VAT Group AG (VACN) Earnings Call Transcript & Summary

Ladies and gentlemen, welcome to the VAT Q1 2023 Trading Update Conference Call. I'm Andre, the Chorus Call operator. [Operator Instructions] The conference must not be recorded for publication or broadcast. At this time, it's my pleasure to hand over to Mr. Mike Allison, CEO of the VAT Group. Please go ahead, sir.

Thank you, and good morning, ladies and gentlemen. Welcome to VAT's Q1 2023 Trading Update Conference Call. With me this morning are our CFO, Fabian Chiozza; and Head of Investor Relations, Michel Gerber. After my introductory remarks, we will start the Q&A session and the call moderator will take your questions in the order you enter them. So as you can see from the media release this morning, the first quarter of 2023 is kind of a mixed bag of results. On the positive side, we've achieved revenues that were slightly ahead of the guidance high point at CHF 233 million. The strong execution of our order backlog contributed around CHF 100 million in the first 3 months of the year. Nevertheless, overall sales were down 12% year-on-year as we approach the bottom of the semi cycle. Across the 3 businesses, we saw the following performance, and please note that we moved the Former Display business into the Semi Business unit and the solar activities into the Advanced Industrial business unit as of January 1. In Advanced Industrials, we achieved 28% higher sales year-on-year and 2% higher sales in Global Service. However, this was more than offset by the 23% sales decline in our largest business unit, Semiconductors. On the rather negative side, order intake declined strongly by 54% year-on-year to CHF 136 million, leading to a book-to-bill ratio of 0.6. On a sequential quarter-to-quarter basis, orders also declined substantially and were 45% below the level of Q4 2022. The net order intake in Q1 2023 also includes order cancellations in the range of CHF 25 million. These cancellations were predominantly from semi customers and reflect the inventory corrections going on across the industry. We estimate that this number will slow down in future quarters. A focus during the quarter were the continuing operational cost adjustments to align the structure to the current business volumes. As this process was already initiated back in the third quarter of 2022, we are confident that we can maintain the margin resilience that VAT has successfully demonstrated in previous down cycles. This operational cost discipline is important as we navigate the uncertainties of the down cycle. We believe the bottom will occur in the first half of 2023, but given the macroeconomic challenges, we're constantly adapting our operational plan to ensure we successfully deal with the market conditions. We are, however, not sacrificing R&D investments nor the ability to ramp up our production sites, as you will hear later in the update. So what do we believe 2023 will look like? It's very hard to say exactly due to the constantly adjusted forecast and you will have all seen that there's a widespread of opinions as to how the market will develop and by when we can expect the semi industry to turn positive again. Our working hypothesis is that we expect a low point in the first half of the year and then a gradual improvement into the second half of 2023 as our customers prepare for a ramp-up in 2024. We still see unprecedented midterm demand for the semiconductors, which is clearly seen in the high number of new fabs getting built around the world but a variety of macro factors currently weighs on chip demand from both the business and consumer segments. In this environment, investment in memory capacity is especially hard hit, may be down by as much as 40% this year as market research indicates. The expected decline in the area of advanced logic for the use in data centers or artificial intelligence is currently only about 10%. Additionally, we see some strength in the legacy 200-millimeter fabs supplying areas like automotive, power and microcontroller chips as demand is still very strong in these areas. Taking all this information into account, we expect Q2 orders to remain at a low level, but cancellation should be less pronounced than in Q1. Sales-wise, you have seen our guidance of CHF 200 million to CHF 230 million in the second quarter. These sales will again be supported by the ongoing backlog execution, which we expect to be again around CHF 100 million, leading to a backlog after the first half of around CHF 300 million and a good basis for our second half sales development. We expect the second half order intake to pick up again as the OEMs prepare for a rebound in the first half of 2024. And the shortening order to sales conversion times driven by reduced lead times will contribute to sales in the second half. As a result, we expect sales in the second half of 2023 to be slightly above those in the first 6 months, but again, this fully depends on a capital equipment rebound in the first half of 2024. So what did we see for the business units in the first quarter? And what do we expect in 2023? Global Service had a slow start of orders in Q1, while revenues were in line as previously mentioned. Service customers are still sitting on slightly elevated inventory levels in spares and consumables but we expect this to improve as fab utilization starts picking up again in the second half of 2023. In addition, the upgrade and retrofit demand is expected to gain traction as the year moves on. So we continue to expect growth here. In Advanced Industrials, we have seen a weaker first quarter order pattern. This, however, has to be seen in the context that Advanced Industrial customers typically spend their budgets in the fourth quarter. In addition, we did see preordering due to the price increases, which were effective from January. The Coating business started the year on a weaker note but scientific instruments, research spending and solar all saw positive gains. Again, 2023 should see growth in the Advanced Industrial business. In the Semiconductor business unit, we have seen the expected sharp drop in sales and orders driven by the WFE CapEx decline of about 20%. Expected market share gains and an expanding adjacencies business will not be able to offset the significant negative impact of the semi [indiscernible] cycle. We did see some very positive developments in Q1 with some strong advanced module spec wins and fresh qualification orders for our new high flow ALD valves. We also continued to see strong spec wins across the industry. This shows the importance of maintaining continuous R&D spending across the cycle. In summary, and as already communicated with our full year 2022 results, we expect lower full year 2023 sales, EBITDA, EBITDA margin, net income and free cash flow. I elaborated on our order and sales pattern expectations for 2022 already. On the cost side, we started to take measures already in Q3 in 2022, and we will continue throughout 2023. As a result, we are optimistic that we can be around the low end of the EBITDA margin band of 32% to 37%, despite a substantial sales drop. The EBITDA margin for the first half, however, will be below the 32% threshold. Achieving this full year result would represent a full 5 percentage points improvement in our EBITDA margin compared to the 2019 down cycle. You can expect that VAT will follow the same model as the last cycle, that is strong continued investments in R&D and infrastructure so that we can take full advantage of the next growth cycle. As such, there is no change to our investment in our new Swiss R&D center and also to the new production facility in Malaysia, which will be ready for qualification in the second half of this year. Spending in R&D, both in the core business and in adjacencies will increase, and we are confident to continue our organic growth in the future. Therefore, we expect CapEx to between -- to be between CHF 80 million to CHF 85 million in 2023, ensuring strong long-term success on organic sales growth. Last but not least, we will closely manage our trade working capital to support our free cash flow generation despite higher CapEx. This concludes my prepared introductory remarks, and I'll now turn back to the operator for the Q&A session. Thank you.

