ASML Holding N.V.
Q4 2008 Earnings Call Transcript

Published: 15 Jan 2009

[Operator:]

Thank you for standing by and welcome to the ASML 2008 fourth quarter and annual results conference call on January 15, 2009. Throughout today's introduction, all participants will be in a listen-only mode. After ASML's introduction, there will be an opportunity to ask questions. (Operator instructions) I would like to turn the call over now to Mr. Craig DeYoung. Go ahead please, sir.

[Craig DeYoung:]

Thank you, operator, and good afternoon and good morning, ladies and gentlemen. This is Craig DeYoung, Vice President of Investor Relations and Corporate Communications at ASML. I would like to welcome you to our investor call and web cast. As the operator mentioned the subject of today's call is ASML's 2008 fourth quarter and annual results. Hosting the call today are Eric Meurice, ASML's CEO; and Peter Wennink, ASML's CFO. At this time I would like to draw your attention to two items. The first one starts with an apology. We discovered that the call-in number in the press release for Dutch participants or European participants is incorrect in the press release but is correct on the web site and in the e-mail invitation. So let me give that correct number to you now. It is [phone] and our sincere apologies on that. I would also like to draw your attention then to the safe harbor statement contained in today's press release and to the fourth quarter results presentation both of which you can find on our website at www.asml.com. Of course the safe harbor statement covers the discussion we will have today and we may and likely will be referring to both the press release and the fourth quarter presentations, again which you can find on the website. The length of the call will be 60 minutes. And now, I would like to turn the call over to Eric Meurice for a brief introduction.

[Eric Meurice:]

Okay. Thank you, Craig. Good afternoon, good morning. Thank you for attending our 2008 annual and Q4 results conference call. As usual Peter and I would like to provide an overview and some commentary on the quarter before we open up the call for questions. Peter will start with a review of our Q4 financial performance with comments on our short-term outlook and we will complete the introduction with some details on our near term strategy and overall review. So following this introduction, of course, we will open to questions. So Peter please.

[Peter Wennink:]

Thank you, Eric. First of all, I would like to review some of the events in the last quarter as well as some details on our fourth quarter results. And as a reminder in our December 18th press announcement we communicated that he saw a severe deterioration in order intake due to the global economic crisis. We experienced a sudden drop off in lithography demand triggered by an unprecedented mix of falling end demand for semiconductors, weak memory device prices, and very limited access to capital for our customers. The associated steep decline of our business outlook forced us to take immediate action to lower our cost base. We took a set of measures to lower our cost base, which included a work force reduction of 12% worldwide, a partial relocation of customers support activities from the United States to Asia, a scheduled shutdown of production facilities for four weeks in 2009, reduced labor hours, and a dramatic reduction in various discretionary spending. The combination of these efforts will reduce ASML’s expenses roughly 50 million euros per quarter starting in the first quarter of 2009. The costs associated with these measures and the impairment charges was 137 million euros, which we took in the fourth quarter. Virtually all of these items were non-cash by the way. Now on our fourth quarter results, our fourth quarter sales fell just below 500 million. We shipped 25 systems, 15 of which were new and ten were used. Amongst the new systems 11 were immersion systems keeping the new system ASP above 20 million euros with an ASP of all systems being 15 million euros. This is obviously the result of a large portion of used system sales. Our (inaudible) revenues were higher than anticipated due to an unexpectedly large number of lithography system moves, and due to the restructuring charges we reported a net loss in the fourth quarter of 88 million euros. Excluding the net effect of these charges, we would have booked net income of 11 million euros. That is 2.3% of net sales. Full year 2008 sales were close to 3 billion. That is a 22% decline as compared to 2007. Net profit for the full-year including the fourth quarter restructuring charge amounted to 322 million euros or 11% of net sales. Now as I said before orders in Q4 were understandably low with net bookings at only 13 systems with a bookings value of 27 million. The orders included only five new systems of which three were immersion with an average selling price of close to 25 million euros and 8 used systems with an average selling price of 400,000 euros. This leaves our order backlog as of December 31st at 755 million euros with a total of 41 systems, 19 of which were immersion. Due to the impact of the fourth quarter restructuring charges the company’s gross margin was 8%. And excluding restructuring and impairment charges gross margin would have been 35.5% and this is well within our guidance range when taking into account the lower net sales in the quarter. The fourth quarter R&D costs were 127 million euros and SG&A were at 47. Both items included by the way, each a 1 million euro restructuring charge. In the fourth quarter the sudden and steep business decline obviously left its temporary mark on our cash position. We deliberately made payments to support our suppliers that provided us with inventory that ultimately pushed out by our customers. Customer shipments in the quarter were back-end loaded creating unpaid accounts receivable at the year end. And a final tax assess – and a final tax settlement for 2007 and investments toward the completion of our EUV factory, which is due to be completed mid this year created additional cash outflow in the quarter. Now clearly such cash outflow will not repeat itself in the quarters to come. Positive cash flow from operations in the first half of 2009 will be fueled by a collection of our accounts receivable balance and the use of our inventory to support virtually all of our systems and service sales. Reduced capex in the second half of 2009 due to the completion of the EUV factory and the cash benefits from lower cost of operations will further help to control our cash position. As a result of the above, we expect to stay within the bandwidth of our gross cash target of 1 billion euros to 1.5 billion euros. This would by the way include an expected dividend payment of 86 million euros in Q2. As for our outlook, the severity of the global economic downturn has caused most semiconductor manufacturers to delay their investment plans. The uncertainty as to the timing of the pickup of semiconductor demand and the time needed to complete the current consolidation and restructuring activities in – especially the memory sector makes the prediction of the pickup of the semiconductor equipment market impossible. In this environment of an investment freeze it is understandable that the wait and see attitude of our customers has a significant impact on our system shipments. We therefore expect Q1 2009 net sales to be in the range of 180 million euros to 200 million euros. And since our cost base to support our normalized sales levels amounts to approximately 120 million euros the gross margin at this very low level of Q1 sales will be about 8%. Net R&D expenses are expected to be at 170 million while SG&A are expected to be at 44 million euros. As a closing remark I would like to inform you that based upon our confidence in our ability to keep generating cash ASML also made a proposal to the 2009 general meeting of shareholders to declare a dividend of 0.20 per share which is approximately 86 million euros which compares to 0.25 euros per share made in 2008 on the profit of 2007. And now I would like to turn it back over to Eric for more on our strategy for this downturn. Eric?

