[XFAB] X-FAB: When Being Behind the Curve Is the Point

Sometimes semiconductor success means skipping the nanometer race entirely

Origins

If you're looking for a classic Silicon Valley origin story – brilliant founders in a garage dreaming of disruption – you've come to the wrong semiconductor fab. X-FAB's tale begins in 1992 among the industrial ruins of East Germany, where a Belgian entrepreneur with a penchant for football clubs and basic income philosophy acquired a communist-era chip factory during Germany's post-reunification fire sale.

That entrepreneur, Roland Duchâtelet, wasn't interested in competing with Intel or chasing Moore's Law into the nanosphere. Instead, he saw something the market had overlooked: while everyone else raced to make chips smaller and faster, entire industries still needed reliable, specialized chips for cars, factories, and medical devices.

The early DNA of X-FAB was shaped by two crucial factors. First, their inherited East German workforce brought deep technical expertise in analog and mixed-signal processes – unfashionable but essential technologies that would become X-FAB's bread and butter. Second, Duchâtelet's other company, Melexis, provided a ready-made customer for their chips. It was capitalism with training wheels, allowing X-FAB to develop capabilities without the typical startup death march for customers.

What followed was a methodical expansion that looked more like a game of Risk than a tech startup playbook. They acquired a Texas Instruments facility in Lubbock (1999), added Malaysian capacity through a merger with 1st Silicon (2006), and picked up a MEMS specialist in Germany (2012). Each move added specific capabilities while staying true to their focus on specialty processes.

The genius – or luck – of X-FAB's evolution was in recognizing that not every chip needs to be cutting edge. While titans like TSMC and Samsung fought for supremacy in the single-digit nanometer realm, X-FAB happily operated in the comfortable mid-range of 130nm to 1µm processes, where margins were healthy and competition was limited. It's the semiconductor equivalent of finding a profitable niche selling vintage vinyl in the age of digital streaming.

By 2024, that former communist factory had transformed into a global specialty foundry with six manufacturing sites across three continents. They never became a household name like Intel, but they did become something possibly more valuable: a crucial supplier of specialized chips that make modern cars, factories, and medical devices possible. It's proof that in technology, sometimes the best strategy isn't to chase the bleeding edge, but to perfect the overlooked middle ground.

The irony of a communist factory becoming a master of market segmentation isn't lost on us. But perhaps that's the perfect metaphor for X-FAB's pragmatic approach: they succeeded not by revolutionizing semiconductor technology, but by carefully choosing which semiconductor revolution to sit out.

How The Money Actually Works

Here's how X-FAB actually makes money: They take a dinner plate-sized silicon wafer worth about $50, add about $300 in processing costs across 400 steps, and sell it for $500 to automotive companies who will use it to make chips that control your car's brakes. They'll repeat this process about 600,000 times this year. What makes this profitable isn't technological wizardry – it's their willingness to keep running processes that bigger manufacturers consider passé, but carmakers consider essential. Think of it as a semiconductor version of the guy who maintains vintage Porsche engines: not the most advanced technology, but try finding someone else who can do it reliably.

Their 2023 revenue of $907 million flows primarily from three sources:

1. Automotive chips (59% of revenue) - everything from brake controls to power management

2. Industrial applications (23%) - factory automation and power systems

3. Medical devices (7%) - where reliability matters more than speed

The simple truth about X-FAB's business: They make chips using older technology that nobody else wants to maintain anymore,
but that cars, factories, and medical devices still desperately need.

The cost structure follows the remorseless logic of semiconductor manufacturing: high fixed costs in equipment and facilities, significant energy consumption (those plasma etchers don't run on goodwill), and the constant need for skilled labor. Yet X-FAB achieves 28.5% gross margins by focusing on mature processes where the equipment is largely depreciated and the yields are predictably high. Their Texas facility still produces profitable chips on 150mm wafers, a format TSMC relegated to museums sometime during the Clinton administration.

The sustainability of this model depends on X-FAB maintaining their position in specialized analog and mixed-signal processes. Their customers commit to long-term agreements not out of loyalty, but because redesigning a power management chip for a different manufacturing process costs more than several years' worth of wafer purchases. When your product controls the braking system in a luxury car, price sensitivity becomes a theoretical concept.

