Cymer Q2 1999 Conference Call Transcript, Part 2 - Q&A
Robert Marie, Bear Sterns: Hi, congratulations on a continued recovery. Couple of questions: could you give us your best estimate as to the percentage of different types of steppers that are currently being shipped, I-line and the different wavelengths of laser source? Also, listening to the numbers it sounded like the Nikon numbers were down substantially – I would assume those would pick up again in the future, and correct me if I'm wrong, but I would also assume that that's due to previous higher inventories that may have been worked down. And third question, in terms of the upgrade kits that you had made available to customers, is that, is most of that upgrade activity complete and or do we still have a substantial amount of that upgrade activity to go?
Angus: Let me take the last two questions first, you had quite a laundry list there. But, just to refresh everyone's memory, in the first quarter, Nikon percentage of our revenue was 13%, and now on a year-to-date basis it's increased to 16%. So we actually did more business with Nikon in Q2 than we did in Q1. One the subject of system upgreade kits, we are just now starting to really ship those, and we're going to be shipping more, and we'll have more business of that in Q3 than we did in Q2. And I think Pacal may be able to shed some light on what we're seeing as far as the industry mix.
Pascal Didier: On the industry mix, deep-UV is still growing at a much higher rate than I-line. I don't have an exact percdntage on a per-regional basis of what that means, but deep-UV has been growing very strong, mainly because of the adoptions of the new scanner generations over the last two quarters. The volumes where many [unintelligible. Sounds like ‘char are drivens'] ArFl scanners have been shipped over the last 3 months, but mainly for process development purposes and not for any kind of mass production. We expect continue to see ArFl shipments at very low volumes as people are not only going from process development but also to precharacterizations of volume manufacturing and process integration. And also on the 5005, maybe to clarify a little bit more, what usually happens is that we book and ship on those upgrade kits in the same quarter. And installations are taking place right away. So we have also accumulated a fair amount of product performance [data] with our 5005, which are well in-line or even in excess of the performance that we sell them for.
Marie: OK, and one last question: given the current recovery of the industry, is it your estimation that there's been an acceleration of the conversion to shorter wavelength sources over the last couple of quarters?
Didier: No, I don't think so … well, depending what you look at it, from an I-line to a KRF , I would agree to that. I think there is still excess I-line capacity at the chipmakers when you look to their transitions to .20 and .18, so you have an acceleration of KRF. When it comes down to ArFl, I think we're pretty much in-line with the expectations of a limited number of tools, many as I said before driven for process development and process configurations and characterizations of some of the new technologies like copper and so on.
Mark Fitzgerald, Merrill Lynch: On your guidance for revenue for the fourth quarter, are you basically going to ship all your backlog, or is there a certain percentage of your business that's turned into turns business here?
Angus: Mark, basically all our spares business and much of our upgrade business, as Pascal just alluded, is definitely a turns business.
Fitzgerald: And you're including that in backlog?:
Angus: To the extent that we have some orders, yes.
Fitzgerald: And on the tax rate here, going out, uh next year what tax rate are you looking at, is it …?
Angus: Yeah, uh given what I know today, my best estimate would be a 38% tax rate for the year 2000.
Fitzgerald: And can you give a sense of what the cycle times are for, uh, actually the lead times for your lasers at this point?
Angus: It depends on what laser you're ordering. I know there's a minimum 4-month lead time on the 6000 series right now. And the … Pacal, what's our understanding on the booking there as far as the KRF is concerned?
Didier: Well usually we like to quote a minimum of 4 months lead time for all of our integrators, up to 6 months for the 6000. The main reason is because we have now worked relationships with our direct customers where we're not only managing their shipment, but we're helping them and us managing their inventory of product mix. So we like to get enough visibility from a backlog standpoint, especially moving forward as the business continues to strengthen.
Fitzgerald: So you've started to get visibility on the fourth quarter, I assume?
Didier: We're starting to get some visibility on the fourth quarter, and in fact as Bob mentioned in his comments, we're actually working with our customers to define their product mix for the fourth quarter.
Fitzgerald: And can you say anything in terms of does it look like there's another strong quarter here, or …?
Angus: Yeah, uh we definitely believe that sequential growth, that uh will be good, going into Q4.
Fitzgerald: And on Nikon, can you make any comments how much of their business is going to Komatsu? Are we still looking at 50% of the business?
Didier: Right now, our best estimation is that the total number of Komatsu shipped to date is 26 units. Versus our installed base and shipment, uh, for example our shipments just this quarter alone of 94 units. So, the estimated 50% market share in Japan has not materialized to our estimation to date.
