-6% of an active growth/quality portfolio. Capped due to 38% top-2 customer concentration and Carl Zeiss SMT single-source risk.
  6. Stop-loss: $1,200 (below 200-day SMA). Below = trend break, re-examine thesis.

Trader

Transaction Proposal — Buy

Action: Buy

Reasoning: ASML is the only viable EUV monopoly with €38.8B of backlog covering 115% of FY 2026 guide midpoint, +27% 2025 net income growth, and best-in-equipment-class ROE of 52.2%. The post-Q1 pullback to $1,427 — down 7.8% from the $1,547 52-week high — provides a constructive entry just above the 50-day SMA. Margin compression and the TSMC high-NA deferral cap upside near-term, but the customer-funded reshoring wave and AI memory order surge underwrite a 12-15% revenue CAGR through 2030.

Entry Price: Tactical scaled entry: 50% at $1,427 (current), 30% at $1,375 (50-day SMA test), 20% at $1,300 (Bollinger lower / 2-sigma).

Stop Loss: $1,195 (below 200-day SMA at $1,112 with 7% buffer for volatility).

Position Sizing: 4-5% of an active growth/quality portfolio. Tier 1 position, capped at 5% due to customer concentration (top 2 = 38%) and Carl Zeiss SMT single-source risk.

FINAL TRANSACTION PROPOSAL: BUY

Investment Debate

Bull Researcher

ASML at $1,427 is a once-a-decade compounder. Specificity bar items: (1) Dated catalysts: 2026-07-15 Q2 earnings, 2026-Q4 Investor Day. (2) Memory math: SK Hynix + Samsung + Micron 2026 capex ~$95B, EUV intensity 12-15%, ASML capture ~$8-10B = 23-28% of 2026 high-end guide (NOTE: bear later argued this overstated by ~50%, real $4-5B). (3) Budget cycle: TSMC $45B + Samsung $25B + Intel $20B = $90B+ logic capex. (4) Consolidation: TSMC AZ $65B, Samsung TX $17B, Intel OH $28B, TSMC Japan $16B reshoring wave. (5) Kill-switch: FY 2026 net bookings <€25B = thesis broken.

Numbers: 2025 revenue €32.7B (+15.6%), NI €9.6B (+27%), ROE 52.2%, GM 52.8%. Backlog €38.8B = 115% of 2026 guide midpoint. EUV: 48 systems shipped in 2025, 67 new + 12 used in Q1 2026. 2030 model €44-60B revenue at 56-60% GM. €12B buyback through 2028.

Rebuttal to bear on margin: 80bps of GM compression is real but offset by Installed Base Management growth (€2.5B Q1 2026, 28% of sales). On TSMC high-NA: not abandonment, just deferral; Samsung, SK Hynix, Intel still committed. On China: 19% is the floor; further step-down to 14% = €1.9B revenue gap, already in guide. On valuation: 29.9x forward is consistent with EUV monopoly premium vs AMAT 22x, LRCX 24x, KLA 28x. On Zeiss: 1984 partnership, 24.9% ASML ownership, integrated R&D — risk real but mispriced as binary.

Recommendation: Buy. Conviction: high. Time horizon: 12-18 months. Target: $1,650-$1,750.

Bear Researcher

The bull's specificity bar is partial. Memory EUV intensity is 6-9%, not 12-15% — bull's $11-14B is overstated by ~50%; real $6-8B globally with ASML capture $4-5B. Kill-switch of €25B 2026 bookings is too soft; real threshold is Q2/Q3 quarterly <€4B.

Five points the bull skipped:

  1. Margin compression is structural. Q1 GM 53.0%, Q2 guide 51-52%, FY 51-53%. ~80bps compression on €38B = €304M gross profit loss = €250M net income hit (~2.6% of 2025 NI). 3-year cumulative ~€900M. (NOTE: bull conceded; counter-argued Operating Margin is range-bound, not expanding, in management's own 2030 model)
  1. TSMC high-NA deferral is bigger than bull acknowledges. ASML installed base of EXE:5000 was supposed to ramp from <10 in 2025 to 40+ by 2027. With TSMC out, realistic 2027 figure is 20-25 — ~50% reduction. High-NA was THE story for 2027-2030 EBIT step-function.
  1. Valuation has front-run recovery. +111% from 52-week low. 29.9x forward assumes 2027 EPS growth >20%, but with margin compression and TSMC capex flatness, 2027 EPS growth could be 8-12%. 40-50% derating risk if multiple compresses to 22-25x. (NOTE: bull rebutted that 22-25x is Lam-multiple, wrong for EUV monopoly)
  1. China revenue cliff not over. EU export controls still being implemented; US Oct 2025 controls on metrology/software not yet fully in run-rate. Q2-Q3 2026 likely 14-17% China share. Chinese installed-base service revenue (high margin) being eroded.
  1. Carl Zeiss SMT single-source. 20-F: "we would effectively cease to be able to conduct our business." Binary risk with no hedge. (NOTE: bear conceded this in Round 2 — base rate too low to short on)

