Asm International Nv

Good afternoon. This is the Coral School Conference Operator. Welcome and thank you for joining the EASM first quarter 2025 earnings call. As a reminder, all participants are in listen-only mode. After the presentation, there will be an opportunity to ask questions. Should anyone need assistance during the conference call, they may signal an operator by pressing star and zero on their telephone. At this time, I would like to turn the conference over to Mr. Victor Barreño, Head of Investor Relations. Please go ahead, sir.

Thank you, operator. Good afternoon, and welcome, everyone, to our 2025 Q1 earnings call. I'm joined here today by Hicham Massad, CEO, and Paul Verhagen, our CFO. ASM issued its first quarter 2025 results yesterday at 6 p.m. Central European time. For those of you who have not yet seen the press release, it is accessible on our website, asm.com, along with our latest investor presentation. We remind you, as always, that this conference call may contain information related to ASM's future business and results, in addition to historical information. For more information on the risk factors related to such forward-looking statements, please refer to our company's press releases, reports, and financial statements, which are available on our website. Please note that the profitability measures mentioned in this poll will be primarily based on adjusted non-IFRS figures. For the reported IFRS figures, as well as the reconciliation between IFRS and adjusted results, please refer to the quarterly results press release. And with that, I will now hand the call over to Hisham Massad, CEO of ASM.

Thank you, Victor, and thanks to everyone for attending our first quarter 2025 results conference call. We will follow the usual agenda for today's call. Paul will first review our first quarter financial results. I will then continue with the discussion of the market trends and the outlook, followed by Q&A. I will now turn it over to you, Paul.

Thank you, HM, and thanks, everyone, for joining our call today. Our revenue in the first quarter of 25 amounted to 839 million euro, which is above the midpoint of our guidance of 810 to 850 million euro. At constant currencies, revenue increased 26% year on year. Equipment sales increased 25% at constant currencies and were led by ALD, followed by EPI. Spares and service sales were up by a strong 32% at constant currencies. Next to solid underlying growth in our outcome-based services, there was again some benefit from accelerated demand in China, albeit not as strong as in the fourth quarter of last year. Note that the increase of 32% is compared to a leveling Q1, which has been revised for the change in the definition of our spares and service sales. As a reminder, we announced this change last quarter, and it only affects the breakdown between equipment and spares on service sales, and further details are available in the press, please. In terms of customer segments, revenue was led by foundry, followed by logic, and then memory. Logic foundry sales accounted for the majority of sales and were up both year on year and compared to Q4. Memory sales increased somewhat, both year-on-year and compared to Q4, but were below the higher levels in Q2 and Q3 of last year. Sales in this segment continue to be predominantly driven by high-K ALD tools for high-performance DRAM and HBM-related applications. 3D NAND remains the smaller part of our memory sales. Sales in the power analog wafer segments continued to be under pressure and were down year on year and compared to Q4. Gross margin in the first quarter increased to a high level of 53.4%, up from 50.3% in the fourth quarter and 52.9% in Q1 of last year. The increase in the gross margin was mainly explained by a favorable product and customer mix, including a solid contribution from China. The gross margin also benefited from a gradual impact from cost reduction programs that we have been implementing over the past few years. We expect the gross margin to normalize in the rest of the year, and we now project it for the full year to be in the upper half of the target range of 46 to 50%. This excludes any impact from potential tariffs. As mentioned in the press release, We have been preparing several scenarios to mitigate potential direct impacts. Over the past several years, we have already taken steps to increase the flexibility of our global supply chain and manufacturing operations. Next to our expanded capacity in Singapore, we are also stepping up our manufacturing capacity in Korea. And as part of the design of our new facility in Scottsdale, Arizona, we also have the flexibility to start local manufacturing in the U.S. SG&A expenses marginally increased year-on-year at constant currencies, demonstrating our ongoing focus on cost control. As a percentage of total revenue, SG&A decreased to 9.1%, down from 11.4% in Q1 of last year. And for the full year of 2025, we project SG&A expenses to be flat to slightly up and to be down as a percentage of revenue. Net R&D in Q1 increased 35% year-on-year at constant currencies. Gross R&D increased by 14%, explained by steady expansion of R&D activities and a growing pipeline of opportunities. Similar to Q3 and Q4 last year, the faster growth in the net R&D compared to the gross R&D is explained by the fact that several development projects entered the commercial release phase in the course of last year, which triggered the start of amortization of the related capitalized development expenses. By 2025, we expect net R&D to remain at the upper end of our target range of high single-digit to low double-digit as a percentage of sales. Operating profit increased 41% year-on-year, which was the result of solid revenue growth, a slightly higher gross margin, and moderate growth in SG&A. Below the operating line, financial results include the currency translation loss of $40 million. This compares to gains of $23 million and $54 million in the first and fourth quarter of 2024, respectively. As most of you know, we hold the largest part of our cash in US dollars, and the translation differences are included in our financial results. Let's go to HMPT shortly. Our share in income from investments, reflecting our stake of approximately 25% in HMPT, amounted to 3 million euro in the first quarter, down from 5 million in the same period last year, reflecting the continued soft conditions in the back-end equipment markets. Our reported net results in Q125 included an impairment of 250 million euro, of our stake in ASMPT, triggered by the reduced market valuation in the recent period. There is no cash impact. Following this impairment and in line with our accounting policy, the changes in the market value of ASMPT will be included in our quarterly net results in case of further decline or until the impairment charge has been reversed. I'll go back to ASM. Let's take a look at our order intake now. Our new orders increased to 834 million, up 14% at constant currencies compared to the first quarter of last year. Looking at the breakdown by customer segments, Logic Foundry was the largest segment, followed by Memory and then Power Analog Waiver. Logic Foundry orders were up both year-on-year and compared to the fourth quarter of last year. Orders for two-nanometer gate-order rounds were up strongly year-on-year and also increased compared to Q4, as leading customers are moving towards high-volume manufacturing this year. Mature Logic Foundry orders, mainly from Chinese customers, were also relatively strong in the quarter. Memory orders decreased year-on-year and compared to Q4, with the largest part of orders for HBM-related DRAM applications. Power, analog and wafer bookings, including silicon carbide, dropped year-on-year and compared to Q4, reflecting the cyclical downturn in this market. Next, the balance sheet. ASM financial position continues to be solid. We ended the quarter with over €1.1 billion in cash, up from €927 million at the end of the fourth quarter. This increase was primarily driven by strong free cash flow of 264 million, up from 62 million in Q1 of last year. Next to a strong increase in profitability, free cash flow was supported by a reduction in working capital, in particular by improved collection of account receivables. Days of working capital decreased to a relatively low level of 43 days at the end of March, down from 50 days at the end of December. CapEx amounted to €30 million in the first quarter. With the last part of spending on our new Korean Innovation Center and with the construction of our new facility in Arizona in full swing, we expect CapEx this year to be towards the higher end of the annual target range of €100 to €180 million. In terms of share buyback, the €150 million program that we announced with the Q4 results started today. Finally, a few words on our currency exposure. As we discussed on previous occasions, the largest part of our revenue is built in US dollars. For the full year 25, the dollar part is expected to be more than 80% of total revenue. The recent depreciation of the US dollar may affect our sales as translated in euros. In view of the recent increased volatility, we decided to change the guidance to growth rates at constant currencies. Our currency exposure is for the largest part translation-related, as the currency exposure over cost of goods to a large extent measures our revenue exposure. Impact on the margin percentage is therefore limited. And with that, I'll turn the call back over to Hichem.