This explains why X-FAB can generate consistent cash flow from technology nodes that leading-edge manufacturers discarded years ago. They're not selling commoditized manufacturing capacity; they're selling the certainty that comes from decades of process refinement.

The Numbers That Tell The Story

Forget nanometers. X-FAB's story boils down to three numbers that actually matter.

First, there's the tale of two production lines. Their newer factories process 200mm wafers - discs of silicon that pack in more chips per production run and use more modern equipment. These lines are running flat out because they're perfect for today's automotive and industrial chips - modern enough to handle complex designs, but mature enough to be reliable and cost-effective. Meanwhile, their older facilities work with 150mm wafers - an aging format that's like the floppy disk of semiconductor manufacturing. These lines sit half-empty because, while some legacy customers still need them, most new designs target the more efficient 200mm format. The industry moved on, but X-FAB keeps these older lines running because some loyal customers still depend on them, and redesigning old chips for newer processes would cost more than several German luxury cars.

Then there's the number that would give any MBA professor heartburn: 45% of X-FAB's revenue comes from a single customer, Melexis. Having nearly half your business depend on one customer is typically a recipe for disaster – it gives that customer enormous pricing power and leaves you vulnerable if they switch suppliers or face difficulties of their own. But as we said, Melexis and X-FAB share the same major shareholder. It's less a supplier-customer relationship and more a carefully orchestrated industrial partnership, though one that leaves minority shareholders hoping the family stays happy.

Finally, there's the metric that shows who's actually buying what X-FAB is selling. The "book-to-bill ratio" – simply the ratio of new orders to actual sales – tells a fascinating story. For their traditional business, they're getting $1.05 in new orders for every $1 of current sales. But in their new silicon carbide chip business (crucial for electric vehicles), they're only getting 38 cents in new orders for every dollar of sales. It's a number that suggests their core business remains robust while their venture into next-generation technology faces headwinds.

Together, these numbers paint a picture of a company that's both incredibly stable and surprisingly precarious. Their $481 million backlog (plus $114 million in unscheduled orders) shows sustained demand for their traditional offerings. But the real question is whether they can maintain their position as the master of mature technology while successfully expanding into new territories.

Some context on silicon carbide: It's the semiconductor industry's equivalent of a better battery – chips that can handle higher voltages and temperatures than traditional silicon, perfect for electric vehicles. X-FAB bet they could master this complex process and ride the EV wave. They've made technical progress (their new process offers 30% more chips per wafer), but they're caught in a classic semiconductor trap: They need large volumes to make the economics work, but customers won't commit volumes until they see better economics. That's why they just abandoned their target of $300-350M in silicon carbide revenue by 2026 – being pretty good at an emerging technology isn't enough when giants like Infineon are pouring billions into the same space.

A few telling metrics that round out the X-FAB story:

• A $1 billion capacity expansion program for 2023-2025, roughly equal to their entire market cap. It's like betting the farm, if the farm produced silicon wafers and had a German accent.

• 4,500 employees representing 45 nationalities, which for a company based in Erfurt, Germany (population 213,000) suggests they've managed to make eastern Germany surprisingly cosmopolitan.

• Revenue per employee of $201,511 (2023), significantly lower than TSMC's $730,000 but logical for a specialty manufacturer focusing on complex, lower-volume runs. Sometimes being artisanal means being less efficient.

• 458 patents in their portfolio - modest by semiconductor standards, but focused almost entirely on analog and mixed-signal processes. It's not how many patents you have, it's whether they cover the things your customers can't live without.

• Zero memory or digital logic products. While their competitors play Moore's Law limbo, X-FAB has thrived by simply refusing to join the race.

The Moat (if any)

To understand X-FAB's competitive position, forget everything you know about the semiconductor industry's relentless race to the bleeding edge. X-FAB operates in a different world entirely – one where stability trumps speed, and being boring is actually a competitive advantage.

While competitors could technically manufacture similar chips, X-FAB has built quite the moat: As CEO Rudi De Winter noted in their Q3 2024 earnings call, they are "the sole supplier for >90% of products manufactured." Their customers are so focused on securing this capacity that they've paid $275.9M in advance deposits – essentially lending X-FAB money for free just to guarantee their spot in line.

