ASM International NV (ASM) Earnings Call Transcript & Summary

Good afternoon. This is the Chorus Call conference operator. Welcome, and thank you for joining the ASM International Q1 2023 Earnings Call. [Operator Instructions] At this time, I would like to turn the conference over to Mr. Victor Bareño, Head of Investor Relations. Please go ahead, sir.

Thank you, Sherri. Good afternoon, and welcome, everyone, to our Q1 earnings call. I'm joined here today by our CEO, Benjamin Loh; and CFO, Paul Verhagen. ASM issued its first quarter 2023 results yesterday at 6:00 p.m. Central European Time. For those of you who have not yet seen the press release, it is available on our website, asm.com, along with our latest investor presentation. As always, we remind you that this conference call may contain information relating 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. And with that, I'll now hand the call over to Benjamin Loh, CEO of ASM.

Thank you, Victor, and thanks to everyone for attending our first quarter 2023 results conference call. We know it is a busy day for all of you. I will follow the usual agenda for today's call, Paul will first review our first quarter financial results. I will then continue with a discussion of the market trends and the outlook followed by Q&A. I will now turn it over to you, Paul.

Yes. Thank you, Benjamin. Maybe first, that in my remarks, I will focus on normalized results. Details of the acquisition-related PPA expenses and reconciliation of normalized and reported results can be found in the press release and in the investor presentation. So let's now move to the financial results. In the first quarter of '23, we delivered revenue of EUR 710 million. At constant currencies, this was an increase of 40% year-on-year and unchanged compared to the fourth quarter of 2022. Revenue came in slightly above the top end of our guidance of EUR 660 million to EUR 700 million due to a few shipments that were pulled in by customers from the second quarter. Equipment revenue increased by 43% year-on-year at constant currencies, led by ALD and Epi. Momentum was also strong in our vertical furnace product line, which achieved a new quarterly high in revenue. Spares and services revenue had a solid quarter, up 21% year-on-year at constant currencies. Then in terms of customer segments, revenue was led by foundry, which increased both year-on-year and compared to Q4 and at a new quarterly high. Power analog was the second largest segment strongly up and also at a new quarterly high. Logic sales were still at a healthy level but decreased both year-on-year and quarter-to-quarter. This was followed by the memory segment with sales substantially lower, reflecting the significant cuts and investments in this market. Gross margin increased to an exceptionally high level of 51.1% in the quarter, up from 46.9% in the fourth quarter of '22 and 47.8% in Q1 of last year. The increase in Q1 gross margin was explained by mix, including an exceptionally rich application mix within our ALD sales and also an increased contribution of China sales. In Q1, cost inflation further increased, although we still had some benefit from early ordering of materials that we did last year at relatively lower costs. In the following quarters, we expect an increased impact from cost inflation to a certain extent, compensated by continued focus on cost reduction programs, including value reengineering and by passing on higher cost and the pricing of our tools. SG&A expenses increased 15% compared to Q1 last year, mainly reflecting increased head count. The year-on-year increase in SG&A slowed down compared to the quarterly increases last year, which at that time were explained by the extra investments we made in strengthening our organization. For the full year of '23, we expect SG&A to be slightly lower compared to the run rate in Q4 of last year. Net R&D expenses increased year-on-year by 65%. Net R&D grew at a slightly faster rate than the increase of 51% in gross R&D, explained by a somewhat lower increase in capitalization of R&D expenses. The increase in R&D is explained by higher headcount and an increased number of R&D projects. We continue to step up our investments in R&D to position ASM for the growth pipeline opportunities that we see in the midterm. For the full year of '23, we project net R&D in line with or somewhat above the run rate of Q4 '22. Operating profit jumped 55% year-on-year. The operating margin increased to 31.2% in Q1, driven by the exceptionally high gross margin. In line with what we communicated with the Q4 results, we believe it is possible that the full year 2023 operating margin could be slightly below 26%. We remain committed to the structural target range of 26% to 31%, and we expect the operating margin to move back into this range beyond 2023. Below the operating line, financial results included a currency translation loss of EUR 7 million. This compares to a translation gain of EUR 9 million in the first quarter of last year and a translation loss of EUR 36 million in Q4 of last year. As most of you probably know, we hold the largest part of our cash balance in U.S. dollar and the translation differences are included in our financial results. Then now I will move quickly to ASMPT. Our share and income from investments, reflecting a 25% share of net earnings from ASMPT amount to EUR 9 million in the first quarter compared to EUR 8 million in the fourth quarter and down from EUR 24 million in the first quarter of last year, reflecting the slowdown in the back-end equipment market. Our reported results include an impairment reversal of EUR 215 million. This was related to the earlier impairment due of the value of the stake in ASMPT in the third quarter of last year. Combined with the reversal in Q4 of last year, this impairment has now been fully reversed as per the end of Q1 and reflects the recovery in the market valuation of ASMPT. Now turning back to ASM operations again. Our new orders in the first quarter were EUR 647 million, an increase of 6% year-on-year and a decrease of 20% from the fourth quarter at constant currencies. Towards the end of the quarter, order trends started to soften, including the impact from some pushouts in the delivery in the logic/foundry segment and, to a lesser extent, incremental weakness in the memory segment. Looking at the breakdown in bookings by customer segment, foundry was the largest segment, followed by power, analog, logic and then memory. Combined logic/foundry orders were fairly stable compared to Q4 and year-on-year. Power analog bookings dropped compared to Q4, but were still at a very solid level. As a reminder, Power analog in Q4 included exceptionally high orders for LPE, supported by a strong backlog, we believe LPE continues to be on track for the 2023 revenue target of more EUR 130 million that we communicated last quarter. Memory orders in Q1 dropped both in DRAM and NAND and were at the lowest quarterly level since 2019. Let's now look at the balance sheet. The financial position continues to be strong. We ended the quarter with EUR 573 million in cash, up from EUR 490 million at the end of the fourth quarter. The increase in cash is mainly driven by strong free cash flow of EUR 155 million, and this compares to a free cash of EUR 99 million in Q1 of last year, excluding the cash paid for the acquisition of Reno in that quarter. This increase in free cash flow was primarily driven by increased profitability. Working capital slightly decreased compared to the end of Q4 with modestly higher inventories offset by lower accounts receivable. CapEx amounted to EUR 26 million in the first quarter. We continue to project CapEx of EUR 150 million to EUR 200 million for the full year. A significant part of this will be spent on infrastructure expansion, primarily for our R&D activities. This also includes the first spending related to the EUR 100 million investment that we announced earlier to substantially expand our activities and global development center in Korea. The EUR 100 million share buyback program that we earlier announced will start on April 27, and it will run ultimately until November 15. And in terms of dividends, as a reminder, we announced a regular dividend of EUR 2.50 per share, similar to last year for approval by the AGM on May 15. And with that, I hand the call back over to Benjamin.

