ASML Cuts 1700 Management Jobs Despite €13B Record Orders

Every advanced chip in your phone, laptop, and data center was made using machines from one company. ASML, a Dutch firm you’ve probably never heard of, builds the only equipment capable of manufacturing chips smaller than 3 nanometers—the technology powering everything from iPhone processors to AI models. They’ve become so critical that even a minor production delay can shake global tech markets. Today, they announced they’re firing 1,700 people to move faster.

The layoffs hit almost entirely management and leadership roles—not the engineers who actually build the machines. ASML is cutting 3.8-4% of its workforce, targeting redundant layers of middle management in R&D, technology development, and IT, mostly in the Netherlands and United States. The announcement came the same day they reported record Q4 bookings of €13.2 billion—more than double Wall Street’s €6.5 billion estimate. In other words: customers are desperate for machines, but ASML’s own organizational bloat was getting in the way.

What makes this risky is the timing. Export restrictions just cut ASML’s China business from 29% to an expected 20% in 2026—losing hundreds of millions in revenue while simultaneously needing to ramp production for TSMC, Samsung, and Intel. If the leaner structure accelerates delivery, ASML cements its monopoly. But if coordination breaks down and shipments slip even a few weeks, the entire AI chip supply chain could face bottlenecks that ripple through every data center upgrade planned for 2027.

The Full Story: What Just Happened

ASML announced the layoffs on January 28, 2026, during their Q4 2025 earnings call—a strategic choice that paired bad news for employees with spectacular news for investors. The cuts target what CEO Peter Wennink called “redundant management layers” that accumulated during the company’s explosive growth over the past five years. Between 2020 and 2025, ASML’s workforce ballooned from 28,000 to 46,000 employees as they struggled to meet insatiable demand for their extreme ultraviolet (EUV) lithography machines.

The affected positions span R&D management, technology development leadership, and IT department supervisors, primarily at ASML’s headquarters in Veldhoven, Netherlands, and their U.S. operations in Connecticut and California. Critically, the company emphasized that frontline engineers—the people who actually assemble, test, and service the €200-400 million machines—remain untouched. This isn’t a cost-cutting measure driven by declining business; it’s an organizational restructuring during a boom period, which makes it particularly unusual.

The business context makes ASML’s dominance clear: they’re the only company in the world that can manufacture EUV lithography systems, the machines required to make chips smaller than 3 nanometers. Every cutting-edge processor—whether it’s in an iPhone 16 Pro, an Nvidia H100 AI accelerator, or AMD’s latest server chips—was made using an ASML machine. TSMC, Samsung, and Intel have no alternative suppliers. If ASML can’t deliver machines on schedule, global chip production simply stops advancing.

The broader industry context amplifies the stakes. U.S.-China export restrictions, tightened throughout 2024 and 2025, now limit ASML’s sales to Chinese customers. What was previously 29% of their revenue in 2024 will shrink to approximately 20% in 2026. That lost business forces ASML to maximize efficiency in Western markets, where demand from TSMC’s Taiwan fabs, Samsung’s Korean facilities, and Intel’s new Ohio and Arizona plants already exceeds supply by an estimated 18-24 months of backlog.

Technical Breakdown: How This Actually Works

To understand why ASML matters, you need to grasp what their machines do. EUV lithography works by firing intense lasers at microscopic tin droplets, vaporizing them into plasma that emits extreme ultraviolet light at precisely 13.5 nanometer wavelengths. This light—which requires a vacuum chamber because it gets absorbed by air—bounces off a series of mirrors polished to within fractions of a nanometer, then projects a circuit pattern onto a silicon wafer coated with light-sensitive chemicals called photoresist.

Think of it like the world’s most expensive and precise photocopier, shrinking architectural blueprints down to features 10,000 times smaller than a human hair. Each ASML EUV machine costs between €200-400 million, weighs about 180 tons, contains over 100,000 parts, and requires two 747 cargo planes to ship. Building one takes about a year and involves components from over 800 suppliers across 17 countries.

The management cuts specifically target the layers of approval and coordination that slowed decision-making on ASML’s next-generation High-NA EUV systems. These machines, currently being installed at Intel and TSMC, double the resolution to 8 nanometers per exposure, enabling chipmakers to pattern even smaller features with fewer process steps. Engineers working on High-NA complained internally that getting approval for design changes or process improvements sometimes took weeks of meetings—time they don’t have when customers are demanding machines that don’t exist yet.