Who are they really competing with?

On paper, X-FAB competes with GlobalFoundries, Tower Semiconductor (now part of Intel), and UMC in the specialty foundry space. But that's like saying a craft bourbon distillery competes with Diageo. For 90% of their products, X-FAB faces no direct competition at all. They're not competing with other manufacturers; they're competing with their customers' fear of obsolescence.

In semiconductor manufacturing, switching suppliers isn't just expensive – it requires completely redesigning the product, requalifying it for automotive or medical use (a multi-year process), and betting your company that nothing goes wrong. Most customers would rather pay a premium than take that risk.

X-FAB has built their moat not through technological superiority, but through mastery of processes others consider obsolete. They still run 150mm wafer lines (ancient by industry standards) because some customers' products depend on them. More importantly, they've perfected the art of analog and mixed-signal manufacturing – the black magic of turning real-world signals into digital data and back again.

Customers stay for three reasons:

1. Supply security (they don't abandon old processes)

2. Automotive/medical qualifications (which take years to obtain)

3. Process stability (crucial for analog chips)

Despite their strong position, X-FAB faces real risks:

• Customer concentration (45% of revenue from one customer)

• High fixed costs requiring constant high utilization

• The transition to new technologies (their silicon carbide struggles)

• Chinese competition in more commoditized processes

The moat appears remarkably durable. Their customers have essentially locked themselves into X-FAB's manufacturing processes through their own product designs. As long as cars need power management chips and medical devices need precise analog sensors, X-FAB's position looks secure.

The real question isn't whether X-FAB can maintain their current position – it's whether they can replicate their specialty manufacturing success in new technologies like silicon carbide. Their recent struggles in this area suggest that building new moats may be harder than maintaining old ones.

While TSMC and Samsung race to build tomorrow's chips, X-FAB profits by keeping yesterday's cars running. X-FAB has turned that unglamorous role into a profitable niche, proving that in technology, sometimes the best position isn't at the cutting edge, but at the reliable center.

The World vs. X-FAB

X-FAB's world is splitting into three competing realities.

1. In China, aggressive EV adoption drives 55% year-over-year growth in their automotive business.

2. In Europe, their traditional customers want them to keep running decades-old processes that control luxury car brakes.

3. Meanwhile, geopolitical tensions turn their global manufacturing footprint from an efficiency play into a strategic imperative.

Their Q3 2024 results read like a geopolitical briefing. Chinese automotive revenue surges while European and American orders weaken. Their Malaysian fab becomes not just a cost advantage but a hedge against European energy prices. Even their struggles in silicon carbide (revenue down 60% year-over-year) tell a larger story about betting on global transitions that move at regional speeds.

X-FAB faces a €550M question: how much to invest in new technologies when their competitive advantage lies in mastering mature ones. Their traditional automotive and industrial customers want stability and continuity – keep those proven processes running, don't change what works. Meanwhile, the electric vehicle transition demands new technologies and massive capital investment, just as geopolitical tensions make such investments riskier.

In Eastern Germany, where their headquarters sit, X-FAB faces another macro challenge: competing for semiconductor talent in a region not exactly known as a tech hub. They've managed to attract 4,500 employees from 45 nationalities to places like Erfurt (population 213,000), suggesting they've found ways to turn geographical challenges into recruitment advantages.

Here's the macro reality for X-FAB: They're a European company with American and Asian operations, making chips for Chinese EVs and German luxury cars, using processes that are simultaneously obsolete and irreplaceable. They need to maintain expensive fabs in high-energy-cost regions while competing with Chinese manufacturers in lower-cost regions.

Yet this seemingly awkward position might be exactly where you want to be in a fragmenting world. When automotive and industrial customers worry about supply chain security, X-FAB's mastery of mature processes becomes strategically valuable. When governments push for regional semiconductor capacity, X-FAB's existing geographic footprint becomes a competitive advantage.

The world seems to be moving away from the type of globalization that produced semiconductor giants like TSMC. In a world of friend-shoring and regional manufacturing priorities, being a medium-sized specialty manufacturer with fabs in three continents might be the new smart play. Sometimes, the macro forces that look most threatening on paper create the most interesting opportunities in practice.