Thank you, Paul. If we look at the industry environment, the other parts of the semiconductor market continued to slow down in the first quarter, the uncertain economic outlook, inflation and monetary tightening further dented end market demand. Market segments such as PCs and smartphones continue to be impacted by substantial inventory corrections. TechInsight recently lowered its forecast for the overall semiconductor market to a drop of nearly 8% in '23 compared to a decrease of 4% previously expected. The biggest drop is forecasted for memory, but TechInsight now also predicts a small decrease for the logic/foundry related part of the semiconductor market versus growth that was expected only a couple of months ago. Automotive is still a very spot for which continued growth is expected. Against this backdrop, our results were still good in the first quarter, as Paul just discussed. Our revenue increased somewhat more than initially expected in the first quarter, driven by pulling demand. The fact that we could ship a number of additional tools is in part thanks to the supply chain situation, which has improved substantially compared to last year. However, it has not yet fully normalized as we are still facing constraints and extended lead times in specific areas such as for certain components and specialty materials. Logic/foundry was again the key driver for us in the first quarter as leading customers continue to add capacity in the most advanced nodes. Supported by the record high order backlog at the end of '22, our logic/foundry sales were very robust in the first quarter. As we highlighted on previous occasions, the new logic foundry nodes that went into volume manufacturing in the recent periods have been positive inflections for ASM with strong double-digit increases in ALD layers and applications. Towards the end of the quarter, demand started to soften in leading-edge logic foundry, and this has continued in the recent weeks. We have seen that a few customers in leading logic and foundry have started to push out some to deliveries in certain advanced nodes. In the first place, we believe that this reflects the further softening in end market conditions. Although investments in the advanced nodes are part of multiyear programs, we have observed that some logic and foundry customers are taking a more prudent view to their investment plan for this year, pushing out some to shipments by a few quarters. In addition, we believe that some customers have seen a delay in the completion of new fab infrastructure and as a result, have shifted some of the related equipment investments from '23 into '24. This softening trend in demand impacted our order intake at the end of the quarter. We expect these pushouts to have a more pronounced impact on orders in the second and the third quarters. To be sure, despite the decrease in the second half, overall investments in leading-edge logic and foundry will still be at a very decent level in full year '23. We still expect our strong exposure and leading position in logic and foundry to drive an outperformance for ASM relative to WFE this year. For the next year, our view is unchanged, the leading-edge logic and foundry will remain an important growth market for our company. Importantly, we have not seen a change in our customers' focus to move to the next nodes. Gate-all-around nanosheet technology is expected to enable further improvements in device performance and power efficiency, which in turn will be the key for future data-intensive applications such as in AI. We believe customers finalization of the gate-all-around process is on track with volume introduction planned for 2025. We expect gate-all-around related demand to start supporting our orders from the end of '23. We have already secured tool of record selections for what we believe to be critical and high-growth ALD applications in gate-all-around devices. In Epi, we are also strongly engaged with leading customers, and we expect gate-all-around to drive further share gains. In the memory market, market conditions further worsened in the first quarter, evidenced by further pressure on memory prices. Memory customers are not only cutting down on investments, but have also been lowering their production and utilization levels in order to reduce excess inventories. CapEx is currently largely limited to some modest technology investments. We do not anticipate a meaningful recovery in memory spending in the remainder of the year. Memory WFE is expected to be down by a significant double-digit percentage this year. Looking beyond the current downturn, we believe that ASM is going to benefit once memory spending recovers. Last year, we won a number of new positions in 3D NAND with our ALD gap fill solutions. In DRAM, we expect a further increase in high-k metal gate adoption in the next node transitions. In addition, we have expanded our R&D engagement for new applications in next-generation DRAM and higher stack 3D NAND, which we expect to further strengthen our position in memory ALD. The trend in the analog power sector remains solid. Consumer-related applications are weak, but automotive-related demand continues to be very strong. Revenue in this segment increased strongly for us, even excluding LPE. In the Power analog segment, we are also benefiting from strong momentum with our products, our Intrepid ESA tool for 300-millimeter power and wafer manufacturers continues to do very well. Demand for Sonora, our new 300-millimeter vertical furnace that we launched last year is also very strong with both existing and newly-won power analog customers. With multiple Sonora systems shipped, we had record high vertical furnace sales in the first quarter. A meaningful driver for our power, analog and wafer business is also the Chinese market. As just discussed by Paul, the contribution of China to our sales increased in the first quarter, both compared to the fourth quarter and year-on-year. In addition to power analog, our sales saw increase in the logic/foundry segment of the Chinese market. In part, this reflected the impact from the rebookings in the fourth quarter. As a reminder, in the third quarter, we conservatively derisked the backlog by taking out all orders for the Chinese market that could be impacted by the export controls that were announced on October 7. At the end of November, we announced our assessment that an estimated 15% to 25% of our China sales would be impacted by the export controls. We subsequently rebooked the orders that we could still ship fully complying with all rules and regulations. A meaningful part of these rebookings contributed to sales in the first quarter. Some customers in the Chinese market continue to invest in equipment for the relatively advanced nodes, which have not been impacted by export regulations, while we believe that some other customers have decided to shift their focus to somewhat less advanced nodes. At the same time, we have also noticed a much stronger increase in logic and foundry fab projects in the older technology nodes such as 45-nanometer and older. Our participation in the older nodes is limited to the niche segments of power analog and wafer manufacturers, where we have achieved a solid position as just discussed with our vertical furnace and Epi solutions. However, in order not logic and foundry notes, ASM is relatively underexposed, so we don't really benefit from increased investments in that area. Next, moving on to LPE, let me provide you with an update on LPE. We believe the products and technologies that LPE has in manufacturing and also in development are very competitive. Now that LPE is part of the ASM Group, we already are seeing tangible benefits for customers such as strengthened global customer support. We have seen a strong increase in R&D engagements for next-generation 200-millimeter tools now with virtually all of the leading players in the silicon carbide sector. Last quarter, we already mentioned the first orders from a new U.S. customer, and we believe we are very well placed for additional customer wins in the coming periods. For this year, as mentioned earlier, we continue to expect strong growth for LPE with revenue in excess of EUR 130 million. For the coming years, we expect LPE to further strengthen its position in the fast-growing silicon carbide market. Now turning to the outlook. Following strong revenue in the first quarter, we expect sales to remain at a good level of EUR 650 million to EUR 690 million in the second quarter at constant currencies and again, supported by our strong order backlog. The trend in logic and foundry softened towards the end of the quarter with some pushouts in advanced node investments. This had a limited impact on first quarter orders with a more pronounced impact expected on second and third quarter orders. Due to pushouts in leading-edge logic and foundry and to a lesser extent, further weakness in memory, we now expect our revenue in the second half to decrease by 10% or more compared to the level in the first half at constant currencies. This compares to our previous expectation for somewhat lower revenue in the second half. For the year as a whole, we expect our revenue to increase by a single-digit percentage, including LPE and at constant currencies. This compares to the WFE market, which is now expected to drop by a high-teens percentage in '23. For the market in total, the impact of pushouts in advanced node spending is partly compensated by increased material spending in logic and foundry, especially in China, but this is a market where we have only limited exposure. The longer-term trends remain positive for our company. Despite the correction in '23, structural growth is expected for the semiconductor market, driven by new applications in cloud computing, artificial intelligence and electric cars. Our customers remain committed to their technology roadmaps. The next-generation devices with higher density will be enabled by new 3D architectures, further miniaturization and new materials, which in turn will drive increasing ALD and Epi requirements. In logic and foundry, the gate-all-around transition will drive a significant increase in ALD applications and opportunities to further increase our Epi market share. In memory, we expect to increase our ALD position in next-generation DRAM and 3D NAND devices. In power analog, we have significantly expanded our position. And with silicon carbide, we have added a new growth engine to our portfolio. With that, we finish our introduction. Let's now move on to the Q&A.