Fitzgerald: And just one last question, it's kind of a general question here, but when you look at what you guys gotta do to build out the infrastructure here, is the market large enough for one of these second players in the market to come in and kind of duplicate the global infrastructure and be able to make money at it?
Akins: Mark, that's the question that we have of course been asking ourselves for many years,and I think that it's a good assumption to say that the industry demand for a second source is going to make do with whatever distribution and/or serviceing infrastructure can be put in place for that second – or third – source of lasers. We have been attempting, I think successfully so, to differentiate ourselves as a value-adding company that ties in not only to our technology development and manufacturing capability, but also to our global service infrastructure as well. And I will say that there's a growing trend that shows that when Cymer provides direct support for the laser, the overall uptime and chipmaker satisfaction are significantly higher. So we think that we've got certainly the best way to conduct the business, and it's now up to the competition to try to duplicate that.
Bret Hodess, Nations Bank Montgomery Securities: Given the strong order recovery and the shift back to a much more unit-driven business uh volume pickup that you'll see in the next couple of quarters, can you talk a little bit about what's going to happen with the gross margins going forward. With stable pricing you'd think that you'd get a lot of fixed-cost absorption.
Angus: You're right Bret, we should start tp see margins improving in Q3 over Q2, and then any incremental improvement in Q4 will be basically a function of a number of things, including our comfort with reducing additional warranty reserves we always take when we ship new product, specifically here thinking of the 6000, as well as the ultimate product mix that we finally achieve in Q4, depending on how our plans work out with the customers. But we're going to be broaching the 40% margins here hopefully by the end of the year/early 2000.
Hodess: Next part of the question I had is can you give us a feel – you talked about I-line versus deep-UV mix and some of the other things – can you give us a feel of how much of the 5010s are starting to go into scanners as opposed to steppers, and does it have any impact what the end-market is if it's a step-and-scan system versus a stepper system in terms of your margins?
Didier: First, the first part of the question I would say more than 65% of our 5010 shipments are going into scanners versus steppers. And when it comes down to the 1 kilohertz equipment, no, there is no impact if it's a stepper-design laser or scanner-design laser on the gross margin at all. Fpr us, it is the same product. It's mainly how people are using the software which is embedded into our lasers that make them suited for scanners or steppers.
Leonard Sanders, from [unintelligible] Company: Could you go over your operating expense outlook? You said in your prepared statements that some of your sales and marketing expenses were below planned levels, and could you go through the expectations there, and whether … what you expect to see going forward on your plans? And some of the limitations that you've seen there?
Angus: Yeah. Uh, I don't want to get into a detailed model discussion right now, Lenny. But we are going to see some incremental increase in the sales and marketing area. Ithink I gave guidance as far as R&D. R&D is going to go up on absolute terms, but we will finish the year in about a 16% of total revenue point. In the area of sales and marketing, I think you're definitely going to see a bit of a step-up in Q3 and perhaps a little bit more in Q4. And then on the G&A side, I think there's just going to be only small incremental growth going forward in the G&A costs.
Sanders: What have been the issues in the sales and marketing that have prevented you from ramping as rapidly as quickly as you thought you might?
Angus: Part of it was actual savings that we were trying to get all along, but you couldn't be sure of and you couldn't put in the bank. And it has to go with certain expenses that we may have had to spend on certain projects that we were trying to do in that area. And, frankly, we were able to save the money. That was good. In the area of hiring, I'll let Pascal talk to that directly himself.
Didier: On the hiring side, we've been pretty selective, just because of the business model that we have, and kind of the triangular relationships that we have, between our direct customers and the chipmakers. We are looking for very specific type of people.; The other side is, as mentioned, we have made some progress on our business processes, and the sales and marketing organizations being pretty young at Cymer, we are able to get those business processes in line as we go along, and realizing some benefits out of that from an expense standpoint. And it had no significant impact whatsoever on our ability to support the customers direct, or the chipmakers. As we said, our 5005 is on-line and our orders are growing.
Sanders: And could you just give us the number of employees then, please?
Angus: Yeah. Our employees at June 30 totaled I believe 724. And I'm trying to confirm that …. Uh 727.
Sanders: And the other question I had is pricing in your backlog versus pricing in the quarter, and what trends you're seeing there.
Angus: Well, obviously the pricing on the 6000s is higher than the 5010s and the 5000s, so the ASP and the backlog is headed North. We are still in the middle of negotiations with some customers, so exactly where that's all going to settle out is yet to be determined. But the trend is still upward.
Sanders: And it's upward due to mix versus anything else?