Round 2 additions: IBM growth is +7-8% YoY, half of system sales growth — not a real offset lever. China revenue swap (DUV high-margin out, EUV ramp in) is slightly GM-negative.

Recommendation: Underweight. Conviction: high. Time horizon: 6-12 months. Target: $1,150-$1,250 (forward P/E 24-26x on realistic 2027 EPS).

Risk Assessment

Aggressive

Round 1: 7-8% position, scale to $1,300, target $1,800 by Q3 2027. Risks (TSMC high-NA, China, Zeiss) priced in. Asymmetry favors aggressive long: downside bounded by backlog (~$42B vs $550B mkt cap), upside unbounded by AI capex.

Round 2: NVIDIA at $4T traded 35-40x; ASML at half scale + clearer monopoly should trade similar. 32x forward * €31.50 EPS = $1,710. Plus buyback impact ~2%/yr EPS accretion = $1,750 by 2027. Move to 6%.

Conservative

Round 1: 2.5-3.5% position, hard stop $1,195, mechanical trim above $1,500. Three risks: (1) margin compression structural ~€900M cumulative 3-yr; (2) customer concentration at risk-flag (top 2 = 38%, TSMC slip 15% = €1.2B revenue hit); (3) Taiwan 25.5% concentration = catastrophic tail.

Round 2: NVIDIA precedent — multiple compression on great compounders happens. "Can't hedge tail = size down." Top-2 absolute exposure now 4x larger than 2018 (€7.8B vs €1.5B). Concede 4% but fund from broader semi exposure (AMAT, LRCX) to keep tech weighting constant.

Neutral

Round 1: 4% position correct (narrow from 4-5%). Tier 1 quality with tier 2 concentration risk haircut. Fair value 12-mo: FY 2026 EPS €31-32 * 28x = $1,545-$1,610 = +8-13%. Modest single-digit appreciation, not rerating.

Round 2: 4% holds. Q2 2026 print (July 15) is next major info event. >50% of intended size in by entry. Flat-to-down stock between now and Q2 = full position add condition. Rip above $1,500 between now and July 15 = mechanical trim.

Analyst Reports

$661.5 $876.9 $1,092 $1,308 $1,523 May 2025 Oct 2025 May 2026 10.8M ASML · PRICE & VOLUME$1,427 +115.7%

Daily close (warm umber) with under-fill, volume bars in tan. Trailing range capped at 252 trading days.

Market Analysis

Market Report

ASML closed at $1,427.02 on May 1, 2026, sitting in a tactical equilibrium near the 50-day SMA ($1,398.76) and 10 EMA ($1,426.00), but well above the 200-day SMA ($1,112.33) — a +28% trend premium that signals a strong intermediate uptrend has been re-established after the early-April drawdown. The stock traded in a $675-$1,547 range over the past 52 weeks, putting current price at the 92nd percentile, meaning longs are sitting on substantial cushion but bulls have already paid for most of the obvious good news.

Trend structure. The price action over the past month tells a 3-act story: (1) early-April rally from $1,300 to a peak of $1,528 on April 14 (+18% in 9 sessions, front-running Q1 earnings), (2) post-print whipsaw to $1,375 on April 22 as TSMC's high-NA EUV deferral landed and OpenAI growth fears rolled through semis, (3) recovery rally to $1,438 on April 30 / $1,427 on May 1 as institutional money returned. This is textbook "consolidation under prior breakout" — the $1,500 zone is now resistance, $1,375-$1,400 is intermediate support, and $1,300 is the must-hold level.