[Operator Instructions] All right, Sherri, we are ready for the first question.

So the first question comes from Robert Sanders of Deutsche Bank.

I guess the first question would just be on the China business, domestic China business this year. I think ASML has talked about their business doubling in the current year versus last year. I was just wondering if you could give us some feeling about how much growth you're going to see in the domestic China business in the current year. And I have follow-up points.

So as we explained a little bit earlier, I think we will see probably more business coming in from China compared to the previous years, but it's not going to be at the same level as what perhaps one of our peers have mentioned, and this is explained by the stuff that are really in what you call mature nodes, 45-nanometer and above. There is very little play for us there. I mean, you don't really need ALD at 45, 65, 90 nanometer. So there is very limited exposure for us. So some of our peers would be benefiting from that, but very much less for us.

Got on the follow-up would just be around OpEx. If I just take your comment that you expect to be just below the bottom end of your of your profit margin guidance of 26% upwards. Does that mean your OpEx will actually grow not that much this year, maybe 10%. So just can you give us some idea of how -- how you think OpEx will grow in this year?

Yes. No, of course, Rob. For OpEx, our target the year will be that we keep, let's say, SG&A costs somewhat below the Q4 run rate, the last year, Q4 was EUR 75 million. And our target is to actually stay just below that in the coming quarters. Q1 was maybe lower than what you should think of. But it is expected and targeted to be below the EUR 75 million, somewhat below the EUR 75 million. Then for R&D, Q4 was also at EUR 75 million. I would expect that to be more or less the run rate for 2023, we have some pluses and minuses from quarter-to-quarter. But for the full year, I would expect more or less Q4 run rate for '23.

The next question is from Sandeep Deshpande of JPMorgan.

My first question is on your backlog. When we look at your backlog at the end of last year and the order intake you had in Q1, you were only a couple of hundred million of orders away from what the consensus revenue for the year was. So can you talk about has the backlog being pushed out or canceled? And then secondly, with regard to your guidance on the second half that sales are slightly softer than what you had previously guided to. Can you talk about the companies where you're seeing this?

Sandeep, thanks a lot. So on the first question, yes, I think with some pushouts, we probably have some adjustments in the backlog. And as we have always explained, what we report as backlog is only what needs to be delivered over the next 12 months. So should anything drop out or, let's say, get pushed out as is more than 12 months, that we will have to take that out. And that is the adjustment that we have done as far as the backlog is concerned at the end of Q1. On the second half, the way that we look at this, and we have really tried to give a pretty good view of what we think is going to happen. So a couple of things here. One, of course, we look at some of the logic and foundry, the leading edge logic and foundry push outs. Some of them do have an impact, and we made an adjustment for that. And of course, it's our view that I think memory is going to continue to be weak. The rest should hold up pretty well. And with that, we kind of changed our guidance. Previously, we said it will be somewhat lower than the first half. But now we think that you'll be at 10% or more. So that's the reason why we changed the second half guidance.