Angus: That's right.
Hi Min Peng [phonetic], SG Cowan: Hi, this is Tim for Min. Couple of things here. First of all, I'm not sure if I heard you properly, but your guidance for Q3 when you said that revenues would be up between 28 and 30 %, but that shipments would be up between 60 and 65% …
Angus: Unit shipments of new systems, that's correct. And remember when you look at our overall revenue, it has two major components: one is new system sales, the other is the sales of spares and service and upgrade kits. So I'm talking units, I'm only talking new systems.
Tim: Secondly then, why were unit shipments down this quarter over last quarter? Is there an overall industry issue there going on or … ?
Didier: No, the main reason is really in two stages. One is, we had a final inventory correction that needed to take place, to bring us into the 250 lasers inventory level, which is in line with the goal of inventory which is projected moving forward. The other thing is a strong shift to our 6000 product line, and we're starting to see the beginning of that shift. And what we want to be careful is working with each of our customers and making sure that the inventory level on 1 kilohertz powered machine versus our 2 kilohertz will be in line with their future goal. So it was mainly working directly with our customers, understanding what their 1 kilohertz needs, what their projected 2 kilohertz needs, were going to be, and then making decision on unit shipments in this quarter and moving forward.
Tim: One more thing: do you expect , given that unit shipments will be up between 60 and 65%, do you expect ASPs to be about flat?
Angus: Yeah, that's what we said. Q3.
David Wu, AB and Amerault : Two things. Number one, is the uh, as you move forward with higher shipment rates, gross margin will be improving. I remember last October when we talked about improving the operations of the company, I was wondering whether gross margins are improving just from sheer overhead reduction, or is there something about your lean or new method of manufacturing the products that would give you a boost in the gross margin? And lastly, how much does the mix of 6000 do to your overall gross margin versus 5010?
Angus: To say than any of the improvement in margin is due to any of the process improvement at this time is too early. As I have said before, we really expect the effect of the whole business process development program for the most part, aside from Pascal has already alluded to, to basically be to have a significant effect on the company starting the second half of next year. These things take time to fully mature. That doesn't mean that you won't get spots of encouragement and saving here and there, but nothing significant to drive major trends between now and then. In the area of the 6000, we're still as I said in price negotiations with a number of customers, so until that settles down I'm a little concerned about drawing any conclusions vis-à-vis the total margin improvement based just on the 6000. But we obviously are expecting that as they kick in,and as we are able to reduce any added warranty reserves that we always take with the shipment of a new product. As has been our practice with past product introductions. I thnk there's more to follow on that, and we'll be able to update you better at the end of Q3.
Wu: On a cash flow basis, I guess if you are accelerating your growth rate, I assume that you'll be … you interest expense, net interest expense, will be rising, right?
Angus: Yes, assuming that we're going to be starting to burn up some more cash as receivables and inventories expand, although however, particularly in the area of inventory, I don't want to get into making a prediction, but I'm hoping to keep that expansion to a minimum here. Obviously receivables will expand to some extent as the revenues increase, that's just very natural for that to happen. And there will be some increased capital spending as well, particularly towards the end of this year and into next year. So, your assumption that the interest expense will become more in excess of interest income is a fair statement.
Wu: Do you have any ideas, preliminary ideas, about capital spending depreciation for the year 2000?
Angus: No, not yet/ We're just beginning our 2000 planning process. So anything that I have now is very premature.
Akins: Let me add to a postscript to your questions here. You talked about the process, and we didn't speak specifically to that in our prepared comments. But as you are aware, our Senior Vice President of Process, Ted Holtaway, has been spearheading an effort in the company to put in placed process in many sectors of the company. And in fact we have a number of teams all running simultaneously, that have been evaluating Cymer's processes, benchmarking in the second quarter against best-of-breed within our industry and outside of our industry. They've done the conceptual resdeisng of processes which have been presented to and approved by myself and the exec committee, and work is now underway to flesh out those basic redesigns, with much more detail. Ted Holtaway expects that we'll be able to actually start profiting from these processes in a significant way later this year. And as Bill mentioned, we'll be looking to see the benefit of those, especially in the second half of the year 2000.
Mark Fitzgerald, Merrill Lynch: Do you guys still agree: or disagree with the Dataquest forecast for 400 deep-UV systems this year?
Didier: We are in the process to review that with Dataquest. I think that we are very close to there, to their set of numbers. So I would say that for this year that we have reached an alignment there.
Fitzgerald: And are we talking about 418 systems? That where they're at?