[Operator Instructions] The first question comes from the line of Sebastian Kuenne from RBC.

My first question was on your comment on the order backlog. Is it correct that you assume CHF 300 million by June and if so that would imply, if you have CHF 200 million to CHF 230 million revenues, it would imply another quarter with orders of somewhere between CHF 130 million, CHF 140 million. Would that be correct?

I think directionally, that's correct. Hard to tell exactly where orders will be in the second quarter. We expect them to be above the first quarter. But directionally, I think you're in the right area.

Okay. And then yes, on the cancellations, you had about, I think, you calculated some [ 9 million ] cancellations. And for your existing order book, do you expect similar trajectory of cancellations going forward and also is there a risk that you have to book out orders that you have in your book, but because of project delays, you have to remove them from the order book. So I just want to get an idea of the quality of the current order book.

Yes. We manage our order book very carefully. I did mention in the remarks that we saw around [ 25 million ] of cancellations. Okay. It was slightly higher than that, but we expect that number to drop in the second quarter. We have a very close relationship with our customers, and it's really in our best interest to quickly align our production plans together. There's no point in us shipping as much as we can in the first quarter than seeing a very low production volume in the second quarter. So we've learned through experience and through previous cycles that it's better to work with our customers and very quickly align the backlog. What we try to do, obviously, is where we have goods in production or in shipment, our customers normally take those goods and we work with them. But when we have orders for maybe 4 to 5 months out, those are the ones that we cancel or adjust or reschedule. So it's a very dynamic process. And I think together, we worked very closely with the customers to ensure the best outcome for both. And as I said before, it's really important that we align very quickly on what the coming 2 to 3 quarters are going to look like.