The next question comes from Adithya Metuku of Credit Suisse.

Yes. Just firstly, just on revenue. I just wondered if you could talk a bit about the second half, how you see the revenue split between third and the fourth quarters? Are you thinking third quarter will be the trough or do you think it will take a bit longer than that? And secondly, I just wondered if you could give us some color on the Eval tools, which went up sequentially in the quarter. Any color on which products and end markets growth, the increase in events would be helpful. And I have a follow-up if I'm allowed to ask it.

I don't think we will give a split -- a breakdown between the third and the fourth quarter. I think we have shared whatever we could, based on our view, realistically what potentially could happen in the second half a combination of Q3 and Q4. So hopefully, that's suffice. On the Eval tools, you are correct. Maybe let me explain what is really happening in terms of the technology development, which actually drives the Eval tools. So of course, we are right in the thick of working together with all 3 customers on the, let's say, first generation of gate-all-around. And some of those -- some of the Eval backlog or, let's say, a balance is, of course, for tools that are still there, and they are still in the process of development and evaluation. But at the same time, we are increasing Eval tools that will be used now going forward over the next couple of quarters as we ship them for second-generation gate-all-around. So I think I previously shared that all 3 customers are very active. And besides just finishing off the first generation of gate-all-around, work has already started on the second generation gate-all-around. And that's actually helping or in other words, causing the increase in the, let's say, balance of Eval tools. Net-net, it's very positive for us because we are heavily involved with all of them, like in the take of things, and it will take a couple of years, especially for second-generation gate-all-around. But if you are not in there now, then you don't have a chance. So we are starting very early.

The next question is from Francois Bouvignies of UBS.

So my first question is on a follow-up to gate-all-around. So you mentioned in the release that you will see probably see some orders by the end of the year from gate-all-around. Now if we look at what Lam Research said publicly, as I said, they got some multiple customer wins on the ALD locate spacer and that gate-all-around is driving some kind of orders. Can you explain maybe why you maybe being a leader in ALD, you would probably see that at the same time? And what magnitude can you help us on the magnitude of orders you would get by the end of the year? And the second question is on your gross margin, obviously, significantly higher than maybe most of us expected. And you mentioned 2 reasons for that is a mix in China and the within ALD. And I was wondering, China is going to probably continue, I mean, in terms of trend at the moment. So I wouldn't expect the mix to change from China but what about ALD mix. In other words, the mix that we have seen in Q1, should we expect a drastic change in the next few quarters.

Thank you, Francois. So maybe I'll take the gate-all-around questions, and Paul can give you more color on the gross margin. So first of all, on the gate-all-around, let's say, penetration. So all 3 of them are in the phase where they are fast approaching the finalization of the process. Now the way that this works is some of the work that is done is through evaluations, to joint development, but they also buy a couple of tools themselves. And we do have some of that as well. Now specific to what you have mentioned about one of our peers claiming that they have been winning certain applications. We are actually very confident of maintaining our market share, which is the largest in leading-edge logic and foundry. And of course, there will be some wins, some losers, but we are very confident that we will maintain our leading position there. So we are not so concerned about that. In terms of the timing for ordering. So I think if you look back and say, look, all 3 of them are messaging their high-volume manufacturing will start in -- will be in 2025. And then you look back maybe another year where they start pilot manufacturing. You are correct. We do expect to see at least the orders for gate-all-around happening or starting from the end of this year.

Let me then take the question on gross margin. So indeed, there's a few reasons for, what I would say, an exceptional high margin in Q1. As already mentioned, we had a very strong mix within ALD, which for sure we will not have every quarter just to make that clear. The mix in China was good and strong, which helped a lot, but also given that we had some rebookings that were basically taken out at the end of last year, I think that the China sales in Q1, although China sales will continue, as you say, but maybe not to the same level as in Q1, that's still to be seen. And there was also a third reason, especially looking forward in Q1, we still have the benefit of, let's say, some benefit of these of advanced purchase of materials, where there was some cost inflation, but not let's say, to the extent that we expect to see that in the subsequent quarters, while at the same time already in Q1, we had some benefit from increased sales prices already. So there was some small benefit from that part as well, which will be a little bit less going forward into the coming quarters. Having said that, the margins would still remain nicely within our guided range in the quarters to come. So that should definitely be the case. But it's very unlikely that you will see a similar level as Q1.

The next question comes from Amit Harchandani of Citi.

I'm Amit Harchandani from Citi. 2 questions, if I may. My first question goes to the topic of pushouts and your backlog evolution. And I'm trying to join the dots and figure out what's the level of visibility you may have today looking towards 2024. For example, the pushouts, which are going out of your backlog, do you get the sense they are still there, but due for shipments next year? When you talk to your memory customers, is there any talk of rebound, anything on power and analog. So my first question really is, what's the level of visibility and confidence you have in 2024, particularly because you also did say that you expect the operating margin to go back up beyond 2023. So that would be my first question. My second question is really more in terms of what you said earlier about cost increases. So you talked about your own costs going up. Could you give us a sense for how your own pricing negotiations are coming together? Are you a net -- is it a net headwind for you inflation today? If not, how should we think about inflation playing out for you over the subsequent quarters?

