X-Fab Silicon Foundris Se

Hello and welcome to XFAB second quarter 2025 results conference call. On today's call we have Rudy De Winter, CEO and Alba Morganti, CFO. Please note this call is being recorded and for the duration of the call your lines will be on listen only. However, you will have the opportunity to ask questions at the end of the call. This can be done by pressing star 1 on your telephone keypad to register your question. And if you require assistance at any point, please press star zero and you will be connected to an operator. I will now hand over to your host, Rudy De Winter, to begin today's conference. Thank you.

Thank you very much. Welcome to the XFAP second quarter 2025 results call. In the second quarter, we recorded revenues of $215 million, up 5% year-on-year and up 5% quarter-on-quarter. Excluding the FRS-15 impact, the quarterly revenue was $218 million, which is well above the guidance of $200 to $210 million. The second quarter revenue in our core markets, automotive, industrial, medical, was $206 million, up 8% year-on-year and up 9% quarter-on-quarter. This is the highest quarter in the last six quarters and is now growing for two quarters in a row. 225 is progressing more favorably than initially anticipated, and we upgraded our full-year revenue guidance to $840 to $870 million. The order intake has increased for two consecutive quarters now. Our business is no longer constrained by capacity, and factory cycle times have shortened significantly. And customers place orders later than usual and more frequently and at short notice, resulting in a reduced visibility. The shorter ordering behavior is also a result of the geopolitic uncertainties where everybody is more careful. All in all, the backlog in the second quarter increased by 26 million to 413 million. Now, when looking into our different end markets, starting with automotive, the revenue totaled $143 million, up 1% year-on-year and up 6% sequentially, mainly driven by EV-related applications. The inventory levels is a mixed picture. On average, it's still on the high side, and at the same time, we see products with rush orders indicating that the industry is conservative in ordering. Industrial revenue came in at $47 million, recording a very strong growth of 37% year-on-year and 20% quarter-on-quarter. The highly fragmented industrial end market is picking up again. Our industrial business also benefited from increased demand following the last time buy, announcement for some of the 150 mm CMOS technologies. This will continue to be strong throughout 2026, and the industrial also benefited from new business generating revenue by prototyping new projects. Ixfalt Medical Business recorded a quarterly revenue of 15 million, up 14% year-on-year, and 9% quarter-on-quarter. The growth in the second quarter was mainly driven by medical-grade contactless temperature sensors using microsystems technology. In the second quarter, all technologies showed progress compared to the previous quarter. The CMOS progressed thanks to additional capacity that became available and also the strong demand for EV-type products. The microsystem progressed well and also had good bookings. The silicon carbide is now clearly progressing. The gradual recovery of XFAB silicon carbide business is not fully reflected in the evolution of the toplines. Sequentially, the silicon carbide revenue rose 32%, while the number of silicon carbide wafers produced grew by more than 60% quarter on quarter. This is due to the greater portion of silicon carbide consigned wafers. During the first half of 2025, XFAP started production of more silicon carbide wafers in its factory in Texas, than it did throughout all of 2024, primarily due to demand from data center applications, and this sets us up for further growth in the second half year. On the silicon carbide business development side, it's also progressing well. Our latest Technology platform that we released last December is driving strong engineering and prototyping activity, and the customers are very pleased with the results that they achieve. Quarterly prototyping revenue was $21 million, down 1% year-on-year and up 30% quarter-on-quarter. The achievement of key milestones in customer-specific microsystems projects has contributed to that. Now, let me update you on the operations side. The end of the second quarter marks the completion of XFAP's three-year program to expand manufacturing capacity across the group. Main focus in the first half of the year was on equipping the new clean room in Cochin in Malaysia. All equipment has been delivered and is at different stages of installation and qualification. A number of tools and machines have already completed qualification allowing for phased ramp-up of our 180 nanometer technologies where we have particularly strong demand. and we expect further growth in the third quarter. Capacity in the second quarter amounted to, sorry, CapEx in the second quarter amounted to 54 million, a clear decline from the previous quarter. For the first half of 2025, the CapEx totaled 155 million, coming in slightly lower than expected due to the deferral of some CAPEX into the second half of this year. Due to this shift, we might see a small increase in the third quarter to drop further in the fourth quarter. The full-year capital expenditure projection remains unchanged at $250 million. And now I would like to pass the word to Alba for the finance.