For X-FAB, the challenge isn't surviving in a hostile macro environment – it's managing the contradictions of a world that simultaneously wants the stability of the old and the promises of the new. Their ability to balance these opposing forces while maintaining their special sauce in analog and mixed-signal manufacturing will determine whether they emerge stronger from this period of global transformation.

🐻 Why Smart Money Worries

If semiconductor manufacturing were a game of chess, X-FAB has been playing a clever gambit: sacrificing the pursuit of cutting-edge nodes to dominate the mature technology middlegame. But their Q3 2024 earnings call reads less like a grandmaster's triumph and more like someone realizing they've run out of safe moves.

Let's start with the elephant in the clean room: X-FAB's silicon carbide business just posted a 60% year-over-year revenue decline and a book-to-bill ratio of 0.38 (Q3 2024 earnings). Management has officially abandoned their 2026 target of $300-350M in SiC revenue – a target that, until recently, was their key growth narrative.

The problem isn't just cyclical. As CEO Rudi De Winter explained in Q3:

"The silicon carbide capacity expansions at our Texas fab, which has been slowed in line with current market demand, will be resumed as soon as the silicon market recovers."

Translation: We're caught in semiconductor manufacturing's version of a catch-22 – we need scale to be competitive, but we can't achieve scale without being competitive first.

While every other semiconductor manufacturer is battening down the hatches – even TSMC cut their capex – X-FAB is doubling down with a €550M spending spree for 2024. Two-thirds of Q3's €150M went into their Malaysian expansion, prompting Deutsche Bank's Robert Sanders to ask the question on everyone's mind: "Why are you not slamming on the brakes? Every one of your peers is slamming on the brakes on CapEx, but you continue to have the same plan, which is very strange."

The answer? They're "really pushing the throttle" for 180nm capacity. It's either brilliant contrarian thinking or the semiconductor equivalent of building luxury hotels in 2007.

Here's where it gets interesting: Their 200mm lines are running at 100% utilization (Q1 2024 transcript), while their 150mm lines languish at 50%. This isn't just inefficient – it's a structural problem masquerading as a cyclical one. As CEO De Winter admitted in Q1: "The 150-millimeter lines are relatively -- I think the loading is in the range of 50%. So it's really low."

Remember when having one customer provide 34% of your revenue was concerning? Well, Melexis now accounts for 45% of revenue (FY 2023 report). Yes, they share a major shareholder, which is either comforting or concerning depending on how you feel about Belgian family business dynamics.

The regional numbers tell a story that would make a geopolitical strategist wince:

• China automotive revenue: +55% YoY

• EU/US automotive: "weakening" (Q3 2024)

• Malaysian fab: Three-day production slowdown due to "an impurity in a chemical system"

They're increasingly dependent on Chinese growth just as the world is rethinking semiconductor supply chains.

The numbers folks should be losing sleep over:

• Inventory days increased to 136 from 128 (FY 2023)

• Cash conversion cycle stretched to 143 days

• Receivables up 101.2% YoY
  Think of receivables as IOUs from customers. When they double in a year, it's like your salary doubled but your employer started paying you two months late instead of one. In X-FAB's case, they went from €73.12M to €147M in receivables (FY2023 report). Industry standard is to get paid in 45-60 days.

• Working capital efficiency deteriorating

X-FAB's bear case isn't about imminent doom – it's about death by a thousand cuts:

1. Their silicon carbide pivot is faltering just as mature technologies face pricing pressure

2. They're spending like it's 2021 while their peers prepare for winter

3. Their geographic diversification is becoming a liability rather than an asset

   1. High-energy-cost fabs in Germany (where electricity prices tripled post-Ukraine war)

   2. A Malaysian fab that just had a three-day shutdown over chemical impurities

   3. A Texas fab running at 50% utilization

   4. Growing dependence on Chinese automotive customers (+55% YoY) just as the U.S. and EU impose restrictions on semiconductor trade with China

4. Their customer concentration risk is increasing, not decreasing

Or as Michael Roeg from Degroof Petercam delicately asked in Q3: "Do you think that some of these customers are holding off on ordering, especially because they think that next year they can get lower prices than in Q4?".