So I will take the first questions on the pushouts and backlog, let's say, visibility. Again, what we report in our backlog is what needs to be delivered over the next 12 months. So you are correct. Some of these push outs, especially some of the recent leading-edge so-called logic and foundry pushouts, they will be pushed out probably into 2024 and that -- we do have visibility there. Some of the, let's say, pushouts that happened in the past or previously for memory, they may be a little bit longer, still a little bit uncertain depending on the memory recovery. So I would say it's a mix. But generally, I would say that if we look at 2024, I think we have some visibility into how our backlog is going to convert. But to give an overall view of how 2024 is going to develop, that is still a little bit early. So we will not -- or we will refrain from giving any guidance as far as 2024 is concerned. Paul?

Yes. On the pricing, I think it's fair to say, yes, is there a net benefit there from inflationary cost pressure and sales price increase that we have been implementing. As I just mentioned in maybe in the first quarter, there was maybe a small net benefit because there we were still benefiting from advanced bill material purchases, where we had less, let's say, cost inflation than what we will see in the quarters to come. In the subsequent quarters, I think it will be more or less a wash. I don't think there will be a net benefit. We've put a lot of work and effort in implementing sales prices, and we have been reasonably successful, I would say, not everywhere, but in quite a few cases, we have been successful. And I do believe with that, we will at least be able to offset inflation. But I think it goes too far to say that there would be a net benefit other than maybe Q1, but had a different reason as just explained, that was because of early purchases that we did already early last year and that we still had in inventory.

The next question is from Didier Scemama of Bank of America.

First question is on bookings. Can you help us understand sort of the level you think bookings could reach in Q2 or Q3, so that we have a feel for what your backlog might be at the end of the third quarter? And related to that, on your press release, you mentioned that you've got a meaningfully, I think, stronger position gate-all-around presumably versus the traditional ALD applications. And you expect that to start contributing to bookings in Q4. Should we read from this that Q4 bookings could start to inflect sequentially? Or are you not prepared to make that comment?

So I think we will share -- we have shared that basically, Q1 bookings were slightly impacted by some of the logic foundry or leading as logic foundry pushouts. And we have also shared that this would have a more impact as far as the second and the third quarters bookings are concerned. And we will probably just leave it as that as we do not guide bookings anymore. The gate-all-around impact in the fourth quarter. So this is what we can say. As the 3 companies start to go into preparing for pilot production, this should happen in 2024. And if we just work back on late time and so on, we do expect that some of these orders are going to land in the fourth quarter. And we do expect that that's going to help us with the order situation exactly in terms of timing, whether all 3 of them do at the same time or again, there's a difference of a couple of months and so on. I think that at this moment, it's a little bit difficult to tell.

All right. And then on the meaningfully better position in gate-all-around, would you be prepared to say that once you are in volume production, let's say, 2025 or 2026, your gate-all-around market share, let's say, for ALD, I guess, would be better than 50% to 55%? Or you're not prepared to make that comment?

It's probably, again, a little bit early for us to make that kind of guidance. But perhaps I can share with you this way. So we do think that we are in a position where we will maintain our leading position as far as overall leading-edge logic and foundry is concerned. And based on the traction that the engagements that we have with all 3 main customers, we think that we are in a very good position to continue doing that. I think once the process development is finalized, probably middle of this year or, let's say, a little bit in the third quarter, I think we would have a better view whether we have gained share. And I think we are prepared to share that information when we have our Investor Day later on in the second half of this year.

The next question is from Lukas Tamulevicius of Jefferies.

This is Janardan Menon dialing in on Lukas' line. 2 questions from me. One is on the pushouts on the foundry logic side. Given the size of your backlog and the fact that China presumably is interested in taking equipment when they can. What are the chances that you could ship some of the pushed out machines to customers other than those who pushed out, especially those in China. Is it that these machines have already started being built and therefore cannot be recustomized? Or is it that they are more high-end systems, which cannot really be sold into China at this point in time?

Sure, Janardan. I think it's a mixture. I think some, if we can do it, of course, we will repurpose or reposition them to meet more, let's say, urgent demand. But I think in general, when you look at leading-edge logic and foundry, the ones that are at 5-nanometer and below, these are equipment that you they are slightly different, I would say, and that you don't need at the 28-nanometer node. So there is no one-to-one type of just repurposing. But of course, again, if we can, we want to do it, and we have been doing that. Some of these, I would say, leading-edge logic and foundry equipment, we have them, let's say, we have them in our build plan, but some of them are in various stages, some are being built. Some are actually waiting to be built. So it's all over the place. I cannot give you a definite answer except that when we can repurpose, we will.

So if you can repurpose, is that upside to your current guidance? Or has that been somewhat factored into your current guidance?

I think if you look at 2023, as a whole, what we have guided for the first half and the guidance that we have given for the second half, I think we have given, I would say, a realistic view of what potentially would happen. I think we are quite clear on the first half given that when we announce the first half type of, let's say, guidance, it was only 2 months ago. So we had a good view. Second half now, we also have a better view. I think it's a pretty realistic picture. Any guidance or let any upside or whatever has already been baked into the forecast.

And just a follow-up on the spares and services. It was quite a big jump quarter-on-quarter and well above your trend previously. Can you explain why -- what caused that jump? And how sustainable is this new level of revenue? And what kind of growth we can expect there going forward?

Yes, we had a good bump up the expense and services. I think we grew by 21% in the last quarter. But the spares and services business, they tend to be a little bit lumpy. Some quarters, you see they're growing much more than others. And part of it is, of course, it depends on what kind of services customers require, what kind of contracts we have. But at the same time, also whether we are able to deliver all the spare parts that our customer wants due to supply chain constraints. And that's why it makes it a little bit lumpy. Q1 just happened to be great, and we hope it will continue. But overall, I think longer term, we do believe that the spares and service business is going to grow at a very nice growth rate for us over the next couple of years.

The next question comes from Timm Schulze-Melander of Redburn.

The first question was on tool development. You've talked about being deeply engaged with customers on gate-all-around Gen 2 development. Could you maybe just talk about what kind of synergies might be available to you given that you are working with all 3 major players? And does that mean that R&D pressures maybe can moderate in 2024? And then I have a follow-up.