Thank you, Rudy. Good evening, ladies and gentlemen. We will now go to the financial update. I would like to start by highlighting that in the second quarter, we succeeded to increase quarter on quarter our gross profit by 12%, and our EBDA grew by more than 5%. Our revenue increased as well by 5%. totalizing $215 million, which was in the upper part of our guidance of $200 to $210 million, and at the highest since more than a year. More specifically, our second quarter came in with $51.6 million EBDA, with an EBDA margin of 24%. If we exclude the impact from revenue recognized over time, the EBITDA margin for the second quarter would have been 24.3% within the guided range of 22.5% to 25.5%. Our profitability remains unaffected by exchange rate fluctuation because our business is naturally hedged still. at a constant US dollar-euro exchange rate of 1.08 as experienced in the previous year's quarter, the EBITDA margin would have been at the same level. As said, from a business perspective, we are fully hedged. But if you look at our financial results in the second quarter, you will see that we recorded a loss of $17.4 million. which includes an unrealized foreign exchange effect for $17.2 million, primarily related to the reevaluation of Euro-denominated debts. But it's still unrealized at this stage, so it's not a cash item. Cash and cash equivalents at the end of the second quarter amounted to $157.7 million. nearly unchanged from the previous quarter. As Rudy already explained, our capex payment went significantly down in the first half of 25, totalizing $156 million, compared to $510 million in the full year 24. And we should then stay around $250 million for the full year this year. And to conclude this financial section, I would like to share our new quarters guidance. Our Q3 2025 revenue is expected to come in within a range of $215 to $225 million with an EBITDA margin in the range of 22.5 to 25.5%. The guidance is based on an average exchange rate of $117 and does not, sorry, $115 and does not take into account the impact of IFRS 15. We have upgraded our financial full year 2025 guidance, projecting an annual revenue in the range of 840 to 870 million, with an anticipated EBDA margin between 24 and 27%. And now I would like to give the word back to Rudy.

Thank you, Alba. I'm very pleased with the progress we made in the second quarter, and we expect this trend to continue in the second half of this year. I'm particularly pleased about the above-average order intake for our microsystems business. It stands for highly complex technologies and applications with great purpose for medical and road safety. This business has very high barriers to entry and will be long-term. With the completion of our capacity expansion programs, we are ideally positioned to grow profitability together with our customers. Sufficient capacity is now available for our most popular technologies, BCD, on SOI and that is in great demand for a broad range of applications in automotive, particularly battery monitoring systems for EV, as well as industrial piezo applications or automatic test equipment for semiconductors, and in medical for ultrasound applications. So with this, I would like to open the line for questions.

Thank you. Ladies and gentlemen, as a reminder, if you would like to ask a question or make a contribution to this call, please press star 1 now on your telephone keypad. And to redraw your question, hit star 2. The first question is coming from Guy Sips calling from KBC Securities. Please go ahead.

Yes, thank you. And first of all, congratulations with these nice results. I have a question on the industrial segment. We saw very nice year-on-year and as well quarter-on-quarter lift-up of the sales in industrial. And the question is, can you elaborate a little bit on this, where it is coming from, and is this sustainable going into the second half of this year? Thank you.

Thank you. Yeah, so in this portion of our business, so this is on the one end a very fragmented industrial business that's typically CMOS type that is stable, is recovering, and is looking good also going forward. There is also our silicon carbide that is to a large extent industrial, which overall silicon carbide increased with 2 million compared to previous quarter. And as I mentioned, there is data center infrastructure power management. I see that in the silicon carbide that is particularly driving this. I mentioned that we have started a lot of wafers in the first half or as much as the whole of 2024. So this will also, they will also further increase in the second half. So that will also have a positive effect there. And then we also had some good prototyping from new contracts that also contributed. So altogether it's a,

strong uh industrial and i expect this to hold for the second half of next this year okay and thank you and and on the yield of the of the the new machines can you give us some clue how that is improving and how what what can we expect of that going forward thank you

When you refer to the yield of the new equipment, the installation is progressing well and the qualification, so far they help with the additional capacity so that we are not in a location and we can support the full demand that is out there. the on the yield side so the production quality is okay so there is no no issues there so everything looks okay if that was the nature of your question yes okay that's that's it from my side thank you ladies and gentlemen as a second reminder if you'd like to ask a question please press top one on your telephone keypad at any time

The next question comes from the line of Robert Sanders calling from Dutch bank. Please go ahead.

Yeah, hi, good evening. Maybe you could update us on your China strategy. I understand you're serving some of the fabulous companies like NovoSense and you obviously have a direct relationship with BYD, but Melexis, your largest customer, is wanting to work more aggressively with local foundries. So are you willing to license your process IP to Chinese foundries in order to support companies like Molexis' localization strategy? If so, which IP and which sort of flavors of IP and which nodes would be interesting? And I have a follow-up.

Thanks. Thanks for that question. So our strategy primarily is to support the Chinese market with all of our technologies, but from the factories that we are operating. The strategy going forward to outsource or license technologies is still under evaluation. There is nothing to report on that right now.

