Why I'm Not Selling ASML After a 230% Gain

ASML is framed as a high-quality long-term compounder, with gross margin rising from 48.6% in 2020 to 52.8% in 2025 and revenue/EPS CAGR of 18% and 24% over that period. The article argues that AI-driven demand for EUV lithography supports analyst expectations for 2025-2028 revenue and EPS CAGRs of 18% and 27%, though the stock already trades at 39x next year's earnings. Overall message: constructive on fundamentals, but valuation is a clear constraint.

Analysis

The key market signal here is not ASML’s absolute growth, but the re-rating of its earnings durability. When a single capital-goods supplier becomes a bottleneck to frontier compute, the business starts trading less like cyclical semi equipment and more like a toll road on AI capex, which helps explain why multiple compression has stayed absent even with a premium multiple. That said, the market may be underestimating how much of the next leg depends on high-NA adoption timing; if customers stretch qualification cycles, near-term upside can be pushed out even if the long-term moat remains intact.

Second-order beneficiaries are the foundries and AI silicon designers that can actually translate ASML capacity into shipped wafers. TSM benefits if leading-edge node demand stays tight, but it also absorbs a larger share of the execution burden and capex intensity, so margin expansion may lag the narrative. NVDA remains an indirect winner because the compute race sustains leading-edge wafer demand, yet any delay in lithography deployment creates a bottleneck that can cap GPU supply growth before end-demand does.

The main contrarian risk is that investors are paying for a smooth multi-year ramp when the path is likely lumpy. ASML’s order flow is vulnerable to customer digestion, export controls, and any pause in AI capex by hyperscalers or memory makers; those are months-to-quarters risks, not days. If AI spending normalizes from “scarcity” to “optimization,” the multiple can de-rate even if revenue still grows mid-teens.

The consensus is probably right on quality but too complacent on timing. The premium is justified only if backlog converts without meaningful slippage and if high-NA transitions begin to contribute before low-NA demand matures. In other words, the stock is less about whether ASML wins and more about whether the market is willing to keep underwriting a very long-duration monopoly at a growth-stock multiple through a potentially uneven capex cycle.

Key Decisions for Investors

• Stay long ASML, but prefer staged entry or add-on only on 7-10% pullbacks; upside is durable over 12-24 months, but downside from capex digestion can be 15-20% if orders decelerate.
• Pair long ASML / short a basket of lower-quality semi equipment names with weaker pricing power; ASML should hold premium valuation longer if AI lithography remains the bottleneck, while cyclical peers should re-rate down first.
• For directional exposure to frontier-node demand, use a call spread in ASML 6-12 months out rather than outright stock; this captures continued multiple support while limiting damage if high-NA adoption slips.
• Overweight TSM only as a relative-value trade versus semiconductor cyclicals, not as a clean beneficiary; its return profile is more execution-sensitive and less asymmetrically attractive than ASML over the next 1-2 years.
• Monitor any slowdown in ASML order growth or customer capex commentary as the earliest reversal signal; if that appears, reduce exposure quickly because the de-rating can happen before revenue actually rolls over.