Thanks, Timm. That's an interesting question. We, of course, as far as gate-all-around, the -- in terms of ALD, in terms of epitaxy, we're probably the bigger player now. As to synergies, what we do with each customer is specific or, let's say, confidential to each customer. So we also cannot, for example, start sharing them around. Now of course, internally, you could say, yes, well, you have developed a specific process, and you don't have to redevelop it. There is some of that. So there are some synergistic effects. But I would also like to caution that all 3 customers have very different designs. And with the very different designs, there are differences in what they use in terms of materials and the kind of -- the different kind of processes that they do. So there are some synergies, but it's not like a ton of it.

Okay. All right. Helpful. And then just sort of playing around some of trying to get a handle on some of this rescheduling scaling and the comments you made around Q2 and Q3. Is the right way to think about this that it's kind of a mid-double-digit value that was rescheduled coming out of Q1 and that maybe rises to more of an upper double-digit number in Q2 -- in Q3.

Again, we won't give a specific number as far as Q2 or second and third quarters are concerned, except to say that there will be a more pronounced impact compared to the first quarter. I think in the first quarter, we came in just a few percent lower than what was a consensus or, let's say, a couple of percentage points lower than where we ended previously. I would say that just assume that it will be higher than that in the second and third quarter, but we won't give a specific number.

The next question is from Tammy Qiu of Berenberg.

So Benjamin, please. On your foundry logic pushout, can you share if that bleeding edge or that leading edge? And also, at the same time, when you talk about pushout to 2024, what is the kind of market should I be thinking about for 2024? Should I be thinking about then 2024 will be adding this pushed out order or 2024 -- or that will be pushed out into 2025?

Sure. I think of course, whether how you define leading and bleeding edge, I would just say that maybe bleeding edge is what you call the 3-nanometer and the gate-all-around stuff. And leading edge is probably the 5 and the 4. Actually, we are mainly focusing on anything that is 5 and below for us as what we call leading edge. I think you are seeing pushouts primarily in 5 and 4. We don't really see a lot of changes and especially gate-all-around. They are still executing to the technology roadmap, the timelines. We don't we don't see any changes. We don't see anything that speaks to the contrary. So the pushouts that we are seeing in what you would classify as leading edge or 5- and 4-nanometer. I would say a couple of factors that we have explained. I think we just have some customers that are being a little bit more prudent in terms of CapEx spending for this year. We are also well aware that there have been fab construction delays. Fab construction delays, delays in getting permits, et cetera, that is kind of delaying the tool, the moving. So it's a combination of factor, specific to each customer, it might be different, but that is what we are seeing today.

Okay. And how should I be thinking about 2024, please? Should I add on 2024 number?

Yes. I think what -- when you look at what's happening, of course, on the 5-nanometer and below, the investment, of course, is a multiyear investment. But having said that, it is possible that even with a plan, a 3- to 4-year ramp, sometimes things get moved out 1 or 2 quarters. So it's not unusual. I mean that has happened already also in the past. So I think the way that the trade customers are looking at this is they're going to execute according to what their road map is. For example, that gate-all-around high-volume manufacturing is going to land in for 2025. Now whether it lands in Q1 or whether it lands in Q3, I think there's probably some room there, but it will land in 2025. So I think that part is fairly firm. Now if we look at 2024, first of all, I think it's a little bit early to really give a full kind of guidance. I think what we are seeing is that some of the 2023 stuff that were planned for this year is being moved into 2024. We think that there is a possibility in a way that they could be incremental to 2024 plan. But it also could be if the demand is not there, the demand pushed out, let's say, what was planned in '24, a little bit further. But at this moment, it's too early to tell and too early to say anything.

The next question is from Stephane Houri of ODDO.

Actually, I've got a question on silicon carbide expectations. So for the year, you have confirmed more than EUR 130 million. But at the same time, in some communications, you were saying that the 25% CAGR is becoming a little bit conservative. So can you share with us your view on your future, let's say, potential in silicon carbide and talk about the commercial advancement. And today, with the current capacity that you have, what kind of revenue you can reach?

So yes, we did share -- first of all -- I think we, first of all, shared that the CAGR was 25%, but we are kind of figuring out that actually, the CAGR is probably higher. We don't have a specific number to share and we will probably have to wait until our Investor Day. And then we initially shared EUR 100 million and then we upgraded this to EUR 130 million because of much more better demand. What I can say for the silicon carbide business is it's probably right now, at this moment, the strongest part of anything that is in the semiconductor space. And that is primarily driven by all the electric vehicles, the renewable energy stuff and so on. We are very heavily engaged especially, I would say, on the 200-millimeter transition and the 200-millimeter transition is going to be, I would say, a big, let's say, opportunity for not just for LPE, but also for their competitors because you could see fairly significant investments in 200 millimeter over the next couple of years. And we believe that we are in a very good position because we are particularly engaged with every silicon carbide device manufacturer or wafer manufacturer. And at this moment, we are involved in a lot of discussions looking at a couple of options, a straight purchase of the first tool for 200-millimeter evaluation, et cetera. So we are right now in the take of it. And over the next couple of months and hopefully, by the time we get to the Investor Day, we will be able to update everybody what is the progress. We also look at in terms of our capacity. Of course, LPE in Milan was a small facility, and we are trying to expand there as much as possible. But at the same time, we're also having to look at what about using our other so-called ASM facilities to help them with the increased demand so that we can have enough capacity. So all those are being looked at, at the same time.

The next question is from Ruben Devos of Kepler.