The next question comes from the line of Didier Scemama from Bank of America.

I guess my question is as follows: Mike, would you say that the current cut to production at the memory customers, SK Hynix, Micron, Samsung, et cetera, and elevated inventories at your direct customers might create a delay, if you want, between a pickup in demand for memory chips, and orders and recovery for your own business? Or do you think that we should not worry too much about that?

It is a dynamic situation. I think the cut to memory output is the right thing. Pricing is everything to the IDMs as you could see from the recent Samsung results. And I think getting -- realigning their output is a smart move. I think the logic spending is still pretty strong. We don't expect to see much production in logic and foundry, maybe about 11%, but very advanced nodes are getting good traction. So we still expect strong WFE into that segment. I think you're also seeing increased spending in artificial intelligence through data centers, which is driving advanced logic, but it also has the need for the most advanced memory chips , both DRAM and flash. So I think you will start to see in 2024 investments coming into capacity expansions in memory again. We mentioned in our remarks that we're expecting a pickup in revenue in the second half, and that really is on that premise that our direct OEM customers will start shipping equipment in the first half of 2024. Therefore, we need to get shipments into them by the fourth quarter. So I think that's the most likely I'd comment at this point, but if you consider that the equipment shipping into the memory manufacturers then in the first half of '24 will only contribute to output in the second half of 2024 because it takes some time, obviously, to get the systems qualified and producing chips. So that's still 18 months away. And I think there's a pretty good chance that capacity needs will be back up again in that time period. So I think that's the most likely hypothesis at this point.

Okay. Thanks for the color. That's helpful. And on your comment, I think you sort of alluded to it already a bit on your comments on the strength in AI and data center. Could you characterize a little bit more the leading-edge logic market? I mean, have you seen any push outs or any cancellation or anything like that? And I think related to that, maybe talk a little bit more about your success in EUV and in ALD as well.

Yes. I mean we -- from what we see, and obviously, we can't triangulate this directly in our sales. But in talking to our customers, we see strength in all the logic segments. So that's foundry, advanced logic and mature logic, that seems to be progressing quite well. Most of the costs apparently are in the memory area, again, but we can't correlate it at [ 1:1 ]. The -- starting first with the mature logic. That tends to be our older technologies. Our share in those areas is not as high as a leading edge. So that doesn't give us the full potential to exploit the share gains we've had in recent years, but it's still an important business for us. As you move into advanced logic, that's where we really start to see improvement in our share and also the percent of wallet we get. That's very strong in ALD and from talking to the OEMs participating in that area, we expect growth in the ALD businesses. They're really critical for 3- and 2-nanometer processes and we should see them grow during the year. The same with EUV. Although I think we're already -- I think the [ SML ] already passively limited and [ improved ] for this year. So I'm not seeing that grow, but it's staying steady as we look at to 2023. Our share of wallet is improving nicely in EUV and looks very good into '24 and '25 as we grow our business there. So 2 very important segments for us. But also I think don't underestimate the impact of edge in the leading-edge logic areas as well. We have very strong share there and edge is a critical component on the 3 and 2 nanometer chips. So these are all very strong for us and will certainly help us this year.

Just 1 clarification of the ALD sort of design win, is it a new customer? Or is it an existing customer?

Well, really, they're all existing customers. We really are heavily supporting all the OEMs in the top 20 list. So they're all existing customers. But these are share of wallet gains at these customers.

The next question comes from the line of [ Christoph Grau from AWP ].

Can you maybe give me some more details on the development of the regions, and we have the losses in order intake being more severe. Can you maybe clarify it a little bit for me?

Yes. Memory is very much an Asia-focused business with the major memory sites in Korea, China, Taiwan and Japan. So actually, Singapore also. So all the reductions in CapEx are coming in those regions. Europe and North America are quite largely focused, both in the mature area and also advanced areas. So this should do pretty well out of this. And also the ongoing government stimulus packages the U.S., [ Chip Bank ] and the pending European one are also creating quite good investment grounds for the chip companies, again, especially in the 2 logic segments. So I think it's really Asia that has been impacted by the memory segment at this point.