Momentum signals — bearish skew. MACD line at 10.79 sits below the signal line at 15.63 (histogram -4.84), meaning a confirmed bearish crossover printed in late April and momentum continues to decay. RSI of 51.6 is neutral. MACD peaked at 28.4 on April 20 and has compressed by 62%, classic post-rally exhaustion behavior.

Volatility regime. ATR of 53.4 implies expected daily range of ~3.7% of price, elevated. Bollinger Bands sit at $1,541 upper / $1,430 middle / $1,321 lower (width = 15.3% of price). VWMA at $1,439.55 is just above spot, mildly bullish.

Key technical levels:

LevelPriceSignificance
Resistance 1$1,500-$1,540Bollinger upper / April 14 high
Pivot$1,427Spot / 10 EMA confluence
Support 1$1,39850-day SMA — first dip-buy zone
Support 2$1,321Bollinger lower / 2-sigma down
Critical$1,112200-day SMA — trend invalidation

Verdict. Tactically neutral, structurally bullish. MACD bearish cross and post-earnings consolidation suggest sideways-to-down chop in 2-4 weeks ($1,375-$1,500 range). Position sizers should fade strength toward $1,500 and add toward $1,375-$1,400.

IndicatorReadingSignal
Price vs 50 SMA+2.0%Bullish (near-term)
Price vs 200 SMA+28.3%Strong bullish (long-term)
MACD vs Signal-4.84 (histogram)Bearish crossover
RSI(14)51.6Neutral
ATR(14)$53.4 (3.7%)Elevated volatility
Bollinger positionAt middle bandNeutral
VWMA vs Price+0.9%Mild bullish flow
52W High proximity-7.8%Below recent high

Social Sentiment

Social Sentiment Report

ASML enters May 2026 carrying analyst goodwill that has been re-tested but not broken. UBS and Deutsche Bank raised price targets following the April 15 Q1 print, both reaffirming Buy; CNN's analyst panel has 82% Buys and 13% Holds across 45 covering shops. Retail tone (Motley Fool's "3 Monster Stocks to Hold for the Next 20 Years", Zacks Buy upgrade) leans long, but sophisticated commentary (Simply Wall St., Morningstar) is openly debating valuation after the +115% 1-year run.

Bull narrative — strongest threads:

  1. Q1 2026 beat-and-raise. Revenue €8.77B inside guidance, GM 53.0% at top end, 2026 guide raised to €36-40B from €34-39B. CEO Christophe Fouquet's "demand for chips is outpacing supply" became the quote of the week.
  2. AI capex secular thesis. SK Hynix and Samsung memory orders filling the order book. Q4 2025 saw memory bookings (€7.4B) overtake logic for the first time.
  3. Buyback + dividend signaling. €12B buyback through Dec 2028, €1.1B repurchased in Q1 2026, dividend raised 17% to €7.50/share.

Bear narrative — strongest threads:

  1. TSMC high-NA EUV deferral to ~2029. Most damaging headline of the past month. Deputy Co-COO Kevin Zhang publicly cited cost as the rationale. Invalidates the "high-NA upgrade cycle starting 2026-2027" thesis.
  2. Margin guide compression. Q1 GM 53.0%, Q2 guide 51-52%, full-year 51-53%. ~80bps compression vs 2025's 52.8%.
  3. OpenAI growth question. April 28 selloff (-3% intraday) was sentiment-driven (less inference compute → less foundry capacity → less litho).
  4. China revenue cliff. China was 36.1% of 2024 sales, 29.1% of 2025, just 19% of Q1 2026.

Insider transactions: None reported. Net neutral.

Sentiment laneReadDirection
Sell-side rating mix82% Buy / 13% HoldStrong bullish
Recent target changesUBS, Deutsche Bank raisedBullish
Earnings reactionBeat, raised, but stock choppedMixed
Retail / Motley Fool tone"Forever stock"Bullish
Most damaging headlineTSMC high-NA deferralBearish (medium-term)
Most supportive headlineSK Hynix/Samsung memory ordersBullish
China revenue trend36% → 29% → 19%Bearish (transitional)
Insider activityNone reportedNeutral

News & Macro

News Report

The April-May 2026 macro tape is dominated by three forces: (1) AI infrastructure capex super-cycle, (2) export-control regime reshaping China revenue, (3) prime-customer roadmap decisions (TSMC, Samsung, SK Hynix, Intel).