But I mean, is there an opportunity? I mean, it sounds like Chinese companies are now preferring European suppliers over American suppliers. So is there either an opportunity from market share in Foundry or is there perhaps an opportunity because you have more of a European skewed customer base?

Well, we have a European skewed customer base. That's right. But we also have good traction in China with our unique technologies like the BCD on SOI that is not available very much in foundries throughout the world. And so we get good traction for such technologies from China, also China design houses, and So for the time being, we have capacity in our factories to serve that, so there is no need for licensing from a capacity perspective. But yeah, we also said that in the longer term, I think it is, we have been licensing or outsourcing technologies in the period where there were global capacity shortages. So going forward, that is definitely also part of the strategy rather than aggressively further expanding internally.

Got it. And then on TSMC exiting Gallium Nitride in July 2027, I mean, Navitas, one of their larger customers in Gallium Nitride, has announced that they will move to Powerchip. I mean, were you in the running for that business? How much other business do you think is out there to win as TSMC exits in just two years' time?

Yes, Navitas also disclosed on Bloomberg, so we are their silicon carbide supplier, and we have a very good relationship with them, so we're not particularly competing with them. with other sources there, but we are collaborating with Navitas on a broader perspective.

And are there other opportunities in gallium nitride from outside of Navitas?

Yeah, so the fact that TSMC made this announcement, as we have a quite broad pipeline of developments that we are doing in gallium nitride, particularly from our factory in Dresden. And this announcement, of course, gave a bit of a boost on the interest. But, okay, it all first needs to be developed before things can turn into volume production.

Got it. And just lastly, on the data center opportunity that you see, are these small silicon carbide US-based fabulous companies? In silicon carbide, is there a gallium nitride opportunity? What is the opportunity that excites you the most? Obviously, Navitas is already working with NVIDIA. Which particular part is the part that you think you can gain a lot of traction?

Sorry, it's the silicon As I mentioned, the data center applications, they are silicon carbide. And so that is running in quite attractive. So that's the main growth driver for now for the silicon carbide in the past quarter and for this year.

But is that MOSFET? Is that just a standard MOSFET? Where is it? Is it for cooling?

No, it's not for cooling. It's for the power conversion. These are high voltage silicon carbide trench and junction FETs.

Okay, thanks a lot.

We are now going to take the question from Trion Red calling from Derenberg. Please go ahead.

Thanks for that. Hi, yes, Trion here from Derenberg. I just wanted to ask about the comments about shorter term orders. It seems like some of the Q2 sales strength was driven by these later orders. Are you able to increase pricing for these maybe to offset a bit of the reduced visibility that you also talk about?

Yeah, so the pricing in this year is rather stable. So we do not have, with also the capacity that is becoming available, that's a good thing. The shorter term orders, they are not coming at the normal pricing.

Okay, thank you for that. And then the second question, just on the comment about silicon carbide, you're seeing a greater proportion of consigned wafers, which I guess is not great for revenue, but you still get the same EBITDA, which should mean a benefit to the EBITDA margin. So I guess the first point is, is that the correct logic? And as a sort of secondary point, obviously the sales were quite strong in the quarter, but the EBITDA margin was still within the guidance. And usually you guys have quite nice operating leverage. So I just wondered if you could comment about, one, that impact from consigned waivers in silicon carbide, and two, maybe why the EBITDA margin wasn't even better than it was. Thank you.

Yeah, so on the one hand, you're right. So the If the consigned wafers, they result for the same quantity of wafers or products we sell in a lower revenue. But the value add from XFAP side is the same because we don't have a markup on wafers that we supply. And on the overall leverage, yeah, so our top line, if it's better, normally it gives a better margin. However, you also need to take into account that the top line is better, but we are reporting in U.S. dollars. And we have around 44% of our revenue was in euro, which translates to a higher dollar revenue. But we also have costs that are related to the higher costs related to the euro price. that offsets also this leverage effect. So that clarifies the fact that we do not see in this quarter, despite the higher revenue, a better EBITDA.

How much of the revenue growth was due to the weak dollar?

It depends on what the reference point you take. I would have to calculate that. It all depends on what reference point you take. If you compare it with, so I don't have the numbers right in front of me.

We can come back to you with that.

Okay. I was just trying to get an idea of how much of the beats. is due to the effects, but I guess I can go back and work it out myself as well.

I think if there wouldn't be an exchange rate difference, if the exchange rate would be the same as previous quarter, maybe we would be on the high end of the guidance, maybe slightly above, but not as much as we are now.

Got it, yeah. Okay, thanks a lot, guys. The next question comes from the line of Michael Roog calling from DeGroof Petrcam. Please go ahead.

Good evening. I have a question about the silicon carbide business for the data centers that is doing nicely, especially given your comments about the way for start so far in the first half of this year. I was wondering where this is coming from all of a sudden. considering that data center business has already been pushing HBM and GPUs for two years now. Is this market share gains or contract renewals?