Thank you, Paul. Let's now continue with the review of the market trends. As discussed, we delivered a strong performance amid continued mixed market conditions. AI-related markets, in particular Gatoral ramp and HBMV ramp, remained strong. But in the other segments, softer conditions continued. Global uncertainty has increased significantly following the recent Paris announcement. We have several scenarios in place to mitigate potential impact, including local manufacturing in the U.S., as just discussed by Paul. In terms of indirect impact, tariffs are likely to create a more challenging microeconomic environment, but it's too early to tell what the exact impact on GDP and the semiconductor market is going to be. We continue to monitor potential impacts closely. So far, we have not seen any meaningful changes in our conversation with our customers. Based on these conversations, Our customers continue with their capacity expansion plans for the most advanced nodes as they expect structural increases in demand for AI in the coming years. Our focus remains on investing in R&D and working with our customers to develop the next technology nodes. Increasing complexity, 3D structure, and new materials of next-generation semiconductor devices will require more ALD and EPI steps and create new growth opportunities for our company. In addition, we stay disciplined in costs and focused on further efficiency improvements. As mentioned, ongoing programs to reduce the costs of our tools are bearing fruit. New platforms in development will see further cost reductions on the back of increased commonality in parts and components. These savings will help us on delivering the most competitive cost of ownership for our customers, as well as sustaining healthy growth margins. Let us now review the trends in the market segments in more detail, starting with logic and foundry. In the advanced logic foundry segment, gate all-around sales increased chart-free year-on-year, and also compared to Q4, as leading customers continue to invest in capacity for high-volume manufacturing. Some of our customers have reported strong initial demand for 2 nanometer, which is already exceeding the previous 5 nanometer and 3 nanometer nodes at the same period in time. Following the introduction of 2 nanometers this year, our customers are planning for new sub-node versions of 2 nanometers in the coming years, including, for instance, backside power delivery. Despite some capex shifts between customers that we mentioned last quarter, our forecast for a substantial overall increase in gate-all-around related sales in full year 2025 remains unchanged. This increase is also supported by solid increases in our service available market with many new ALD applications as discussed in previous calls. ASM is also strengthening its position in silicon EPI. In the gate all around transistor, the channel structures are defined by EPI and in the future, we also expect further penetration in other applications. We expect advanced logic foundry to remain a key growth area for ASM in the mid-term. We have active R&D engagement with key logic foundry customers for the 1.4 nanometer node, the second generation of GatorRound, which is expected to be introduced in the 2027-2028 timeframe. As mentioned last quarter, we have already been shipping R&D tools for multiple critical applications in this new node. In the first generation of Gate-al-Around, we focus on the architectural change. In the next generation, the industry aim is on further unlocking the potential of Gate-al-Around in order to boost device performance and power efficiency. We are closely working with our customers on such performance enhancement processes, key applications include enhanced films in the gate stack, increased adoption of metal ALD, and selective ALD. Next, let's look at the memory market. Demand in Q1 continues to be healthy and primarily driven by demand for HBM-related DRAM. In this segment, we have a leading position with high-K ALD application for high-performance DRAM devices. For the year as a whole, we also expect healthy memory sales in total, but below the very strong level in 2024. HBM-related DRAM demand is expected to remain solid. In 3D NAND, we are seeing some continuous technology buys, but overall levels in this segment are fairly small compared to high-performance DRAM. the memory market remains a strategic growth opportunity for ASM. There are more and more logic-like technologies on the DRAM technology roadmap, which play to our strengths. In the next DRAM notes, additional layer in the high K-gate stack are expected to transition to ALD. In the mid-term, within the next three years, DRAM will transition from 6F squared to 4F2 technology or vertical channel transistor, VCT. This involves not only a shrinking of the memory cell, but also a more pronounced three-dimensional structure, and as a result, more complicated integration schemes. We expect this will increase both ALD and epitaxy intensity, underpinned by our current R&D engagement with all the leading DRAM players. Finally, the power analog wafer market. Demand remains subdued as customers are still facing inventory corrections and sluggish and market conditions, such as in automotive. After a significant double digit drop in 2024, the silicon-based part of our power analog wafer sales is expected to remain weak in the rest of the year. For the silicon carbide epi market, Prospects for this year have also weakened. After substantial capacity expansion in the last couple of years, several customers are now slowing down or postponing new investments. We believe long-term prospects for this market remain positive, and once investments pick up again, ASM will be well-positioned to benefit. Following several new customer wins in 2023 and 2024, We achieved the number one market share position in silicon carbide FE last year, according to recent third-party reports. Our PE two-way tool will help us to further expand our position as the silicon carbide industry prepares for the 200 millimeter wafer size transition. Finally, our outlook. We expect 10% to 20% revenue growth in full year 2025 on a constant currency basis. We have reasonable visibility, as we mentioned in the press release, to meet the lower end of the guidance range. To achieve the higher end of the range, some opportunities need to materialize. We expect our sales to grow ahead of the wafer fab equipment market, which is still expected to show a slight increase this year. A strong increase in gate all-around related sales would be the key driver for this year, as just discussed. HBM-related DRAM sales will also be solid, but overall memory sales would be lower compared to the very strong level in 2024. The outlook for power analog wafer, including silicon carbide, has further deteriorated compared to our previous updated in February. For the second quarter, we expect a continued healthy development in sales with an increase of 1% to 6% at constant currencies compared to the first quarter. And with that, we have finished our introduction. Let's now move on to the Q&A.