AI infrastructure capex — the rising tide. Hyperscaler capex guidance running at all-time highs, ~$300B+ collective 2026. SK Hynix and Samsung confirmed large memory orders (€7.4B EUV bookings in Q4 2025, memory taking majority share for first time). TSMC capex at record ~$45B for 2026.

Geopolitics — the China cliff is here. NL export controls (Jan 2025), US export controls (Oct 2025) added metrology and software, EU-level controls (Nov 2025) extended Dutch-style restrictions across all member states. Aggregate effect: ASML's China revenue dropped from 36.1% of 2024 sales to 29.1% in 2025, to just 19% in Q1 2026. The 2026 guide bandwidth (€36-40B) explicitly accommodates "potential outcomes of ongoing discussions around export controls."

TSMC high-NA EUV deferral. April 23, 2026: TSMC publicly confirmed it will not adopt High-NA EUV (EXE:5000-class, ~$370M each) for A13. Market read this as 2-3 year deferral of the high-NA revenue ramp. ASML's response: pivot the narrative to memory.

Capacity/manufacturing reshoring. TSMC Arizona ($65B+), Samsung Texas ($17B), Intel Foundry Ohio ($28B), TSMC Japan ($16B) — 4 mega-fabs being built in parallel, all EUV-equipped. Largest single customer-funded capacity push in semiconductor history.

Macroeconomic backdrop. Fed funds at 3.75-4.00%, disinflationary glide path intact, IMF revised global growth to 3.3% for 2026.

Geopolitical tail risks. Strait of Hormuz tensions in late April, US-Iran negotiations stalled, cross-strait Taiwan rhetoric elevated. ASML's 25.5% Taiwan revenue concentration (TSMC) is the largest geopolitical risk in the portfolio.

Macro lane2026 readingNet for ASML
Hyperscaler AI capexRecord highsBullish
Foundry capex (TSMC, Samsung, Intel)All-time highsBullish
Memory capex (SK Hynix, Samsung memory)Re-accelerationBullish
US/NL/EU export controlsTightening, China at 19%Bearish (transitional)
TSMC high-NA EUV roadmapDeferred to ~2029Bearish (medium-term)
Customer reshoring (US, Korea)ActiveMild bullish (mix shift)
Fed policyEasing bias, ~3.75%Bullish
Cross-strait Taiwan riskElevated rhetoricBearish (tail risk)
US-Iran / HormuzStalled, episodicNeutral (tail)

Fundamental Analysis

Fundamentals Report

ASML's fundamental profile is that of a structurally advantaged near-monopoly with extreme pricing power being asked to grow into a $550B market cap.

Top line — re-acceleration confirmed.

YearNet Sales (€M)YoY Growth
202118,611+33%
202221,173+14%
202327,559+30%
202428,263+2.5%
202532,667+15.6%
2026E36,000-40,000+10% to +22%

Sequential Q1-Q4 2025 revenue: €7.74B → €7.69B → €7.52B → €9.72B. Q4 alone was 30% of full year.

Profitability — best-in-class.

  • 2025 GM: 52.8% (vs 51.3% in 2024)
  • 2025 OM: 34.6%, NM: 29.4%, ROE: 52.2%, ROA: 15.7%
  • TTM EPS: €30.25

Balance sheet — fortress.

  • Cash + short-term inv. (Q1 2026): €8.4B (down from €13.3B Q4 2025)
  • Total debt: €2.7B; Net cash: ~€5.7B
  • Total assets: €48.1B; Stockholders' equity: €20.8B

Cash flow. Q1 2026 OCF: -€2.19B (working capital seasonal); Q4 2025: +€11.41B. FY 2025 FCF: ~€8.2B TTM.

Capital return. Total dividend FY 2025: €7.50/share (+17% YoY). Q1 2026 buyback: €1.1B. Outstanding authorization: €12B through Dec 2028. 2025 R&D spend: €4.7B (14.4% of revenue).

Order book.

  • FY 2025 net bookings: €28.0B (vs €18.9B in 2024, +48%)
  • Q4 2025 bookings: €13.2B (€7.4B EUV alone)
  • End-Q4 2025 backlog: €38.8B (vs €35.9B end-2024)
  • Annual book-to-bill 2025: 0.86x (Q4 alone: 1.35x)

The €38.8B backlog covers ~115% of FY 2026 guide midpoint of €38B.