Yes. I had a question on your guidance. Basically, you've provided a very helpful guidance for this year, but looking a bit further out, and I know you've talked already about leading-edge logic and foundry, but very curious if you could talk a bit about memory, which you said before, is a bit of a wildcard. And could you also talk about the strong spending in the older nodes within logic and foundry? And how that maybe affect your level of confidence heading into next year? And then the second question is just on the current lead times for your orders. What are they? And what is basically the current state of the supply chain?

Sure, Ruben. So maybe on the memory piece first. I think we are seeing, first of all, the memory customers messaging that they are still having a fairly difficult time with price going down, inventory corrections and they have resorted to reducing, let's say, or cutting capacity -- production capacity and utilization to avoid further, let's say, ASP deterioration. I think that's all starting to have an effect. And hopefully, the demand comes back and they are able to perhaps even increase the level of manufacturing. In terms of timing, I think it's different from the memory chip manufacturers and the equipment manufacturers. We need to wait for them to add capacity. And the way that they add capacity is when they have run out of capacity on the existing notes and they want to add capacity in the new nodes. Now that one is a little bit more difficult to time. It already dependent on when memory prices come back and demand comes back. But generally, we think that as far as 23 is concerned, we will probably not see much of that. And mostly, it's going to be a technology additions. So while they are going to this correction phase, actually all the memory chip manufacturers are still continuing with their technology development. And we are, of course, working with them. And we do see some technology buys, which we are supplying to them. On the mature nodes, again, this is something which is very interesting because if you look at where ASM is coming from, we are really a leading-edge type of equipment supplier. So the more leading edge, it is, the better it is for us. So when you look at mature nodes that are, let's say, 28 nanometer and beyond and older, there is really a very limited play for us in that sense because at those nodes, you probably do not need ALD or you need ALD maybe just for the high-k part of it. So there isn't really a lot of play for us there. Now where we do play in the mature nodes, which is in the old technology, which is maybe 90, 130 nanometer and so on, is in the power analog sector, which we have actually been building our position nicely over the last couple of years with Epi and also with vertical furnace. And as we shared earlier, we even had a record, let's say, amount of revenue as far as vertical furnace is concerned in the last quarter. So we are doing well there. Now beyond where we are playing today, we don't really have a lot of play in terms of, let's say, 45-nanometer, 65-nanometer logic or foundry. That is actually not our, let's say, sweet spot. In terms of lead times, I would say the lead times is getting better. We're probably coming down to the 6 to 8 months range that is more acceptable. In fact, we also still get a lot of pressure from customers asking us to reduce our lead time plan and we are doing that. But a little bit on the supply chain constraint, I think it's gotten much, much better compared to 12, 18 months ago. But again, as we shared, there are weaknesses in certain components, primarily gas delivery and also in specialty materials, which the weaknesses still persist, but we are working through that.

The next question is from Marc Hesselink of ING.

Yes. First question actually on the second generation of gate-all-around. I would first generation, you guided that there's a quite significant step up in ALD intensity and more than the usual setup. What do you expect for the gate-all-around?

Of course, the second generation has kind of started. We think that there would be even more -- I mean if we were to give our view today at this moment, we think that there will be even more usage of ALD primarily because there's even more new materials that will be -- will have to be used. And new materials generally is good for ALD. So that's where we are today. Of course, you have the usual scaling, maybe there's a little bit of change here and there that could also lead to more ALD. But overall, I think the main driver of, let's say, ALD intensity being increased is new materials.

Okay. Clear. And the second question is on -- you said there's a part of the order book is pushed out. Can you quantify that? How much is being pushed out, that's now more than 1 year ahead?

I don't think we will quantify that. But if you look at, for example, of course, those have been pushed out for more than 1 year, we have to adjust the backlog and then we have the new bookings coming. I would say that it's pretty evenly well-balanced, stuff that has been pushed out and I'm making a guess. I'm lucky looking at Paul to see whether knows the answer. I would say that stuff that has been pushed out more than 12 months have been balanced by staff that are coming in as well. So net-net, not much of a change.

Next question is from Nigel van Putten of Morgan Stanley.

I have a question, already a follow-up on the order behavior by your customers. So clearly, most of what you're seeing is customers reacting to a weaker environment. But I guess, at the same time, the supply chain situation is also improving. So my question would be, where are your lead times now? And to what extent have they improved over past months? That's the first question.

So you are correct. We do see improvements in the supply chain. And as I mentioned earlier, we are getting back to the more normalized lead time, which is between 6 to 8 months. We are not 100% there yet, but we are getting close.

All right, 6 to 8 months. As a quick follow-up then, would that be sort of would your gate-all-around tools current iteration -- would that be a similar lead time? Or would you expect that to be longer considering the tools on your -- just also getting a sense of if you do see those orders coming in, in the fourth quarter, I mean, if the customers can get those relatively quickly from you, would that be a more normalized pace from what we've seen maybe a couple of quarters ago? Or any color around that would be useful.

I would say it will mostly be around the same. There could be, for example, what we call a very specific application that perhaps we need to do a little bit more, which might need a little bit more time just because of -- maybe we have special hardware requirements and so on that we need to do that. But I think the way to look at it is it's probably the same as the other lead times or the usual lead times that we have, which is getting back to a normalized state of coming close to 6 to 8 months.

Got it. And just a quick clarification, then if you say 6 to 8 months, then a normalized backlog would also basically mean revenue for the next 2, 2.5 quarters. Does that sort of net hold up?

If under normal circumstances, that should be the case. But where we have on the backlog, of course, last year, there was actually quite some, what we call, early ordering. And that's because we had lead times for certain products that extended up to even a year. So I think as we get things normalized supply chain gets normalized, we should be looking at that.

Yes. Yes. Understood. It is sort of surprising to hear 6 to 8 months. I mean if you look back a couple of years ago, then I think I remember it being more like 3 to 4 months also if you compare the backlog back then? Is that a particular reason why that's changed? Is that just the complexity of your products? Or is that still sort of supply chain complexities that are elevated now? Sorry to put in the extra question.