We'd like to ask you to please limit your questions to not more than two at a time so that everyone has a chance to ask a question. Operator, we are ready for the Q&A. Can you please open the Q&A?

Thank you. This is the Coruscant Conference operator. We will now begin the question and answer session. Anyone who wishes to ask a question may press star and 1 under touchtone telephone. To remove yourself from the question queue, please press star and 2. Please pick up the receiver when asking questions. Anyone who has a question may press star and 1 at this time. The first question is from . Please go ahead.

Yeah, hi. Good morning. Yeah, just when I look at your full year guidance and calculate the second half, it seems to be that, you know, if you take the sort of midpoint of the guidance, there is not much growth into the second half of the year. And I'm just wondering where the different parts, and if I break it, assuming that advanced logic foundry on the gate or round side is staying very strong, you're sounding incrementally more cautious on the power analog wafer as well as on the memory side, where previously you said it could be flat memory and now you're saying it'll be down memory. And then there's also the services side where you've had some elevated levels of ordering from China. So I'm just saying, can you give us a little bit more color into the second half where you see some of the incremental weakness coming through? Is it mainly a memory drop or is it some of the high spares and services which is coming down to a more normalized level?

Yeah, thanks for the question, Jonathan. So I think you're right. We may be first to say that, of course, still the outlook for the year is uncertain. As you know, our backlog on average is six months. So there's still a part that is still difficult to assess. So we have made, of course, certain assumptions based on certain insights that we have. But anyhow, so there's still not full visibility for the full year. So I want to start with that. Then in addition, basically, if you ask us now, what do we see today, based on today's visibility, it's very much in line with what you just said. So GATE all around, indeed, continues to be strong basically every quarter, with some fluctuations, of course, from quarter to quarter, but overall continues to be strong. Customers continue to invest, which is a positive, of course, One customer is very large and the others are somewhat smaller, but overall there is a clear commitment, which is positive for us, so that is developing very well. Then in the other market segments, at least based on what we see today, we see compared to the first half maybe indeed some reduction. It's still a little bit too early to tell if that really will happen or not, given what I said in the beginning. We don't have full visibility for the year yet. It's quite some level of uncertainty given all the geopolitical environment that is going on. And last but not least, indeed, we have also seen in Q1 accelerated demand in spares and services, as we just said. We don't expect that to continue, of course, each and every quarter. So also there we would expect that to normalize more going into the remaining of the year. Understood. Thanks.

And my follow-up is, you know, you talked about visibility for the low end of the range and you need some opportunities to materialize to hit the high end of the range. Can you give us any indication of where those opportunities lie? Are they sort of new fabs on the advanced logic side or is it in some other area?

