Seeking Alpha
We cover over 5K calls/quarter
Profile| Send Message|
( followers)  

Executives

Craig DeYoung – Vice President, Investor Relations

Pete Convertito – Investor Relations Manager - North America

Analysts

Unknown analysts.

ASML Holding N.V. (ASML) UBS Global Technology and Services Conference November 17, 2011 11:30 AM ET

Moderator

Craig, thanks so much for coming. You probably – most of you know Craig DeYoung, Investor Relations at ASML, obviously one of our preferred stocks in Europe at the moment. Craig, over to you. Lots to talk about obviously, but maybe if you want to do a few introductory words.

Craig DeYoung

And I'll let you catch your breath. Never run upstairs with a CK.

Moderator

Thanks, no problem.

Craig DeYoung

Yes, thank you, everybody for joining this morning. What I'll probably do, I'll just do a real quick recap of our quarter and talk a little bit about our Q4 guidance and then we can – and I'll try to anticipate a few of your general questions and then we'll open it up to Q&A. I think we have about 30 minutes. So we posted a kind of modest €514 million in the fourth quarter. We had anticipated that somewhat, as we saw as the year has gone by, a diminishing demand almost across all the sectors and our customers of course are adjusting to that and trying to figure out what next year will hold for them for sure.

So we have guided bookings to be up. I will – we said to the next level if I remember correctly. But I said it would be modestly. It’s not going to be double or anything like that. We don’t anticipate that and Q4 orders are largely going to be driven by the sector leaders taking advantage, as leaders do, of the kind of weakened demand environment. And I’ll choose this time to say clearly that our customers are not discussing any overall and general capacity planning for next year.

So I think currently they just don’t know what next year is going to produce for them in terms of demand. But we do see the sector leaders in DRAM for example going to take and make strategic investments to try to differentiate themselves even further from their competition. We’ll see that with the DRAM leaders and with the foundry leader for sure. So what again is going to drive Q4 in the main is execution on these strategies that are largely independent of the overall demand environment and can even happen and be supported by a weaker environment.

Again the leaders will normally invest in downturns or in weakness to try to exaggerate the difference between them and their competition. So that’s what we’re seeing. If we look at the NAND, which is probably a clearer sector for ASML. So you may or may not be aware. We have an ability to model the scanner demand based on bit growth demand or bit supply growth actually. So in 2012 the expectation today by industry analysts and ASML uses industry analyst numbers. We don’t – we’re not good at forecasting what demands are going to be in our customers end markets. But we use industry analyst forecasts and today we expect, based on those forecasts, about 80% bit demand growth in 2012.

We’re able then to calculate based on the technology NODE transitions that are planned by the individual customers, how much of that bit demand or the bit supply can be gained by the shrinks alone, and then if any how much wafer capacity needs to be added. So if we assume the bit growth to be 80 and given the technology NODE transitions collectively in the NAND sector amongst the four customers, we calculate the bits can be grown by about 55% given those strengths, which means that the industry needs to add some additional wafer capacity to compensate the other 25%. And we estimate today that that would be nominally about 150,000 to 170,000 wafers starts.

So and that’s against an over 200,000 wafers started in this year to meet roughly the same bit growth given the technology NODE transitions. So we’ll see a somewhat similar situation in NAND next year. So we’ve got to get our arms around that. As I mentioned earlier, bit growth and DRAMs, DRAM can be based on the technology NODE transitions planned again by the collective DRAM customers to a level of about 40% and the current industry analyst forecasted bit demand growth next year is roughly 40%.

So if that demand level holds up, that means that we’ll have a reasonable supply-demand balance in bits provided just by the shrink plans of our customers today. So again then the question that remains today for us about 2012 is as I mentioned before, the overall capacity planning of the logic guys generally. So we’re going to have to wait and see a bit. Again we have a little bit of clarity on Q4 now in terms of order intake. The question is whether we’ll support a forecasted level of Q1 shipments that is actually above Q1 shipment levels and we have to wait and see whether those bookings come through to support that level or not. So it’s a bit of wait and see, although we have an idea what the range will be. But we have very, very little visibility into the order levels of Q1 at this point in time.