Thank you very much. And you just mentioned the U.S. CHIPS Act, how, what's the impact in the last quarter? Can I ask you maybe some more details. Last time, you said you are not truly clear how much you have been impacted or you will be impacted, maybe we have some visibility on this now.

Sorry, on the impact of the CHIPS Act, is that the question?

Yes.

Yes, I don't really try to quantify the CHIPS Act because these packages are just incentives to move a chip plant into the U.S. or Europe, probably from Asia. So I only look at the total labor fab equipment, budget and outlook, and that really remains unchanged. Either the fabs in America or as in Taiwan, I don't really care where that is. But I think the impact of the CHIPS Act is not substantial this year in both Europe and America. It's mostly going to be a 2024 story, especially in the U.S. where we'll see the biggest impact.

The next question comes from the line of Jörn Iffert with UBS.

Michael, the first one would be a follow-up on your statement that the second half sales should be a tick better versus the first half and is this more a top-down view because we have some market forecast, CapEx is going up in '24 or do you actually really get already some capacity renovation requirements from your customers for the second half? So just to check a little clarity you have for the second half? And the second question would be, please, on your profitability metrics. Where do we stand here right now? I remember you said 30% of your workforce in 2022 [ at 10. ] Is this fully reduced now? Is short-time work already in discussion and also, can you just remind me what you do with the capacity expansion outside there in the recent times in Switzerland? Is this a little on hold now? Just to check this also from the cost side.

Yes. I'll let Fabian answer the second half, but let me touch on the first part first. Yes, visibility is still, I'd say, pretty cloudy in the second half. I mean, we're in close dialogue with our customers. And the challenge, as you know, when the business comes back, it comes back pretty quickly. So we're in dialogue as to what their needs are in the second half and into the first half of 2024. Again, if you -- it is kind of a top-down view but if you just extrapolate out to what the needs are going to be in chip demand at the end of '24, as I mentioned, and bring that back as to when the equipment has to be manufactured. You start to see the needs for growth in the first half of 2024. So it's a working hypothesis right now. And hence, the importance of us realigning our cost structure as much as we can. And I'll let Fabian elaborate a little bit on that for the second half of the question.

Thank you, Mike, and also good morning, Jörn. Look, we have already started to adjust our cost structure back in Q3 last year where we went in with a temp workforce, as you said, close to 30%. And at this point of time, we have reduced that significantly. Right now, we're looking at around 5% remaining. And besides all the other cost reduction measures that we have in place, including overtime reduction, vacancy reduction, cut on discretionary spending, et cetera, short term work is also one of the further options that are certainly being explored. With regards to the second part of your question about our ongoing capacity expansion programs. This is absolutely unchanged, and we continue to build out this additional capacity maybe in Malaysia at full steam. The facility will be ready for equipment moving in August this year and then qualification will follow by the end of the year. So overall, continued focus on investments, including R&D, but also the preparation for the next cycle with the continued expansion plans.

One other point. Just I think an important point, Jörn. As you know, we outsourced about 75% of our building materials. So as we bring on the second factory in Malaysia, we can in-source a higher percentage of our machine components to ensure we get this factory ready and qualified, and we didn't match to our first factory there. Now that gives us tremendous operational flexibility. We then have 2 factories qualified there. And as the market ramps, we can expand both in our factories, but start increasing that percent of outsourcing again. So it's also a way to help us navigate the development of the supply chain in Malaysia, which is nowhere near as sophisticated as a supply chain here in Central Europe. So it's a pretty nice strategy, again, really developed for this flexible operating model we have.

And really just a 5-second question for clarification. [indiscernible] [ 25 million ] order cancellation, is this booked in the order intake?

Sorry, could you repeat it?

The [ 25 million ] order cancellation. Is this booked in the order intake number of the [ 136 million ]

Yes, absolutely, yes. That is a net number, yes.

The next question comes from the line of Michael Foeth with from Vontobel.