I think I will take this question. Most things that the upside opportunity that needs to materialize is really in the area of more than more or power analog. As you know, the power analog market has been a downturn and for the last sixth quarter from uh first quarter 2020 uh 2024 to uh second quarter of this year so right now we don't see any visibility but in this market you know this market can come back uh anytime so um in case it could recover then we will see uh upside opportunity in that part of the market We're also in HBM too. There might be also more HBM memory coming in. Some of it, you know, there's in for AI chips, we see from here and there some limitation of advanced packaging. So it's really limited. So maybe if those limitation, are taken care of, then you're going to see also more HBM memory orders from that point of view. So these are actually the most things that we see from that point of view. But also, you know, in a Gatorade yield, if you see some yield increase, then some customer might pull in some of the orders in the second half of this year. So those are the upside opportunities that we see for 2025, second half.

Understood. Thank you so much.

The next question is from Nigel van Putten. Morgan Stanley, please go ahead.

Hi. Thank you. Good set of questions just now. I'll move on to perhaps China. It was strong in the first quarter and also seems to have been contributing to the order intake. You previously said it's going to be low to high 20s, but given strength in the first quarters, would it make sense for China to move higher in that range? That would be my first question. Thanks.

I can be very short, Nigel. Based on what we see today, yes, is the likely answer. It's very much still in line with what we expected. To be honest, maybe marginally better, but not a lot. But indeed, China is developing well. Of course, this does not take into account potentially new export controls if they were to come. We don't know, as we also told in previous quarters. But it's very much in line with what we already expected based on the last update and maybe marginally higher, maybe a little bit better. Then for the full year, as we see today, we expect it still to be lower in 2025 compared to the full year of 2024.

Right. Is that no change or is that still, could that imply still you're looking more towards the high end of that low to high 20s range for the full year?

Yes. So we look more towards the higher end of the range, but that's more or less in line with, let's say, what we were starting to see with the previous update already. It's maybe slightly better marginally, but not very much. It's very much in line. And also the fact that 25 is expected to be lower than 24 is also in line with expectation, but likely to be indeed towards the higher end of the range.

Got it. Thank you very much. Switching gears, the order book. I was a bit surprised to see 4% quarter-on-quarter tailwind, 6% year-on-year. Given how the dollar behaved versus the euro, that would imply you received the very large majority of orders very early in the quarter and very little in March when I think the FX was quite different. Is that out of the ordinary? Well, first off, is that correct? And then second, is that out of the ordinary? I think you usually also work with some customers on a schedule. So is that, yeah, again, normal that they would produce the orders early in the quarter, or is something different this time around?

So you're referring to the difference, let's say, between the growth in bookings in constant currencies and nominally. That's assuming you're referring to an idle.

Yeah, exactly. Yeah, this is quite a big difference. And if I look at the FX, it's Yeah, it sort of implies that the orders were placed early in the quarter.

There's a few things there, so I don't know by heart the monthly phasing of bookings, but we have all flavors that I can tell you. Depending on the quarters, it never comes in precisely as you predicted. might anticipate. That's also why we stop guiding for us because even within a quarter to tell when in which quarter it comes already difficult and even more so in which month. So it can indeed sometimes it's front loaded, sometimes it's not front loaded. Two, as you might know, we had that question before when it was difficult to understand currency movement. What happens is we receive orders in dollars. and then these dollars are converted to the local currency of the country that that booking receives and then based on the local currency of that country we convert it back to euros because yeah the local entities report in local currencies and we translate that back in euros so there's a double conversion if you wish most of the time it is very much in line with the dollar euro development that can explain nine out of ten times the movement but not always sometimes we see a a additional effect of a local currency movement relative to the euro. So it's not complex on one end, but to simulate that, you need to literally go entity by entity and see how the local currencies, the dollar versus the local currency versus the euro develop. But this should all be correct, what we have reported, because it's checked, double-checked, and triple-checked.

Oh, I'm sure. Yeah. No. So nothing out of the ordinary. It has to do with all kinds of difficult calculations. Maybe a real quick follow-up then on this.

Next caller, but you can go.

Okay. Appreciate it, Victor. Thank you.

The next question is from Sandeep Deshpande, JP Morgan. Please go ahead.

Hi, thanks for letting me on. My question is regarding what kind of visibility you have into 26 at this point, or is it too early to talk about 26? Given that you're seeing this GAA strength in 25, is there a risk into GAA in 26, or there are other drivers that you see into 26?

I think that's for, okay, thank you for the question. We still see that the strength in Gate to the Round will continue in 2026. The reason we're saying this, because with interactions with our customer and working with them, we see there's some, we can see that there's some need to Gate to the Round in 2026. But also, if you look into our If you look into our colleagues in the south of the Netherlands, they also are very positive about 2026. And as you know, for this EUV, the lead time is very long, so over a year of lead time. So we expect to see more and more booking happening in the second half of 2025 for delivery in 2026. So overall, I think that we still see improvement in gate-tolerant technology. And in 2026, gate-tolerant is very favorable compared to the old technology nodes like 3 nanometer and 5 nanometer because of better performance and also lower power usage. But at the end of the day, you know, There are microeconomic factors that might change it, but right now, things look positive.