So I expect that our customers will wait in general for the holiday seasons around the world to sell through out of that before they can really understand what the volume requirements are for next year. So we’re kind of in a wait-and-see mode on that front.

So yeah, I think that’s the best I can do for an introduction.

Question-and-Answer Session

Moderator

Great. Oh, we’ve got a question already in the audience. So please go ahead. I don’t know if there’s microphones around. If not just yell out, we’ll repeat it.

Unknown analyst

(Inaudible).

Craig DeYoung

No, our customers don’t talk to us about those kind of issues. Even though we are a significant supplier to the industry, we are still just an equipment supplier. So those are not fortunately or unfortunately the kind of discussions that we would have with our customers who very hard for us to get any kind of idea about how those kinds of things would affect their spend plans.

Unknown analyst

(Inaudible) talked about…

Craig DeYoung

Actually I apologize. There’s a fan in there makes it very difficult…

Unknown analyst

Is this better?

Craig DeYoung

Yeah, much better. Thanks.

Unknown analyst

All right. This may be a little too granular, but within the DRAM and NAND bit growth assumptions you mentioned, is there an assumption regarding Windows 8 consumer? And then on the NAND side, is Enterprise SSD and Ultrabooks kind of a meaningful factor as well?

Craig DeYoung

Well, I think I heard the question again, still a little bit hard. I apologize. Again we’re not that smart about the end demands and the demand drivers. There are considerations of all those moving parts within the industry analyst forecast and we have visibility and availability of all that. I don’t have it in my head. But all of that, the considerations of those things would be considered in the NAND or the DRAM bit growth numbers.

Unknown analyst

The reason I say it is I think the bit growth numbers are not that much different than 2011. That’s why I asked.

Craig DeYoung

Sorry. I don’t know what…

Unknown analyst

I’ll catch up with you later.

Craig DeYoung

Sorry. We have another one. (Inaudible).

Unknown analyst

Hi. A couple of questions. Can you hear me at all?

Craig DeYoung

Yeah.

Unknown analyst

All right. In terms of capital intensity, obviously there’s fluctuations in the industry. But in general the long term trend and capital intensity is down. Can you talk me through what the reason for that is?

Craig DeYoung

Double intensity in the industry. No, I can talk more to the litho percent of the spend, which is more important for us which is going up. So as a percent you’ve seen a significant increase in time with a bunch of variability of course in the last few years in terms of the litho percent of the wafer front-end spend. That’s what we look at. So it’s moved from 12%, 13%, 15% range to up as high as 27% of the wafer front-end spend going to litho and we expect – within the last three years it’s varied between about 21% and 27% actually. So we expect that that will continue as the ASPs of our tools increase and in the near term the throughput of our tools will actually go down. So we anticipate that there there’ll be litho intensity in time.

Unknown analyst

Okay. And just a second question. Do you break out your percentage of revenues from Intel?

Craig DeYoung

Do I break out?

Unknown analyst

Percentage of revenues from Intel.

Craig DeYoung

No, we don’t break that out. But certainly the larger – we are a significant supplier to those guys. So the large CapEx spenders are the largest of our customers, almost in order.

Unknown analyst

Just ask for myself. Maybe you’ll be able to hear me and then we’ll go to Nick. I just wanted to ask on EUV. Obviously there’s a lot of – there seems to be more confidence from your company, so post the last quarter in terms of getting to the key level of 60 wafers per hour during the course of the next year. Can you just give us a sense of where that confidence is coming from and I guess what the alternatives are? We’ve heard that it’s extremely expensive obviously to do quad-patterning and what are the timetables for your key customers in terms of locking in that process technology right out?