Just 2 questions. The first is a follow-up on a question already asked on the new design wins. You said in your press release that you added several new key accounts to capture growth opportunities. And then in your answer you said that these are all existing customers. So I'm a bit confused whether they are really new or existing now. So if you could clarify that, without obviously naming the customers? And the second question is, if you could give us an update if there is any on the CEO succession process procedure.

Yes. To clarify the first question, yes, these are all existing customers. But as our customers grow, we have different types of account structures that we build around them. Our largest global accounts, we have a fully dedicated team of sales applications, even engineering teams dedicated for largest global accounts. Then we have key accounts and key accounts for us are probably in the CHF 20 million to CHF 50 million size. And we've been gradually adding customers to the key account status where they have, again, dedicated sales, but we also have dedicated applications and some engineering resources to them. We also fully customized designs for both the global accounts and the key customers. So there's really elevation of some of these growing accounts to a full key account structure. When you take, for example, EUV customer, if you go back 5 years, we almost had no business in the litho segment. So we then started doing business. We added a sales person as we got to [ 10 million to 15 million ] we had applications, people. And today, we have a full key account status for EUV customers that includes sales applications and engineering team supporting them. So there's really a structural growth around the key customers and we added 2 new key accounts in the first quarter of this year. And the CEO succession, no update on that yet. It's a progress of search and as soon as we have any update, we will announce that.

The next question comes from the line of Emrah Basic from Baader-Helvea.

Yes. The first question would be on lead times. So what -- what's the -- what are the approximate lead times at the moment in the month? And how do you expect it to develop for the rest of the year?

Yes. If you go back to, say, at the midpoint of 2022, we were seeing lead times in our core semiconductor products, elevated to 4 to 6 months. And some of that was just getting capacity in place and some was the electronic component shortages, especially that were impacting output. Now that we've got most of the supply chain challenges under control and demand is -- sorry, supply is coming back in the component area, in electronics area, I would say that's coming down by approximately 50%. So back to more historical 2 to 3 months lead times. And that's an important factor as you think about the second half of the year because the order to sales conversion time is going to shorten. So even though we see a weaker order intake in the first half, if that starts to increase going into the third quarter, we see a pretty fast conversion into revenue.

That's perfect. And the second question would be on your guidance. So you issued a 2025 guidance or you revised that a year ago, of around CHF 1.5 billion for 2025. Now does this guidance -- is this guidance still valid? Or was it indirectly also update with your last capital market guidance?

Yes. I think at this point, we wouldn't want to change it until we start seeing the recovery -- expected recovery in 2024. If you think back to '19 and what the growth rates were coming back into '20 and '21, they were pretty significant, they were around 25% to 30%. So I think if the industry gets back up above [ 100 billion ] in WFE, I think we can still get pretty close to that number by 2025. Obviously, as we go through this year and we get better visibility in 2024, we will update our guidance around that as it's a very strong correlation to WFE. It may take a year longer, it may happen in 2025. But I think it still shows directionally what to expect in the next 2 to 3 years.

The next question comes from the line of Jürgen Wagner with Stifel.

I have a follow-up to your -- one of the previous questions regarding visibility. You mentioned the recovery, you see [ next year ] as a bit of a top-down view. Taking all the data points together, how high would you consider the likelihood that this time you end up in a 2-year pause in terms of CapEx growth for you and the cancellations? Are they broad-based? Or where are they coming from in general?

Yes. As already mentioned, I think the cancellations were more around the memory sector and just the fact that OEMs were planning at a higher WFE number for 2023 and as the scale that back, we've just adjusted the capacity outlook for 2023. We can never -- especially because of macroeconomic issues that are affecting all businesses today, you can't fully predict when that capital equipment spending will come back. I mentioned what our hypothesis is. But as Fabian also elaborated, we've adjusted our cost structure so that we can perform well even at this reduced level. But I think given the growth in chips, I just can't see too long of a pause before the need for the advanced nodes, especially in advanced logic and advanced memory are going to be needed. And again, our hypothesis is really is showing additional capacity, chip capacity, not coming on stream until the second half of 2024. So that is almost a 2-year pause, but it does take about 6 to 9 months for equipment to be fully installed and qualified and chips produced. So this is quite a long pause that we're already factoring in here.