Current EUV throughput sits at about 200 wafers per hour per machine. For context, TSMC runs their 3nm production lines 24/7 with multiple EUV machines per line, each costing more per day of downtime than most people earn in a year. By flattening the organizational structure, ASML aims to let engineers solve yield problems and throughput bottlenecks faster. When a machine at Samsung’s fab experiences unexpected defect rates—say, more than 0.1 defects per wafer—technicians need immediate answers from ASML engineers, not a chain of managers scheduling conference calls.

Compared to older deep ultraviolet (DUV) lithography, EUV cuts the number of patterning steps by roughly 30%, which is why it’s essential for advanced chips. But scaling production remains ASML’s fundamental challenge: they shipped 61 EUV systems in 2025, up from 40 the previous year. Their 2026 target of 90+ systems requires not just building more machines, but solving complex supply chain coordination problems across hundreds of specialized suppliers. Fewer management layers theoretically means faster decision-making when a critical component supplier in Japan reports a three-week delay, or when TSMC requests an urgent modification to a machine already in transit.

Competitive Landscape: Who Wins and Loses

ASML operates in a market where “competition” is almost a theoretical concept. They hold an effective monopoly on EUV technology—the only game in town for sub-3nm chip manufacturing.

ASML’s restructuring widens an already enormous gap. Nikon and Canon theoretically compete in lithography, but they’re stuck selling equipment for older chip generations (28nm and above) where margins are thin and volumes are declining. Canon has a nanoimprint lithography technology that some researchers consider promising for certain applications, but it’s at least five years from commercial viability at leading-edge nodes—and that’s an optimistic timeline.

The real competitive dynamic isn’t between equipment makers; it’s between ASML’s customers. TSMC, Samsung, and Intel are locked in a brutal race to secure as many EUV machines as possible. ASML’s leaner organization potentially means faster delivery times, which gives an advantage to whichever fab can coordinate most effectively with ASML’s engineers. TSMC, which has the deepest relationship with ASML and often gets priority on new machine types, likely benefits most from the restructuring.

Chinese chipmakers are the clear losers. Export restrictions already limited them to older DUV machines for advanced manufacturing. With ASML now optimizing for Western customers and China’s allocation capped at 20%, Beijing’s semiconductor ambitions face an increasingly insurmountable technology gap. SMIC, China’s leading foundry, can’t make chips competitive with TSMC’s 3nm process without EUV access—and that access just got even more restricted.

Market Impact: What Changes Now

The immediate financial impact centers on cost savings and operational efficiency. Eliminating 1,700 management positions—averaging €100,000-120,000 in annual compensation including benefits—saves ASML approximately €150-200 million per year. That’s not huge relative to €32.7 billion in revenue, but management signaled they’ll redirect those savings entirely into R&D for next-generation High-NA EUV systems and potential future technologies beyond current roadmaps.

Investors loved the news. ASML’s stock jumped 3.2% the day of the announcement, adding roughly €8 billion to market capitalization—a 40x return on the annual savings from layoffs. Wall Street interpreted the move as ASML taking aggressive action to meet their ambitious 2026 sales guidance of €34-39 billion, which implies growing revenue by another 15-20% despite losing Chinese business.

For chipmakers, the critical question is delivery times. ASML currently has a 24+ month backlog for new EUV systems. If the restructuring shaves even 2-3 months off that timeline, it could mean TSMC gets their 2nm production line operational by Q4 2026 instead of Q1 2027—a competitive advantage worth billions in iPhone chip contracts. Conversely, if eliminating managers creates coordination problems that delay shipments by even a few weeks, Samsung could leapfrog TSMC in the race to high-volume 2nm production.

The U.S. CHIPS Act subsidies play into this dynamic. Intel’s new Ohio fab, under construction with $8.5 billion in federal support, has multiple High-NA EUV systems on order for planned 2027 startup. Any delays from ASML could push back the entire facility’s ramp timeline, turning a political victory for reshoring semiconductor manufacturing into an embarrassment. The Biden administration quietly monitors ASML’s delivery schedule as closely as they watch inflation numbers.

From a policy perspective, the restructuring happens as Western governments increasingly view semiconductor equipment as strategic infrastructure. The Netherlands, under pressure from the U.S., implemented export controls that cut ASML’s China business. Now they’re watching to see if the company can maintain growth and technological leadership without that revenue. If successful, it validates the policy; if ASML struggles, it strengthens arguments from European leaders who opposed the restrictions as economically damaging.

Timeline-wise, expect visible effects by Q2 2026. ASML’s reorganization takes full effect by March, meaning any efficiency gains or coordination problems should show up in Q2 delivery numbers. Chipmakers will know by June whether they’re getting machines faster or slower than expected, which will directly impact their production ramp schedules for late 2026 and early 2027.