Thank you, Shaman. My quick follow-up on that is the two new drivers that you've highlighted in some of your comments earlier. One is, I mean, we know that MOLLE, you're gaining some footprint in MOLLE as well as 4F squared in the DRAM market. I mean, how should we be looking at, you know, what kind of revenue you can generate on a 26 or 27? I mean, I don't know when does this revenue start coming in and how we start modeling this revenue for ASM.

So if you look into the memory market and the move to 4F squared, I think the move to 4F squared in manufacturing will start in 2028 at the earliest. But we're working with all customers right now on the move to on 4F squared opportunities. And because of the structure architecture of the cell, the more complex architecture of the cell and the 3D part of it, there's more and more LD and epi intensity happening in 4F squared. And we are very excited really with the engagement, the engagement with all customers and also the number of opportunity that's happening in 4x squared. So we see that starting in 2028 at the earliest. When I talk about 2028, I talk about HVM, but we're working with them right now on R&D and also on evaluation and qualification. When you talk about for logic, I think logic too. You know, the transition to 1.4 nanometer technology node, which is the second generation of Gatoron, which we are also in partnership with our customers, we see more and more ALD layers happening in the 1.4 nanometer technology node. So, I mean, overall, we are actually very, really upbeat about the engagement that we have in the high performance, both on the logic foundry side and also on the memory side. Thank you, Chef.

The next question is from Didier Shimama, Bank of America. Please go ahead.

Yes, good afternoon. Thank you for taking my questions. I think you're going to hate my questions, but I'm going to ask them anyway. First question is, Hichem, you talked up the gross margins earlier in your introductory comments when you said, You're working on lower-cost platforms across the portfolio. So my question, I guess, maybe to Paul or to you, Hsien, either way, do you expect to trend towards the upper end of your long-term gross margin range as we progress into 26, 27? Or can we even envisage a higher range or new range of gross margins? And I've got a quick follow-up. Thank you.

Okay, thank you for your question. As we mentioned in our press release, I think for 2025, we're going to be on the higher range of our 46% to 50%. We're very happy about it. I think that the effort we as a company have done the past couple of years on improving our operational efficiency, on having a new platform that has more common parts between the different products that we have, on having also one software architecture, really helped us reduce our costs and eventually makes lower the cost of ownership for our customers. and at the same time improve our gross margin. These efforts are continuing every day. We're also continuing to really try to get better efficiency. We think that because of these efforts, we might get better margins going forward in the future years. I think we're only starting right now. Okay, thank you.

And for my follow-up, I guess I hear what you said, Hichem, on ASML EUV order. So on that bit, and I think going back to the point, I think it was Jonathan who asked the question on the upside opportunities in the second half. You mentioned DRAM or memory perhaps. But when you look at ASML order intake, one of their customers, which is also one of your customers, their new HBM fab, and they came back very quickly. They came back in Q4. They came back again in Q1 with more EUV orders. So I would have thought that having pulled also their second HBM FAB, you know, at some point is going to feed into your own order entry. So is that one of the opportunities you're thinking about, or is that already reflected in your order intake in your backlog?

Yeah, we have basically what we know today is reflected in our backlog, is affected in our, yeah, basically in our guidance backlog. So from that point of view, as we said, HBM-related amount is strong, will continue to be strong. We don't think it will be as high as the very high level that we have seen in 2024, where we grew more than 150%. So 2024 is really good. Also last year, we had some lumpy sales, memory-related sales in China that at least this year we don't see yet. Also that plays a role. So the combination of those two give, let's say, the guidance that we have given. Okay.

Thank you.

The next question is from Francois Bouvigny, UPS. Please go ahead.

Thank you. My first question would be on the tariff side of things. So you mentioned your flexibility to manufacture in the U.S. I was just wondering, are you going to do it? I mean, or you're just still, you know, reviewing it? And I'm asking that because obviously one of your big customers is having some big plans in the U.S. And I was wondering if you see any risk that they prefer some local supply, maybe with less tariff impact, if you see any risk on the market share and that's why you would do U.S. manufacturing. That's my first question and I have a follow-up, yeah.

Okay, so as you guys know that we've been in... in the U.S. for over 50 years, specifically in the Phoenix, Arizona area, where most of our big customers are at this point in time. So over the past 50 plus years, we have significant interaction with the local supply chain in the U.S. We have already three different buildings in the Phoenix area with two lots. We have lots of space, clean room space, So we're already starting to manufacture some of the tools for our customer in the US just to get us started. So from that point of view, I think that our global installed base, our global infrastructure allows us to really have manufacturing in many places. And Phoenix, Arizona is one of them. So I don't really see any impact of not us able to manufacture in the U.S. at all. On the contrary, you know, with the development of Epitaxy and ALD in Phoenix area, we actually manufacture the first tools for all the NPI or the new product introduction. We have people there. We have teams that really manufacture the first four or five or six tools. So we have the infrastructure. we have the space and we are starting to do that right now as we speak. And then when our ScotSphere facility will be ready, the second half of 2026, we'll be able to expand that further from that point of view.

Thank you Hichem. My second question is on, you mentioned 1.4 nano, you know, with layers increasing, but actually we will see potentially a different version of 2 nano which is the N2P and A16, you know, next year. So I was just wondering, how do you see your layer count in these versions of N2 next year, the N2P and A16? Do you see an increase versus the first version of N2?