Craig DeYoung

Yeah, let me – those are kind of two parts. Let me answer the first part, what gives us more confidence. So as most of you probably are aware that have followed us for a while, we have reacted to the continued delays of our source suppliers in getting us the power that we in turn had promised to our customers and our reaction is to send and dispatch engineers and program managers to the store suppliers to understand firsthand what the issues are. So we did that starting about February, March timeframe of this year. So we have about six months of living and learning that’s gone on with the two suppliers specifically, although we’ve dispatched people to all three of the potential suppliers.

But the most of the focus has been on the two and so we understand firsthand what the issues are. We’ve helped the individual source suppliers to break down the tasks. We’ve convinced them that they need to resource these programs a little bit more heavily and there’s been agreement on all fronts regards the reorganization if you will of the programs and the identification of engineering tasks and engineering challenge that need to be done. And we are seeing some progress. We reported as they have. So Ushio and Cymer are the two that we’re working with today in terms of have tools delivered to the field. And they both reported now a stable 30 watts of power, woefully short of the requirements that we have to get to 60 wafer power which is 100 watts. But at least we’re seeing progress.

Those are laboratory produced results which will be rolled out into the field as upgrades to the six machines that we’ve shipped in the first part of next year. And then again we see continuous improvement beyond that. So it’s in part the fact that real progress is being made. In fact we’re living and understanding the issues firsthand. No offence to the suppliers, but they had made multiple promises that weren’t kept and after a while you become suspicious of the commitment. So it made sense to us to get involved and just due to the fact that you know what’s going on firsthand gives you a level of confidence. So we expect that we’ll continue to make progress.

We have a third supplier that is going to deliver hardware to us early next year and which is according to the schedule that they’ve always maintained. So there will be three guys in the mix and we expect that again that the – we have clarity on the route to get to the levels of power that we need and have promised, at least the 60 wafer power level by the middle of next year and it may involve the rollout into the individual customer sites late next year. So as our confidence grows our customers confidence grows. And so I think we’re okay for the time being. We’ll happily give you an update let’s say at our quarterly results in another couple of months of where we are. But we feel much more confident now, again just in large part because of the progress and the fact that we’re involved.

The adoption of EUV is going to be – is very difficult to predict in that every critical layer in every application and every customer will have its own technical slash and or economic threshold. So if we look at a couple of examples. We have one DRAM manufacturer today that is telling us that they need EUV to do a couple of critical layers, two critical layers on their 20-nanometer product and their suggestion then is they can’t do it any other way. Now if we can get to the 50, 60 wafer per hour that we’re targeting the middle of next year, it appears from our vantage point that they would be quite happy with that because that would allow an implementation in production in the 2013 timeframe.

Now, if we get – I shouldn’t speculate too much on this, but if there’s no other way to do it, whether it’s with this particular customer and this application or in another application, then maybe 30 wafers an hour is enough, or 40 or 50 and that’s to be determined by the customer. So there are going to be certain areas it appears where there may not be another way to image these layers with I’m assuming the proper yield et cetera, et cetera. So then it becomes a technical question and the option may be a delay of the product rollout for that customer or they buy it at whatever available throughput there is.

But on the other hand, you have the NAND manufacturers that today use something called Nitride Spacer Double Patterning. They’ve used it for at least two or three generations of product. It’s well known, well understood and used across all the four NAND manufacturers and it’s extendible. It’s extendible to at least the team’s node. So we have a few years yet where they can employ this Nitride Spacer Double Patterning technique. The problem with it is that it’s expensive because it involves many, many additional process steps, additional process equipment, larger longer cycle times for those products, et cetera, et cetera.

Everything that our customers would like to avoid, it delivers for them. But we’ve been told that in the NAND Spacer process that kind of a break even or a cost parity level with EUV at €65 million each is at about 50 or 60 wafers per hour, again that kind of magic number. So that may not be – that’s the cost parity if you will, at least in a couple of cases based on what a couple of customers have told us. So we need a level above that to be highly motivating for the customers. So we would anticipate that if we could get to the 100 wafer per hour which is ultimately the next target for us that you would see probably what I will for the sake of this discussion called broader based adoption in the NAND.