He was politely pointing out a nasty trap:

• Q4 is traditionally when semiconductor companies do inventory corrections

• Customers know silicon carbide wafer prices have fallen

• X-FAB's 150mm lines are running at 50% utilization

• Their major expansion won't be ready until Q1 2025

The market's -50% YTD verdict suggests investors are starting to wonder if X-FAB's clever middle-market strategy has an expiration date. In semiconductors, being stuck in the middle was profitable when the industry was expanding in both directions. But in a world of friend-shoring, EV transitions, and regional semiconductor strategies, the middle might be exactly where you don't want to be.

As one analyst summarized in the Q3 call: "You're the only company I covered that is spending more CapEx this year than last in your end markets." Sometimes the bear case writes itself.

🐂 Why Smart Money Buys

Sometimes the most interesting investments are found in the gap between what the market thinks it sees and what's actually happening.

X-FAB's stock price has been cut in half this year, suggesting investors see a company trapped in the semiconductor industry's awkward middle age. But what if being stuck in the middle is actually the point?

Let's start with a heretical thought: What if not chasing Moore's Law is actually the smart play? While TSMC spends $32 billion on 2nm technology and Intel burns cash trying to catch up, X-FAB's customers are paying them $275.9 million in advance deposits just to guarantee access to technology nodes that were considered mature when Friends was still running new episodes.

This isn't just customer loyalty – it's customer captivity. CEO Rudi De Winter casually dropped a bomb in their 2023 report: they're "the sole supplier for >90% of products manufactured."

Yes, they're spending €550M on capex while everyone else retreats. But here's the thing: their 200mm lines are running at 100% utilization (Q1 2024 transcript). When Deutsche Bank's Robert Sanders asked why they weren't "slamming on the brakes" like everyone else, he might have been asking exactly the wrong question.

Consider:

• Their 180nm technology is in allocation (demand exceeds supply)

• The Malaysian expansion comes online in Q1 2025

• Customers are literally paying them in advance to build this capacity

The market sees their 55% YoY growth in Chinese automotive as a risk. But flip the script: What if it's actually evidence they're on the right side of the biggest transformation in automotive history? China's EV transition isn't just about Tesla knockoffs – it's about a fundamental shift in semiconductor demand.

Their Q3 transcript reveals something fascinating:

"China automotive business has grown with 55% year-on-year and 38% quarter-on-quarter while EU and US automotive bookings weakened."

That's not a bug – it's a feature. They're growing fastest in precisely the market that's leading the electric vehicle revolution.

Yes, their silicon carbide revenue dropped 60% YoY. But buried in the Q3 transcript is this little gem: their next-generation technology offers "improved design performance and a 30% increase in dice per wafer." Combined with cratering substrate prices, they're looking at a "potential of 40% cost improvement for the final silicon carbide devices."

Translation: They're using the downturn to perfect a technology that could be 40% cheaper just as the market recovers. That's not a failed pivot – that's playing the long game.

The bear case sees the 45% revenue concentration in Melexis as a risk. But this Belgian family business arrangement might actually be brilliant: They've turned their biggest customer into their biggest advocate. Melexis has every incentive to help X-FAB succeed, and their shared ownership means their interests are perfectly aligned.

While everyone fixates on the -50% YTD stock performance, consider:

• 28.5% gross margins (FY2023)

• $907M revenue (22.6% growth)

• Trading at 6.28x earnings

• $405.7M cash position

For a company that's supposedly in strategic trouble, those numbers look suspiciously healthy.

The smart money bet on X-FAB isn't about believing they'll suddenly become TSMC. It's about understanding that in a world obsessed with semiconductor sovereignty and supply chain security, being a Western-owned specialty foundry with:

• Fabs in three continents

• 90% sole-source products

• Deep automotive/industrial relationships

• Proven ability to run mature nodes profitably

...might actually be exactly what the next decade needs.

The bear case sees a company caught in the middle. The bull case sees a company that's built an entire business model around being precisely where their customers need them to be. In a world where semiconductor manufacturing is becoming as much about geopolitics as technology, X-FAB's "mature technology + global footprint + automotive/industrial focus" might be the new clever play.

As their CEO noted in Q3:

"The fundamental drives for X-FAB business remain intact. These include the growth of semiconductor content in cars and the electrification of everything to drive the decarbonization of our world."