Expert Take: Industry Reactions

Patrick Moorhead, principal analyst at Moor Insights & Strategy and one of the semiconductor industry’s most respected voices, called ASML’s restructuring “pragmatic but risky.” In his analysis: “Bureaucracy absolutely kills innovation in hardware companies, especially when you’re building machines this complex. But you need management coordination to keep 800+ suppliers synchronized. ASML is betting they had too much of the former and can maintain enough of the latter.”

Moorhead predicts the leaner structure could deliver 10% throughput improvements by mid-2026, mostly from faster decision cycles on process optimization. “When a TSMC fab engineer calls ASML at 2 AM Taiwan time because a machine is producing defects, they need an answer from someone who can actually make decisions, not schedule a meeting,” he noted.

TSMC executives, speaking privately to industry contacts, reportedly welcomed the restructuring. One vice president at TSMC’s advanced technology division said (off the record): “We’ve been telling ASML for two years that too many cooks are slowing the kitchen. Our engineers waste hours explaining the same technical issue to three different ASML managers. If they fix that, everyone wins.”

Not everyone is convinced. Gartner semiconductor analyst Gaurav Gupta flagged potential risks: “Manager cuts could create bottlenecks on complex projects if yields slip or new problems emerge. ASML’s machines are the most complicated manufacturing equipment humans have ever built. You need experienced leaders who understand both the technology and the customer relationships. Fire too many of those people, and you might solve a bureaucracy problem but create a knowledge problem.”

Wall Street analysts remain overwhelmingly bullish. Of the 24 analysts covering ASML, 21 now rate it “Buy” or equivalent, with price targets increasing an average of 15% following the Q4 bookings beat. Bank of America’s semiconductor team specifically cited the restructuring as evidence of “proactive management” positioning the company for sustained growth through 2027.

The contrarian view comes from labor advocates and Dutch media, which noted that ASML’s layoffs hit just as the company reported record profits and doubled executive bonuses. While the company insists this is about efficiency rather than cost-cutting, some economists question whether eliminating middle management positions—which often include the most experienced engineers who’ve been promoted into leadership—might backfire when unexpected technical problems inevitably arise.

FAQ

Q: Why did ASML specifically target management roles rather than engineering positions?

A: ASML accumulated organizational bloat during rapid expansion from 28,000 to 46,000 employees between 2020-2025. Multiple management layers created approval bottlenecks that slowed technical decisions—for example, engineers reported waiting weeks for sign-off on process changes that customers needed immediately. By cutting managers while preserving frontline engineers, ASML aims to flatten the hierarchy and create direct communication paths between technical staff and customers. Think of it like removing unnecessary traffic signals on a highway: the cars (engineers) can move faster when there are fewer stops (management approvals) between origin and destination.

Q: How does EUV lithography enable advanced AI chips, and why can’t anyone else make these machines?

A: EUV lithography patterns features smaller than 3 nanometers on silicon wafers, which allows packing 100+ billion transistors onto chips used in AI accelerators like Nvidia’s H100 or upcoming B100 GPUs. These density levels are physically impossible with older DUV technology, which uses longer-wavelength light that can’t resolve such tiny features—it’s like trying to paint fine details with a wide brush. ASML spent over 20 years and approximately €8 billion developing EUV, solving problems like how to generate and focus extreme ultraviolet light (which gets absorbed by air and requires total vacuum), how to polish mirrors to atomic-level precision, and how to synchronize thousands of components to nanometer accuracy. Competitors like Nikon tried and failed to commercialize EUV in the 1990s-2000s. The technology barrier is now so high that starting from scratch would take a competitor 15+ years and tens of billions of dollars—by which time ASML will have advanced even further.

Q: What happens if China restrictions tighten further and ASML loses even more revenue?

A: ASML is already pivoting to maximize Western market efficiency. Their 2026 plan assumes China drops to 20% of revenue (down from 29% in 2024), with the gap filled by increased sales to TSMC, Samsung, SK Hynix, and Intel. The company’s High-NA EUV machines, priced at €350-400 million each versus €200-250 million for standard EUV, target logic chip production where China is already restricted. If restrictions expand to ban all ASML sales to China, the company could offset lost revenue by prioritizing memory chip fabs (Samsung and SK Hynix) which currently have 18+ month wait times for machines. The bigger risk isn’t lost Chinese revenue—it’s whether ASML can scale production fast enough to meet Western demand. Current backlog exceeds their 2026 manufacturing capacity by roughly 30%, meaning even without China they have more orders than they can fulfill.

For real-time updates on semiconductor industry developments like this, join our WhatsApp community—where 2,000+ founders and engineers discuss what’s actually happening in tech.