No. I think it's a mixture of the several factors that you have mentioned. It is true that it is some of the equipment are becoming more complex. That's one to. Supply chain has also become more complex as well. And all of that added up together is leading to a longer, let's say, lead times of 6 to 8 months. Now of course, there are certain products which are a little bit faster, some of which are lower. I'm just giving you an average number.

The next question is from Gianmarco Bonacina of Equita.

A couple for me, please. In the second half, you mentioned that you may have a 10% decline versus first half or even more. So given your backlog, what at the moment, worst-case scenario in terms of deviation on the downside from the 10%? Just a couple of points, could be worse and what could be the, let's say, the biggest risk here. The second one is just on the FX impact because you talk about, let's say, constant currency. We noticed in the first quarter, you already had a headwind on FX despite the U.S. dollar should have given you a tailwind. So at current FX, what's your best guess in terms of headwind for there? Because I calculated EUR 2.6 billion in constant currency revenue, but clearly, we have to deduct also the FX. It could be a mid-single digit or probably less?

Gianmarco, I will let Paul answer you on the FX scenario question. But I just want to clarify your question on the -- regarding the backlog. So you are saying that we have messaged a 10% reduction in the second half? And how does this impact the backlog? Is that what you're asking?

No, I think given you have a backlog, so how much worse can be minus 10%, minus 15%, minus-- because it seems you have a backlog. So it shouldn't be much worse than minus 10% at this point in time because we are in April or you envision a scenario where actually the decline in the sales second half versus first half could be significantly more because in the press release, you talk about 10% or more. So what does it mean more? Basically, it's just a tiny bit or could be also, let's say, a bigger deviation from this 10%.

I think we'll leave the head as it is. But what I can tell you, Gianmarco, is really we have given in our opinion, a very realistic forecast as of today, what we know today. And we do think that the 10% or more reduction in the second half is going to happen. Now of course, that's based on the best view that we have today, and we want to leave it at that. Paul?

Yes. On the FX, yes, Gianmarco, it's -- if we say -- let's say, if we guide, let's say, take in this case, 10% or more for the second half and we say constant currencies, what we refer to is that, that is based on an assumed, let's say, currency level that we have seen in the -- in the last quarter of our reporting. So in this particular case in Q1. So if you would make the same comment in, let's say, after Q2 reporting, we mean it's the currency that we have seen in Q2 on average, et cetera. That's what we mean by constant currencies. So we're not talking about headwind tailwind. We just say we -- in our forecasting, we always use the latest available currency, and we assume that that will stay because you can put an enormous amount of time and effort and debate and intellectual business in trying to forecast currencies. They will be as good or as bad as keeping it the same. We keep it the same and then we'll see what happens. But as you know, in the actuals, we will always convert it back in terms of growth rates at constant currency compared to prior year. But when we guide forward, we look at the last available currency of the last available quarter. I hope that's clear, but that's how we do it.

Okay. Clear. And just a quick follow-up. In the Q1, you had a headwind despite the U.S. dollar being, let's say, giving you a positive effect, I guess, because of shipment in other currencies. So the Q1 was kind of an anomaly or I mean, it's not, let's say, an anomaly in terms of country mix the Q1?

The revenue, I believe, in constant currencies, measured in constant currencies was slightly lower, the growth than measured in reported currencies. Is that what you further, I'm not sure I not sure what you mean...

No. I'm saying in the Q1, you had a 40% constant currency revenue growth and 37% reported. So basically, you are saying you had a negative impact of FX because you are adding this. And I noticed that actually in the Q1, other companies, they are having a benefit from the U.S. dollar year-over-year. So probably because you are shipping in other currencies like the renminbi or the Taiwanese dollar or the -- I guess this is the case. Now you are not shipping that much in U.S. dollar. It's more other currencies in this quarter.

I don't know what the functional currency is of other companies that you refer to maybe U.S. companies and if they see the opposite of what we see, that's true. But the bulk of what we invoice is in U.S. dollars, not everything, but the vast majority is in U.S. dollars and compared to the level of Q1 last year. Indeed, there was -- basically, there was this negative impact that you just said, of around 3%, if you measure it based on reported currency. If you compare it to Q4 of this year, for instance, sequential, there was a benefit. So it depends on the reference period that you take, and it depends if you report in euros or if you report in U.S. dollars. So dollars have, of course, a different impact than what we have when we reported euros.

Thank you, Gianmarco. Operator, can we please have the last question?

Yes, sir. The last question is a follow-up from Adithya Metuku of Credit Suisse.

Yes. Just one question really on LPE. I just wanted to understand what drives your confidence in the share gain you're expecting in 200-millimeter silicon carbide or tools. Is this based on the attributes of the tools, is just based on what you're hearing from the customers you're engaged with. And because your competitors also seem to be pretty confident about maintaining the share they've got. So any color you can help with -- on LPE share gains going forward would be very helpful.

Sure, I think it's a combination of all of what you just discussed. Of course, we do our own work in terms of the performance of our goals, the cost of ownership. But at the same time, the rubber really hits the road when you go into discussions with customers. and then you get the real feedback. And I think based on all the engagements that we have had so far and are still having, we, I would say, have very good responses and the feedback from the customers. And that's why I think we are in very active engagements talking to them, you better straight purchase evaluations, joint development, I think it's because they also see that, look, after doing some work with us, it is a good product. It has good cost of ownership, and that's why they are thinking about it. So that's how we alive in terms of our confidence for 200-millimeter.

I'd like to hand it over to the CEO for any closing remarks, please.

I just wanted to thank everybody for attending our call today. Also on behalf of Paul and Victor. We hope to meet many of you during one of our upcoming investor events virtually or in person. Stay safe, and goodbye.

The conference is now over, and you may disconnect your telephone.