So we see marginal increase, not a significant increase with the subnodes. I think the significant increase really is happening from one node to the other. So for the 2 nanometer to the 1.4 nanometer, definitely you're going to see double digit increase in ALD intensity. But for these subnodes, yes, there's an increase because the subnodes are really geared mainly to improve performance. And for that, you know, you need some more ALD intensity, but that increase is marginal. It's not very significant.

Thank you very much.

The next question is from Jacob Bluestone, BNP Paribas Exame. Please go ahead.

Hi, thanks for taking my questions. Firstly, just getting back on the tariff point, I appreciate there's currently a 90-day reprieve in place on the tariffs. But it sounds from your comments like you wouldn't be able to step up your assembly inside the U.S. immediately. So can you maybe just give us some sensitivities on what would happen if there isn't a resolution? What would be the sort of impact on your business? And maybe if you can also just confirm that there were no pull-ins during the quarter. It sounds like there wasn't, but if you could just confirm that. Thank you.

On the terrace, maybe one clarification. We can act very quickly, almost immediately, to have some products, the most critical ones, produced locally in the U.S. to the extent needed, and of course expand, as Guillaume already indicated, towards the end of next year when we have a Scottsdale facility ready. That's one. It's not yet a given that local production will be more efficient. It might be, it might not be, because today we import most of our tools, if not all, from Singapore, which at least based on the initial announcements were in the lowest tariff bucket. And once we start producing locally, you import parts, modules, subsystems from different countries with different tariffs. it might very well be that continuing to produce where we produce today might still be more efficient. We all have to, let's say, make up final calculations once it's clear how these tariffs will look like. What for us is most important today is that we have optionality. We have created optionality. We have a global manufacturing base. We have three Countries where we can, the manufacturing, although the bulk in Singapore, we still have flexibility as well in other countries. And depending on how tariffs will play out, depending on where our suppliers are, we have to, let's say, in that puzzle, find the best and most efficient, let's say, supply chain solution to mitigate the financial impact as much as possible. And that's what we are planning to do. Then on your questions on pull-ins, I'm not sure I understand the question. There's always pull-ins, but also push-outs. I mean, there is continuous movement of tools from quarter to quarter, orders from quarter to quarter. Maybe not orders in the sense, finally, boom, but let's say planned delivery schedules. Sometimes customers want them earlier. Sometimes they want a month later. I mean, is that what you're referring to, or because of orders that customers are ordering?

Are you seeing any pull-ins relating to tariffs? It sounds like the answer is no.

No, no, no, no. At least not. No, no. We have not seen it, not heard about it. So also today, as you know, there's no tariffs on our equipment. The spare part is a different story. But on tariffs, there's – sorry, on tools, there's no duty at this moment. Understood. Thank you.

The next question is from Stéphane Oury, OdoBHF. Please go ahead.

Yes, hello. In fact, I have two questions. The first one is about the 2027 targets that you have set, the 4 to 5 billion. You have changed the way you give guidance for 2025. Can you maybe update us on 2027 in the same way? And the second question is about the R&D level. I understand that you're saying that you will be in the high hand of the high single to low double digit, but can you be a bit more specific for modeling purpose notably? Thank you very much.

To be very honest, Stefan, I've not really looked to 27 targets yet. We have our 45 billion that still stands. We have an investor day in September and we'll update you there. But for now, for us, nothing has changed. I mean, the reason why we did what we did now in constant currencies is because of the very sharp drop in recent weeks. We have a big difference between Q2 and Q1 currency, almost 8% to 9% difference. If we would not have said anything, you would have been surprised most likely in Q2. And of course, if we proactively say something, you might also be surprised that we suddenly bring this up. But we thought it's better to at least proactively basically say this. Most of the time, there is no issue. I mean, with a few percentage points up and down, we can manage no problem. We've always managed that. But with the current geopolitical events that I think you all are aware of, So many things are happening that we thought we need to now change it for this year. If we change it also for future years, we have not really had that discussion yet. But as you know, we have always included a CAGR in the targets that we have announced, so that CAGR you can take as, let's say, a growth rate at constant currencies relative to the year that we took it to. The other thing is, We only talk pure translation. I just want to make that very, very clear. It's literally converting results from local entities that are invoiced in dollars to euros. That's it. So it's nothing to do with our performance or in terms of how many tools we sell, at what price we sell, et cetera, et cetera. This gives you, I think, more and a better insight into the underlying performance than if you would not have, let's say, given this level of transparency. But ideally, we don't want to talk about currencies. We want to talk about business. But at this time, given the pretty sharp drop, it was unfortunately necessary to also bring up this point. And then R&D, the higher end of the range, you should work towards the higher end of the range, which is high single to low double digits. So I leave it to your imagination, but low double-digit is whatever. I'm not going to mention the number, but you know what it is.

Well, it can be very wide, that's why I'm asking. Okay, thank you very much.

The next question is from Andrew Gardiner, City. Please go ahead.

Good afternoon. Thank you for taking the question. I think a fairly quick one from me, just regarding the point you were making regarding the upside opportunities that may or may not materialize. You've highlighted here under the Q&A that most of those are in the, I think, Hisham, you said more than more camp for power and analog. it seems like you're putting it down to the state of the cycle and the uncertainty rather than whether or not these customers would actually move forward with these projects. So is it something that if they don't happen in the back half of this year, those are projects that probably just slip to 2026 and may help you there? Or are they actually something sort of perhaps more blue sky things that may not materialize at all depending on what the customers decide?