Customer concentration (from 2025 Annual Report):

  • Largest customer: €7,796.7M = 23.9% of 2025 sales (TSMC by inference, +€3.1B vs 2024)
  • Top 2 customers: 38.0% of 2025 sales
  • Taiwan: 25.5% (vs 15.4% in 2024)
  • South Korea: 25.0% (vs 22.7%)
  • China: 29.1% (vs 36.1%)

Single-source supplier risk. Carl Zeiss SMT is ASML's exclusive supplier of optics. The 2025 Annual Report states: "If Carl Zeiss SMT were to terminate its supply relationship with us or be unable to maintain production of optics over a prolonged period, we would effectively cease to be able to conduct our business."

Valuation snapshot at $1,427: Mkt cap $550B; TTM P/E 47.2x; Forward P/E 29.9x; PEG 2.14; Dividend yield 0.62%; Beta 1.38.

MetricValueNote
Revenue (TTM)$33.7B+15.6% YoY (2025)
Net Income (TTM)$10.0B+27% YoY
Gross Margin (2025)52.8%Best-in-class equipment
Operating Margin (2025)34.6%Software-like
Net Margin (2025)29.4%Software-like
ROE52.2%Capital-efficient monopoly
Forward P/E29.9xMid-cycle valuation
Backlog (FY 2025)€38.8B~115% of 2026 guide midpoint
FY 2026 guide€36-40BRaised from €34-39B
Top customer share23.9%TSMC; +7.3pp YoY
China revenue19% (Q1 2026)Cliff from 36% in 2024
Net cash€5.7BPlus €12B buyback authorized

Risk Factors (Item 1A)

ASML files Form 20-F (foreign private issuer); risk factors disclosed in Item 3.D / Strategic Report. Filed Feb 25, 2026. Source: ASML 2025 Annual Report (https://ourbrand.asml.com/m/6ea363f69344ebd4/original/asml-2025-annual-report-based-on-ifrs.pdf).

Top 7 most material risk factors:

  1. Customer concentration — Largest customer 23.9% of 2025 sales (€7.8B); top 2 = 38.0%. Direct quote: "We sell our lithography systems to a relatively small number of customers, making our business vulnerable to customer concentration risk. The loss of any major customer, or a significant reduction or delay in their orders, could materially impact our financial performance."
  1. Single-source critical supplier (Carl Zeiss SMT) — Sole supplier of lenses, mirrors, illuminators, collectors. Direct quote: "If Carl Zeiss SMT were to terminate its supply relationship with us or be unable to maintain production of optics over a prolonged period, we would effectively cease to be able to conduct our business." 100% dependence with no redundancy at the optics layer.
  1. Export controls / geopolitical (international operations) — China customers represented 29.1% of 2025 net sales (vs 36.1% in 2024). Subject to NL, US, EU export controls. Taiwan = 25.5% of 2025 (cross-strait risk). South Korea = 25.0%. Direct quote: "Our business is subject to a range of export control restrictions, sanctions, tariffs, and broader international trade regulations that affect our ability to deliver systems, technology, and services."
  1. Cyclicality of semiconductor industry — Capital expenditures by customers may not continue at current levels. Inflation, interest rates, geopolitical events all flow through. Industry has historically been highly cyclical.
  1. Cybersecurity — Acknowledged prior incidents. Direct quote: "We have encountered such incidents in the past. As ASML's prominence in the semiconductor industry grows, so does the likelihood of being targeted in security attacks."
  1. Technology transition (high-NA EUV adoption) — Direct quote: "the success of our EUV 0.55 NA (High NA) technology — which we view as essential to advancing Moore's Law — depends on continued technical progress by both us and our suppliers."
  1. Limited number of manufacturing facilities — Concentrated in Veldhoven, Eindhoven, Oirschot (Netherlands), Berlin, Wilton, San Diego, Pyeongtaek, Linkou, Tainan. Direct quote: "These facilities may be subject to disruption for various reasons, including work stoppages, fire, energy shortages and access issues, pandemic outbreaks, flooding, cyberattacks, blockages, sabotage or other disasters."
This document is the confidential work product of the Silent Engineering Fund's internal research process and is intended for partners and authorized personnel only. It does not constitute investment advice, an offer to sell, or a solicitation to buy any security. Past performance is not indicative of future results.
Silent Engineering Fund · Internal Research