The next question comes from the line of Harald Eggeling with ZKB.

Three quick questions. First question, what would be your biggest challenge in cost management currently? Are there probably current more pronounced month-over-month swings you need to manage? Second question is the expected WFE CapEx rebound for H1 2024. Is this assumption pre or post TSMC? And what would be your implied assumption on inventory in the value chain? And third question, for Q2, I understand you correctly that cancellations are in absolute numbers likely to come down, but would it be reasonable to assume that the share of logic in cancellations would increase? These would be the 3 questions, please.

Okay. I'll do -- I'll start with the second question, look at the WFE rebound and Fabian can comment on the cost challenges. We're factoring at this point an approximate return to WFE levels around [ 90 billion to 95 billion ] in next year. That's the best estimates we have at this point. I don't think we'll get more clarity on that until really the second half of this year. And so -- I didn't quite understand your point on inventory in the value chain. Could you elaborate on that?

Yes. My understanding currently is that we have at both Tier 1 and also at the OEMs, so to say, seeing still high inventory levels. Would this not speak against a more earlier recovery in WFE?

No, I don't think so. I mean it's like a higher inventory levels than -- they are going to deplete into the second half. The potential rebound into 2024 will require additional shipment builds in early '24 and that's what we are factoring and -- but I don't think they're fully related. Are you talking about high inventory levels of finished systems or of components within the supply chain?

No, I have -- I don't have the granularity, unfortunately. But my expectation would simply be that even if you work down sequentially, like 5% every quarter, your inventory -- I would arrive at the number by end of '23, which seems you are still pretty high compared to historic levels and also taking into account clearly rising interest rates. So my sense will simply be that the recovery cannot be that big that probably currently is assumed in WFE.

Yes. Well, I think it's important. Let's see what the outlook from the OEM community is like in the coming months. I think you'll see revenue levels still at a fairly high level. We're still predicting around [ 75 billion ] of shipments in 2023. So even if you assume a quarter's worth of inventory -- component inventory in the network still means substantial shipments from VAT. And again, that has to be -- we have to refill that inventory starting at least in Q4 to allow the OEMs to build in 2024. And then the other part of cancellations. I don't expect to see cancellations in advanced logic. There's no major change in the buildout of the advanced nodes. And in the mature nodes, they're still pretty tight supply. And in fact, I think the OEMs are seeing very robust shipments into the mature logic area. So let me pass over to Fabian to talk about some of the cost challenges and the month-to-month swings .

Okay. Thanks, Mike. Look, we have started, as I said before, already last year to execute our proven -- call it down cycle protocol, it is basically a toolbox of measures that we apply as we see demand are reducing. And I wouldn't like to make kind of a segregation here between the difficulty of implementing these measures. At the end of the day, they're all painful, especially for the teams that have to drive these measures. And I think for us, the most difficult judgment that we have to apply is that we balance between strategic spend, which we will definitely continue to execute and the cost adjustment measures. So basically, it comes down to this distinction that we have to drive. But overall, I would say the company is very experienced in managing through the cycles, and we have also improved this toolbox throughout the last couple of years. Now let me maybe add another sentence on the pattern of the cost reduction, what to expect on the bottom line. I have commented on this already in March, where we come from full seen growth mode into this down cycle adjusting now. And then when you hit the brakes as hard, obviously, softer for the first, I would say, 3 to 4 months up until we see really the bottom line effect, and that is also now the bridge back to the weaker first half that we expect with the full set of measures and contributing to a boost of the bottom line into the second half.

The last question for today comes from the line of Robert Sanders with Deutsche Bank.

My first question was just around stock levels of valves, the equipment OEMs in semiconductors. I was just wondering whether you thought that the on-hand inventory today would be higher than the last cycle, just given that the impact of the shortage in semiconductor equipment. And then I have a follow-up.