Thank you. If you look into the more-than-more market, it's a very cyclical market, and the downturn for this market is really between 6.25 to 8.25. By the end of Q2, that will be the end of the 6.25. So usually you'll see recovery by either the end of Q4 of this year, which would be Q1 of 2026, or maybe within the second half of 2025. So that really, usually that's what happens. And when the recovery happens, more than more, it's instant recovery. It's really not like the... logic or DRM market which is more structured. This thing really can happen and it can flip on a dime. And that's really the reason that we have uncertainty there. But also that market also depends on the economic uncertainty that's that's there right now. If GDP increases and so on for the second half, if economic uncertainty goes out, then I think it's going to flip positively very, very, very, very fast.

So that's really where it comes from. Thank you.

The next question is from Tim Schultz-Malander, Redburn Atlantic. Please go ahead.

Yeah, hi. Thank you for taking my question. Paul, maybe just coming back to this productivity program, so appreciate the examples, but maybe if you could add some color about the duration, kind of how far through that program are you, and if you're feeling generous, maybe some scale as to the benefits you expect, and then have a follow-up.

So it's difficult at this stage to put really a number to it, Tim, so I'm not going to do that, but what I can repeat and say is that This, of course, is already for quite some time we are working on, of course, continuous efficiencies, but also productivity improvements. In particular, I think all the work, which is, of course, more medium to longer-term work, and we started a few years ago on platform, can be reasonably significant over the years to come. It will not be one bump up. But there will be definitely quite some improvements in the years to come, gradually from year to year, everything else equal. So that's a positive. We also talked about MIT and Merchant Transit, which will bring benefits, which we also will start gradually and then build over the years to come, which is also partially linked to the platform, actually, to be honest. So they will go a little bit hand in hand. In addition, we, of course, have continuous productivity improvement targets that we execute upon. So there is a continuous effort, but that will be a continuous gradual improvement because it's not like that suddenly all our tools will flip to the new platform. It will go step by step. That's one. But it will be a continuous benefit in the years to come. The short-term fluctuations that you see, to a large extent, I know it's maybe not a very satisfying answer, but our mix, customer mix and product mix, and customer mix, of course, that also includes China, that takes care of the short-term, more substantial fluctuations that you see. But having said that, everything else equal, with all the efforts that we're doing, you will see a gradual improvement of the margins.

Great. That's super helpful. And maybe just my follow up for Hisham, just on the technology roadmap. Again, not looking for a specific number, but just some sort of sense of scale. If we think about the step up of your opportunity, your SAM coming into gate all around, the next step into sort of backside power, power rail. Can you just give us some sort of sense as to how we should think about that opportunity in relation to the step up that you're seeing with gate all around? Thank you.

I think that there's a good step up in ALD intensity by going to the super power rail part of the architecture. It's a significant ALD add, but also it really depends from which customer adds that one versus the other customer. Overall, it's a meaningful increase, but also importantly, too, what really matters at the end of the day, it's really the new technology node that I've been talking about, the big transition from the 2 nanometer to the 1.4 nanometer, and also from the 1.4 nanometer node to the 1.0 nanometer node. That's where you really see the biggest increase in ALD intensity going forward.

Thank you.

The next question is from Tammy Kew, Barenburg. Please go ahead.

Good afternoon. Thank you for taking my question. So firstly, regarding to what happened in China and locally, you talked about they are working on various generations of tools. And in my view, one of the technology design they want to do in the coming years will be something similar to 4F Square. Can you confirm that? Do you think your Chinese local competitor can already offer ALD tools to match that kind of manufacturing process? Or you are still sort of the only one globally can handle ALD solution, which can support 4F Square DRAM manufacturing process?

Please hold the line. The conference will resume shortly. Your lines are open. Please go ahead. Hi, guys.

Did you get my question or that didn't go through?

Can you repeat your question?

The line's up. Yeah, sure, sure. Thank you for taking my question. So, firstly, I want to ask your technology capability versus your competitors in China. So, from your view, can your Chinese competitors for ALD solution already support Technologies such as 4f square or you are still the only game in town who can support that kind of advanced generation of design Okay, thank you very much for your question.

I'm sorry that the phone has dropped, but if you look into the our competitors in China, I think that the past few years shows that our market penetration China is good and solid and The reason for that is that we have a very competitive products but also at the same time our China penetration is also limited by export control. So some of the high-end ALD layers we cannot play in from that point of view in China market. Because of export control we cannot ship ALD and also epitaxy products to the China market. But for the other parts of the market that's really been, for the more mature part of the market that we address, We have been very competitive in our EPI market, EPI application, also ALD, and that's the reason that, as you have seen, that our market in China is good, and our market share in China has been very solid last year and also this year from that point of view. But as far as 4F squared is concerned, I'm not really going to talk about that, okay, because that's an area that we're not really working on in China from that point of view. I think that's our engagement in memory with customer on the high end. Part of it is outside of China.