So the NAND driven by more of an economic consideration because there is a solution for at least the next few nodes in NAND. The DRAM is more of a technology, at least in this particular case technology driven. We may not be able to do it any other way. And then you have the larger guys that are kind of in between today. The foundry players I think by their own admission can do a 28-nanometer image, 28-nanometer line. But in many cases it appears that they’re having to do some design constraints on their customers. So they may be saying yes, I can print this particular image size for you, but I need some space around it which likely requires a compromise on the designer’s part, which they would rather avoid.

So it’s doable today with certain design constraints which nobody likes, but can live with for the time being. So the EUV would give the larger guys again the ability to do this single layer, no constraints and that’s what everybody would prefer. So I hope that gives you a sense of all the moving parts in each of the different sectors that make adoption of EUV a case by case, customer by customer, critical layer by critical layer decision. It’s very hard to anticipate that from us as equipment supplier.

Unknown analyst

Yeah. Hi, two questions, Craig. One is, there’s been a fair amount of talks about fairly significant CapEx increase from one of your logic dash foundry customers in terms of an SI next year. I wonder if you could talk about the implications of that for ASML. And secondly, 20-nanometer TSMC will move to double patterning immersion thus rolling out a lot of 28-nanometer capacity still as you know. Is that 28-nanometer rollout or the 20-nanometer capable or are we still to seeing the logic foundry space wave of upgrades driving NXT business to move to the multiple patterning for 20-nanometer?

Craig DeYoung

Certainly anybody that has wafer capacity, whether it be a bigger or small company, we’re happy to respond to that requirement. So if when the big Korean company is going to add significant capacity in any of the spaces in the logic space particular to your question, yeah, that will have a significant impact on us as we are the major supplier of equipment to that customer. So if your question was – if I heard it correctly, it was as general as that. So certainly I mean that would be advantageous for us. If I heard your question correctly, relative to the 28-nanometer in foundries and multiple patterning, to be quite honest with you, I’m not sure all the different techniques that are used at all the different customers as they go from one node to the next.

Certainly depending on the design layouts you could potentially do a 28-nanometer feature with a single patterning technique and overexposure. If you’re talking about – so the requirement to go to double patterning is based on the design layout more than anything else and that I’m not privy to. But we were in the early stages of 28-nanometer capacity addition. So we are very limited. The industry has very limited amount of 28-nanometer installed today and it appears from some of the foundry’s statements that they’re going to accelerate that at a rate that’s I believe I read somewhere 3x what they did at the 40-nanometer node.

So that’s based on real demand I would expect. So it’s the capacity additions that are driving our business now at 28 and will continue to. So 20, the next node is a ways away. It’s probably a couple of years away in meaningful capacity additions. So currently it’s about capacity additions at 28-nanometer and the 20-nanometer, 22-nanometer is in development now and probably again 24 months away from real production ramp.

Moderator

Time for one last question before we go to breakout. Maybe I’ll ask it. Just on EUV, Craig, I wonder what the clean room requirements are and whether we’ll see actually a spike in CapEx just to enable the EUV, even though it may be one or two layers to actually come through.

Craig DeYoung

Yeah. I can’t translate it into a CapEx number, but these things are significant in their size, in their facilities requirement. So, all of our customers are aware at this point in time. We’ve shipped six of these tools. Five to the biggest IC manufacturers in the world in the different sectors and we have 10 of these tools on order. So, preparations in many cases are already being made. Consideration has to be given to the sheer size of these things. Although the 3300 which will ship next year has a reduced footprint over the 3100s that we’re shipping this year, they’re still fairly significant in size. So yeah, I can’t comment as to what the real investment requirements are in and around the installation of these things. But it’s significant relative to the current NXT emerging tools that are being installed.

Moderator

Great. Thanks. There’s a breakout downstairs in the Julie Arts rooms. So for anyone who’s got further questions please do come down. Thanks very much guys.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!

Source: ASML Holding's Management Present at the UBS Global Technology and Services Conference (Transcript)
This Transcript
All Transcripts