Yeah, I think it is maybe on the one hand maybe related to Increased or gradual change in architecture of the power management in data centers using maybe more silicon carbide than in the past. I think that's one. And then in general, of course, maybe also a bit of market share gained by our customers in that field.

And you basically expect the, well, in the first half you had more wafer starts in silicon carbide than in all of last year. You expect that pace to at least be maintained in the second half, right?

Well, these starts, they will now, as we go, increase our revenues in the silicon carbide in the second half. How these bookings will continue, that I cannot predict for sure.

Okay, but for now you see no material change in bookings?

No, it is helping.

Okay, then I do have a question flowing up on 300 about margins. I was wondering, you've been adding a lot of capacity across many different fabs and even clean room in Malaysia. Typically, with large expansions like this, you have a lot of startup costs and learning curves to tackle, and that typically has an impact on margins until you're fully up and running. Have your Q2 margins been held back by such startup costs?

We already have some of the impacts already from mid of last year. The facility is up and running, and the clean room, so the electricity costs and so forth, they started to come in as of maybe September last year. We already started recruiting additional process engineers and technicians a year ago, so these personnel costs, they are already there for a while. And yeah, if we can really start producing, then we will have the positive effect of the economy of scale.

Sure, but say if the capacity you added last year may be seeing better margins today, but today you're adding again a lot of capacity, so that will hold it back. So let's say in six to nine months, you will have no more startup cost learning curve.

We do not expect additional costs. What will come is the depreciation. So we only start the depreciation when the equipments are qualified and ready for production. So there will still come an increased cost. on additional depreciation that will impact on the margins, on the gross margin, but not on the EBITDA, of course. Okay.

That's clear. Good. And my final question is on the operating costs. Typically, there's not a lot of seasonality in operating costs from Q1 to Q2, but this time in Q2, they were a bit higher. I assume that was because of the weaker dollar. Can you confirm that? Or was there also an underlying increase?

Yeah, so there is a bit of effect because we have quite some operating costs in Europe in R&D and administration and so forth. So that had a bit of effect.

Okay, that's clear.

We also have increased R&D activity that also... resulted in higher consumption of wafers that are charged in the R&D.

And then the final question, just one I thought of. If I'm not mistaken, you were given a subsidy by the European Union for your manufacturing expansion. Could you provide an update on where you stand, when that will come?

Well, this is what we call IPSE, the Important Project of Common European Interest, where we benefit from. I think we reported on that. That has, over the lifetime, was something like 80 million. But this is spread over four years, roughly. And that's... More or less equally spread around 20 million in that period. I think that runs to 28 or so.

Okay, so that is already... Are you booking that in your P&L on a quarterly basis?

Yes, so that this is... This goes into the P&L, so this is mostly R&D, so that lowers the R&D reported costing.

So that's about $5 million a quarter. Good. That's it from my side. Thank you.

Ladies and gentlemen, final reminder, if you would like to ask a question, please press star 1 at any time. The next question comes from, we've got a follow-up question actually from Robert Sanders from Dutch bank. Please go ahead.

Yeah. Hi. Intel recently, I think canceled their German project. But the EU Chips Act is obviously in a bit of disarray as well. What is your sense from hearing updates from European leaders about, you know, the strategy going forward? Because obviously without Intel then, There could be theoretically more onus on specialty technologies and more money to come your way.

Yeah, so there is a call for chip manufacturing support in Germany that is running. So we're also in contact with that, but it's too early to report on that. And that's for all kinds of chip manufacturing support.

And just the last question. I noticed that Asian foundries are seeing utilization falling in the second half. They're talking about orders dropping 30% in the second half. And they're also saying that there was pull-in, which has now ended. I mean, if you contrast that with you, obviously you guys address more of an auto-industrial end market, but is there any sense that what you're seeing is somewhat pull-in driven?

It's very hard to say. I have no idea. I have no idea whether there is a... I know what you're hinting to, that there could be maybe some, because of all the tariffs and so, that people are trying to rush with their orders to escape. But I don't think, considering all the relatively long lead times that we have, where most of our We're shipping most of our products to Asia, where they are then being assembled by our customers, subcontractors, before they go to the U.S. But semiconductors, they are excluded from tariffs, so I have no idea.

Okay, thank you.

Ladies and gentlemen, there are no further questions, so I will hand you back to your hosts to conclude today's conference. Thank you.

Thank you very much for the participation today. I would like to remind that on the 9th of September, we have our XFAP Investor Day in Brussels, so you're welcome if you have not yet subscribed. Otherwise, we speak each other for the next call in 30th of October for the third quarter result. Thank you very much for your participation. Thank you. Goodbye.