Okay, thank you. And follow-up is? So you have seen this massive time expansion story over the past year or so on the back of GAA. In the coming years, there are lots of things happening, which is quite messy because you do have some customers already on backside power. Some customers are going to adopt that. And you do have this DRAM manufacturing process moving to high K and more penetration into higher generation or more advanced generation of DRAM and then introducing, for example, something like AOD MOLLE as well. So from your perspective, on a CAGR basis, over the next three years, do you think your time expansion can be as significant as over the past year and a half in the coming three years, or we have to take a pause after the GA penetration is done?

As we've been mentioning the past quarters, and as we have talked about, you see that our R&D intensity is on the high end. The reason that it's on the high end, it's the highest single double digit, okay, the low of the double digit, is because we are very positive about the future opportunity that ASM has in ALD and also in epitaxy. So structurally, the market is actually moving in a direction where it needs more and more ALD and epitaxy. If you look into ALD, the intensity is increasing in gate-to-round, and as you go to the next generation of gate-to-round and also to CFET, which is the ultimate gate-to-round structure, you have more and more ALD intensity. Now, if you look into the HBM, or the high-performance DRAM, the high-performance DRAM needs, because of high-performance performance, that performance is actually given by ALD. So we will see more and more ALD incorporation. into the DRAM area. But further than that, we also see Epitaxy, which was only running for the analog market and the logic market, is actually also moving into DRAM. So we are extremely positive about our opportunity that we see ahead. The fact that the technology now is becoming more complicated, the fact that devices are moving more and more into 3D, you need more dielectric, you need more deposition tools, and you need more high performance, and that's the sweet spot for our offerings.

Okay, so your time expansion pace shouldn't slow down, right, after GAUs? No, it's not going to stop.

And I think that we will give you more and more information about it in our meeting, investor meeting that we're going to have in September in London.

Okay, thank you.

Thanks, Tommy.

The next question is from Aditya Metuku, HSBC. Please go ahead.

Yeah, good afternoon, guys. Thank you for squeezing me in. So firstly, just a clarification maybe for Paul. Paul, I understand you hedge some of your orders, your revenues when they come in, when the orders are placed. So when we're thinking about the effective FX headwinds, is there any hedging on the top line that we should keep in mind? Or can we just assume that the spot rates are what will drive the FX impacts? and I've got a follow-up on silicon carbide.

We don't hedge. There's no hedging. We have a reasonable natural hedge. So our cost of goods, also to a very large extent, is invoiced in the US dollar. So there will be a marginal impact on gross margin percentage, and maybe a little bit more, but still marginal on operating margin percentage. absolutely different because you need to convert different values, but we do not hedge for that reason.

Got it. I think you don't hedge the top-line FX exposure.

No, definitely not.

Got it. And then just as a follow-up on silicon carbide, Hesham, I think you mentioned you were the number one player in the silicon carbide epitaxy market last year, if I heard you correctly. Yes. One of your peers, your German competitor, also thinks that, number one, I suspect there is some difference in the definition, so there is some kind of disconnect there. But I just wondered, how do you think about your silicon carbide epitaxy tools on a total cost of ownership basis versus competition?

I think that to answer your question, I think that when we mentioned that we are the number one silicon carbide supplier, that's really third-party information not coming from us. On the other hand, if you see that our revenue has increased from 2023 to 2024, significant increase in our silicon carbide revenue. We are one of the Let's say only companies where revenue in silicon carbide has increased from 2023 to 2024. So I hope that will answer your question.

Okay, thank you. Operator, we are already past the end of the hour. Can we please have the last caller?

Yes, so the last question is from Robert Sanders, Deutsche Bank. Please go ahead.

Yeah, thanks for squeezing in. I just want to know that the China business this year, what percent of that is actually mainstream logic and foundry for 28 nanometer and above? I guess this is a roundabout way of trying to understand your exposure to some potentially restricted entities going forward. And I have a quick follow up.

In last year, but also this year, the biggest part will be actually mainstream logic foundry, or mature logic foundry, as we call it. Over a full cycle, actually, PowerWave, Analog, and also including Silicon Carbide, then might be maybe even bigger than the mature logic foundry, but specifically for this year and last year, we're Especially PowerWare for analog, of course, is coming down. Material logic finding is the larger part of the exposure there.

Got it. And how do you assess the stockpiling that you see at your customers at present in China? Clearly, there have been stockpiling spare parts, but how do you assess the amount of tools that are actually up and running rather than sitting in warehouses? Thanks.

Most of our tools are up and running. We know a few cases where fabs were maybe not ready, but customers still want the tools to be ready and delivered. They might not yet be up and running, but we install all our tools. We qualify all our tools so we know precisely where our tools go. We know precisely what our tools do. So, yeah, the bulk, virtually all tools are up and running, Rob.

Thanks a lot.

Okay, ladies and gentlemen, that was the last question, so I turn the conference back to Mr. Marigno for any closing remarks.

So I will end this call. I would like to thank everyone for joining our call today. and for the questions that we have received from everyone. So on behalf of both Paul and Victor, I would like to thank each one of you for attending our call today. Bye bye. Thank you for attending our call today. Bye bye. Thank you for attending our call today. Bye bye. Thank you for attending our call today. Bye bye. Thank you for attending our call today. Bye bye. Thank you for attending our call today. Bye bye. Thank you for attending our call today. Bye-bye. Thank you for attending our call today. Bye-bye.