At 6.28x earnings, you're not paying for that thesis to be perfect. You're just paying for it to be less wrong than the market thinks.

Life Cycle Reality Check

If X-FAB were a restaurant, it would be that impossibly old bistro that somehow survives downtown's trendiest district – not because it chases the latest food fads, but because it's mastered the art of serving comfort food to regulars who've long forgotten they have other options.

Their core business shows all the classic marks of maturity: 90% sole-supplier status, customers paying $275.9M in advance deposits just to guarantee capacity, and the kind of steady margins (28.5%) that make private equity folks salivate. This is a business that's achieved what Damodaran calls "competitive equilibrium" – they've found their niche and fortified it with moats made of customer switching costs.

Yet their financials tell a split story. That 22.6% revenue growth and €550M capex splurge (60.6% of revenue) are straight from the growth stage playbook. Their 200mm production lines are running at 100% capacity while they're pouring money into Malaysian expansion – hardly the moves of a company settling into comfortable maturity.

The lifecycle truth? X-FAB is actually two companies. Their mature analog/mixed-signal business generates the cash, while their attempts at silicon carbide and EV chips burn it in pursuit of growth. It's like watching a sensible banker moonlight as a venture capitalist.

For investors, this dual lifecycle position creates both opportunity and risk.

The next stage isn't about choosing between maturity and growth – it's about managing their Jekyll-and-Hyde reality. Can they maintain their profitable maturity while funding their growth ambitions? At 6.28x earnings, the market's betting they'll trip on this tightrope. But then again, the market once thought selling vintage vinyl was a dead business too.

The Bottom Line

X-FAB embodies the quiet genius of mastering the mundane. They've built a billion-dollar business by perfecting semiconductor processes others consider passé, turning "mature technology" from a polite industry dismissal into a €907M revenue stream.

"Sometimes the best position isn't at the cutting edge, but at the reliable center."
- Rudi De Winter, CEO

The core reality packs a beautiful irony: Their 28.5% margins flow from manufacturing chips that control luxury car brakes using processes that predate the iPhone. This isn't nostalgia – it's monopoly by museology, and their customers are paying €275.9M in advance deposits for the privilege.

The opportunity lurks in semiconductor manufacturing's blind spot. While TSMC and Samsung wage nanometer wars, X-FAB quietly collects toll payments on technological bridges that can't be demolished – the automotive designs, industrial controls, and medical devices that keep our current world humming.

The risk? Physics doesn't care about profit margins. Eventually, even German engineering will bow to the economics of newer processes. X-FAB's €550M expansion bet assumes their customers' fear of change will outlast their desire for efficiency.

What has to go right? Their Malaysian expansion needs to deliver. Their silicon carbide pivot needs to find its footing. Most crucially, they need to keep proving that in semiconductors, like in fine wine, age can command a premium.

For investors, X-FAB presents a semiconductor thesis wrapped in a behavioral economics puzzle: At 6.28x earnings, you're buying their ability to keep turning technological inertia into economic moats. It's a bet that in our rush to build tomorrow's world, we'll still need someone to maintain today's.

Think of X-FAB as semiconductor craftsmen in an industry of mass producers. While others chase the bleeding edge, they're banking on the fact that bleeding edges eventually need bandages – preferably ones controlled by trusted, time-tested chips.

---

LEGAL DISCLAIMER AND DISCLOSURE

© 2025 Silba. All rights reserved.

This analysis is for informational purposes only. The content represents the views and opinions of the author, based on publicly available information. While we strive for accuracy, we make no representations or warranties regarding the completeness or accuracy of the information presented.

This analysis is not investment advice. It is not a recommendation to buy, sell, or hold any security. The author may hold positions in securities mentioned. Future performance of any investment discussed cannot be guaranteed.

All financial figures, unless otherwise noted, are approximations based on publicly available information. Readers should verify all claims, statistics, and analytical processes independently. The information contained herein is not intended to be a complete analysis of any security, company, or industry mentioned.

Market conditions change. Past performance is not indicative of future results. Investments mentioned may not be suitable for all investors. Readers should conduct their own due diligence and consult with professional financial, legal, and tax advisors before making any investment decisions.

By accessing this analysis, you agree that neither Silba nor its affiliates bear any responsibility for investment decisions you make based on this information.

---