Didier: Yeah, I think that there are two revisions, one at 418 and one at 405 in the last discussion I had with them.
Fitzgerald: OK, and Bob can you just give us the CapEx and depreciation in the quarter?
Angus: This is Bill. The capital expenditures for the quarter were 3.7 million, and the depreciation was 4.2 million.
Steven Quail, Morgan Stanley Dean Witter: I was wondering if you guys could walk me through the 250 lasers and [what sounds like ‘whip'] out there – the assumptions on market size, the adoption rate, and the type of lasers demanded – what's the math behind that? How do you recognize that as being rationalized at 250?
Didier: The way we're looking at it Steve, is looking at what we know about the manufacturing capacity of each of our direct customers. Having ongoing discussions with them about the status of their order rates and backlog and some of their manufacturing efficiencies that they have put in place during the downturn, and continue to put in place today. Remember also that there was an increase of capacity that was downed by Nikon prior to the downturn that now Nikon is turning on-line. [Unintelligible] has great efforts in manufacturing capacity and are also turning it on-line. So, looking at all those different parameters which is basically their cycle, their cycle time, the number of testbeds that they have, their product between steppers and scanners, we all came up with a different set of numbers for each one of those customers and it added up to about 220 to 255 to a rational inventory rate. Now most of it, you have to remember, is now towards the 5010, and the 1 kilohertz KRF machines. There is no inventory per se in the ArFl world.
Quail: Are there any concerns with lead times for the 6000s of 4 to 6 months over double booking?
Didier: I had some concern on that to be honest with you, and that's why one of the key things that we've done after the introduction of the 6000 and the qualification process that we went through, was to set in place at least with already two of our customers, a different [or possibly ‘definite'] way of planning production needs for their 6000 requirements. Now, you have to remember, that 6 months is a pretty short lead-time from our standpoint when you look at their integration time on their brand new generation scanners. So I think that while there is substantially a risk there, we believe that we've put in place as many things as possible between the different companies to limit that exposure.
Akins: I'll also add, Steve, that Cymer's a very different company when it comes to the number of people that we have in the field, especially working with chipmakers now, than we were a couple of years ago. So now we're in a much better position as well to rationalize new orders from direct customers with our real understanding of perceived need of the chipmaker for those units as well, and deal with all those numbers appropriately.
Angus: Let me just add, I think it's only appropriate to assume, you know, the lessons of 97 are not lost on Cymer. We are very very sensitive to the potential for our customers to double-order or over-order. And we will work diligently to try and prevent that, but I think it will be impossible for us to prevent it. IT will happen to some extent. We just try and have to wrk to minimize it as much as possible. That's about the best we're going to be able to do.
Quail: Bob, last quarter you gave a number that you said, I think over 90% of the lasers shipped in the first quarter of 19999 were Cymer. I wonder if you could give any more color on the competitive landscape. You had said the second and third tier chipmakers moving to deep-UV lithography maybe the competition would have some inroads there. Any more color you can provide on the competitive landscape?
Akins: Getting away from speculating as to the business that may or may not go on between our competition and our direct customers for units for evaluation and so on so forth, the only thing that seems to make sense is to look at the number of units that have shipped by the lithography community to the chipmakers bearing the competitors light sources. And as we discussed at the end of the first quarter, there were with Komatsu there were 20 systems shipped to chipmakers as of the end of the first quarter, and that had increased by about 5 units this quarter for a total of roughly 25 systems now at chipmakers by Komatsu. In the Lambda case, we believe that number is something like 1 or 2 systems in the field total, So you can run your own numbers and your percentages the way you'd like to. Those are the basic numbers.
Quail: And just so .. get me straight here, what's your installed base now?
Akins: Our total installed base is 737 at chipmakers.
Tim, follow-up: Can you kind of walk me through this again. The spares revenue next quarter goes down, but the system revenue goes up. Is that this quarter uh Q2 is that kind of a one-time thing or should we be reading something industry-wide into that?
Angus: No, that's why I gave the guidance I tried to earlier, that the spares and service revenue here in the shorter term should average 23 to 26%, and then on the longer term, 15 to 20. Now that's our best-educated estimate, Tim, on how we feel it's going to play out. Historically at Cymer, it's been very much up and down. It's been very uneven demand. A lot of things come into play. Nuber one it's a new technology, the chipmakers are just getting used to it. The distribution channel for the spares has switched from through the integrators to us and back,. And additionally you have the whole point of the fab utilization is growing in here as well. So, you know, you've got a lot of elements all turning on and impacting here, in the short term particularly. Utilization rates is one of the biggest ones right now. |