Yes. The stock levels in the network. I think it's a more complicated picture than the last cycle because the relevance of the subsystem suppliers today is much greater than it was in the past. The OEM -- larger OEMs, especially have outsourced a lot of their module manufacturing to a whole host of subcontract suppliers. The inventory -- the OEMs in general, let's say, is reasonable, as I mentioned before. I think at, I'd say, historically normal level. Our own consignment inventory is under control. I think what we are starting to see is that there's maybe a bit more inventory in the subcontract module supply us than we originally imagined. So I think dimensionally, maybe something around 3 months between -- or 2 to 3 months in consignment and again stocking and maybe 4 to 5 months, maybe as high as 6 months and some of these guys are the subsystem suppliers. So that's why you're seeing such a short reduction in order intake as this whole thing is rebalanced. And some of the cancellations we've seen have been in the sub-suppliers rather than directly from the OEMs. But obviously, the relationship is there.

And what's the ratio between OEM and subsystem of your business?

I don't know exactly because there's just such a wide range of sub-suppliers. I wouldn't even hazard a guess without going in and calculating it. But it's something we'll look at for next time.

I guess the reason I'm asking is because in the automotive market, a lot of the kind of less strategic components were handled by [ Flextronics ] assembly companies, EMS companies and now those OEMs are building up their own parallel supply chains of these components? I'm just wondering whether now there's inventory at OEM level and subsystem level because of basically kind of scar tissue.

That's hard for us to see really. I think the only visibility I've got is just in general, slightly higher inventory levels at the subsystem suppliers, but I don't know much more detail than that.

Okay. My other question would just be on leading-edge logic. You sound remarkably kind of benign given that utilization is falling quite considerably at 5-nanometer, 7-nanometer, Intel's cutting CapEx, et cetera, et cetera. So what is behind that? Is it just the content gains because I know you're baking in the semi forecast of [ CHF 92 million ] for 2024, but they're not assuming really any kind of issues with TSMC, with Intel beyond what we already have.

Yes. I think the optimism is based on, again, the content and share gains we've had, but also on the build-out of 3- and 2-nanometer, you're going to see a considerable amount of CapEx going into TSMC and Samsung. I think Intel will also start picking up in '24 again, especially with some of the new product launches they have planned. So they'll build -- they haven't spent as much on these nodes as maybe TSMC have. So a little bit of catch-up there. But the spending at the very leading edge, I agree with you, 5, 7 and 10 are slowing at present, but a lot of that's going to be based on the PC market seeing historical lows. As that starts picking up with GDP growth again in 2024, I think you would expect them to see some level of growth, but the biggest growth expected at 3 and 2 nanometers.

Got it. And you're not worried about [indiscernible] Just because of inability to backfill 5 and 7 and 10 and 5.

I think 2 reuses happened in our industry really forever. It's not really been -- it's not a new thing. And customers will do everything possible to reuse our existing tool, tool set. A good example of that was the 3D NAND market, all of a sudden by bringing in an increasing number of metal layers, you've got tremendous usage out of your existing NAND factory. So if they can find ways to do it, especially with the capital cost of equipment growing at advanced nodes, they're going to do that. There's been rumors of some changes maybe to -- in the 3- and 2-nanometer areas with advanced litho systems. But again, I think it's a bit early to say is that going to cause significant impact to the EUV shipments. Again, we'll see that through 2023. But at this point, I'm not really fazed by it. Again, it's been a traditional strategy in semiconductors to reuse as much as possible. Typically, you would see nodes to nodes, probably around about 50% to 60% equipment reused, maybe even higher. It's not that unusual. Okay. So I think that concludes the questions for this call. Thank you very much for joining. As you can see, a challenging first quarter. We expect that to continue into the second quarter. And then hopefully, with the expected rebound in the first half of 2024, we start to see business improvement into the second half of the year. And our next call will be in July with the first half results, and I look forward to seeing you all there. Thank you very much.

Ladies and gentlemen, the conference is now over. Thank you for choosing Chorus Call, and thank you for participating in the conference. You may now disconnect your lines. Goodbye.