Have you ever watched a stock absolutely crush the market for an entire year and still wondered if the story could get even better? That’s exactly where I find myself with ASML right now.
Up roughly 60% year-to-date as I write this in early December 2025, the Dutch giant behind extreme ultraviolet lithography machines has been one of the hottest tickets in tech. Yet one of the sharpest analyst teams on the Street just stepped up and said, essentially, “We’re not done yet—this is still our favorite name heading into 2026.”
And when they bumped their price target from $1,134 all the way to $1,331—implying another 20% upside from here—you sit up and pay attention.
The Big Thesis in One Sentence
After almost two years of investor worry that “lithography intensity” was going to collapse and hurt ASML’s pricing power, the data is finally flipping the script—and the market hasn’t fully priced it in yet.
That single idea is the core of why Bank of America believes ASML is transitioning from a “wafer-fab-equipment-minus” story (growing slower than the overall market) to a genuine “wafer-fab-equipment-plus” story capable of outrunning the sector for the rest of the decade.
First, What on Earth Is “Lithography Intensity”?
Think of lithography as the ultra-precise printing press for chips. Every time a chipmaker like TSMC, Samsung, or Intel wants to shrink transistors and make faster or more efficient processors, they need more (and more expensive) lithography steps.
Lithography intensity is simply the percentage of total factory equipment spending that goes toward lithography tools. For years that number climbed relentlessly higher because EUV (extreme ultraviolet) machines cost hundreds of millions of dollars each and are basically mandatory for cutting-edge nodes.
Then 2023–2024 happened. Memory prices crashed, leading-edge logic spending got delayed, and suddenly everyone worried the intensity trend had peaked forever.
Turns out that fear was overblown.
The DRAM Inflection Nobody Saw Coming
Here’s the part that gets me genuinely excited.
While most investors focused on logic chips (think Nvidia GPUs and smartphone processors), the memory side—especially DRAM—has been quietly preparing for a massive EUV ramp.
Right now ASML’s share of the DRAM equipment wallet is still modest. Analysts see that number climbing toward 26% by 2028, which would be an enormous mix shift for a company that historically made most of its money from logic foundries.
- Samsung is regaining competitiveness and ordering EUV tools aggressively
- Micron is finally committing capital to EUV layers after years of hesitation
- SK hynix never really stopped
Add it all up and the memory trio could drive a meaningful reacceleration in lithography spending starting as early as the second half of 2026.
Customer Concentration Risk Is Melting Away
For years the bear case on ASML boiled down to two words: TSMC dependence.
At the peak, Taiwan Semiconductor accounted for roughly 40–45% of ASML’s EUV shipments. Any hiccup in TSMC’s capex plans sent chills through Eindhoven.
Fast-forward to today and the picture looks radically different:
- Intel is stabilizing its foundry ambitions and placing real EUV orders again
- Samsung is spending to close the gap with TSMC
- AI accelerator companies are moving to angstrom-class nodes earlier than expected
- Micron and SK hynix are layering in EUV for high-bandwidth memory (HBM)
Suddenly the customer base feels diversified in a way it hasn’t in half a decade. That’s a big de-risking event that rarely gets the attention it deserves.
The China Question—Finally Normalizing
Let’s not sugar-coat it: China exposure has been the elephant in the room since 2019 export restrictions started tightening.
At one point China accounted for well over 40% of ASML sales (mostly older DUV tools). Investors feared a cliff if Washington slammed the door completely.
The latest guidance suggests China will settle in the low-to-mid 20% range going forward—painful compared to the old days, but perfectly manageable and, crucially, predictable.
“We think the narrative is about to turn for the better.”
— Bank of America semiconductor team, Dec 2025
When the scariest geopolitical variable in your model starts becoming boring, that’s usually great news for the stock.
Gross Margin Expansion—The Gift That Keeps Giving
One of the more under-appreciated angles is how the mix shift actually helps profitability.
EUV systems already carry fatter margins than legacy DUV tools. More EUV layers per wafer, more high-NA EUV machines (which cost north of $400 million each), and a growing service/annuity stream from an installed base that keeps ballooning—those are powerful margin tailwinds.
Bank of America now expects gross margins to expand again in 2027 and keep climbing toward the end of the decade. That’s the kind of operating leverage that turns good revenue growth into spectacular earnings growth.
In plain English: even if wafer fab equipment spending grows “only” mid-to-high single digits, ASML’s earnings could compound at close to 18% annually through 2030. That’s the definition of quality compounding.
Valuation—Still Reasonable for What You’re Getting
Yes, ASML trades at roughly 34× forward earnings as I write this. That’s not cheap in an absolute sense.
But stack it up against almost any other company with a durable monopoly in a market growing teens percentages annually, and suddenly it looks reasonable—especially when you layer in 18% EPS growth and mid-40s margins.
Or think about it this way: the company essentially owns the toll bridge for every leading-edge chip made on planet Earth for the foreseeable future. How much would you pay to own that bridge?
Risks? Of Course There Are Always Risks
I’d be doing you a disservice if I painted this as a risk-free moonshot.
- Further tightening of export rules remains the wild card
- Any meaningful delay in high-NA EUV adoption would hurt sentiment
- Cyclicality is still baked into the DNA of semiconductor equipment
- Competition in older nodes (mostly from China) could pressure DUV pricing
But here’s the thing: most of these risks have been debated to death for years, and the stock has already priced in a decent chunk of them. The new catalysts—DRAM share gains, margin expansion, customer diversification—are fresher and, in my view, under-appreciated.
The Bottom Line
Sometimes the best investments aren’t the ones nobody has heard of. Sometimes they’re the ones everyone knows but still underestimates how long and how far the runway really is.
ASML feels like that name right now.
With lithography intensity inflecting higher, customer concentration easing, China normalizing, and gross margins set to march upward again, the setup into 2026 and beyond looks genuinely compelling.
If Bank of America is even half right about that 18% earnings CAGR through the end of the decade, owning shares today could look like an absolute gift in hindsight.
That’s why, for the first time in a long time, I’m comfortable saying ASML remains one of the very few semiconductor names I’d be happy to own and hold through the entire cycle—no matter how noisy it gets along the way.