[Eric Meurice:]

Thank you, Peter. In other terms Peter confirmed the following. First, we executed our fundamentals in Q4 in terms of price margin and costs. Second, we recognize that the market will be at a standstill for a period of time, justifying restructuring, impairment and the significant leverage of our flexible cost model that is reduction of flexible workers and contracted activities for labor et cetera. Third, we do not have at this point any data allowing us to call for the timing of the recovery. The semiconductor unit end demand has dropped significantly in Q4. I think it is an historical drop by the way. And the customers have slowed down or delayed their technology transitions either by leveraging their existing installed base or just by delaying their new product introduction for later. We however see a certain number of encouraging factors for the future. For instance, we estimate that the DRAM manufacturers and the flash manufacturers have now reduced their wafer out refill [ph] capacity by respectively 30% for the DRAM and 20% for the flash manufacturers versus an average of 2008. So that is I would say January numbers or Q1 numbers. By a mixture of significant 200 millimeter capacity retirement and the transfer of some 300 millimeter capacity to new technologies which in turn translate into less wafers per installed base because these new technologies are less efficient in terms of number of wafers although they are more efficient in terms of numbers of dyes. This significant reduction of capacity combined with a lower utilization rate is slowly translating into a DRAM and flash chip market price increase as we have seen in the past two to three weeks. We also estimate that the current low booking levels is clearly, clearly not sustainable in case of a semiconductor demand pick up from the current level. In addition we are encouraged by the progress made towards the restructuring of the overall DRAM industry led by the Taiwanese Government towards the creation of a sustainable consortium. During this, what we could call transition, we’re focusing our effort on efficiency and on continuing to invest in technology without any impact on our strategic programs namely the XT4 [ph] cost efficient platform, the NXT Double Patterning platform and the EUV next generation technology. In detail, we will indeed transition our customers to the cost effective XT4 platform within the year 2009. Our TWINSCAN and XT multipurpose platform is being finalized for introduction and delivery in the first half. With substantially higher speed up to 200 wafer per hour and overlay accuracy that any – that are better than any current ASML competitive product. It will provide customers with the capability to image to 32 nanometer and potentially (inaudible). Our EUV production system roadmap supports cost effective chip manufacturing to 22 nanometer and smaller and deliveries will start in 2010 as previously planned. In addition, ASML has developed a range of resolution enhancement techniques, which leading IC manufacturers are implementing to reduce cost per chip. These techniques from advanced illumination shaping to radical mask pattern enhancement require powerful computation of modeling and intimate knowledge of the scanner systems. This is an unique combination in which ASML has established a leading role thanks to the know-how acquired through Brion, its division that we acquired two years ago. We describe this concept as holistic lithography and it will now enable continued aggressive lithographic devices scaling while optimizing our customers cost of ownership. In parallel to our focus on R&D as I said at the beginning we have accelerated our operational efficiency program in order to be able to weather the current crisis and at the same time become structurally stronger company. By this current quarter by Q1, we will have as Peter mentioned already cut operating expenses by about 50 million euros per quarter. Most importantly we expect that about 50% of this will be sustainable when the economy recovers making ASML much leaner and in turn resulting in greater profitability during the market recovery. In summary, we are confronted with the demand freeze in the short-term during which we will not burn cash, we will continue in our technology investment which will not be impaired, and we will be able to work and focus our attention; our management attention, our people attention on creating a more efficient cost structure for the future. So, in other terms, we will leverage the period of low activity during this recession. We cannot however, yet call the timing of the demand pickup obviously which will naturally be semiconductor end-demand driven. That is in numbers of units and the application or the new application pickup, which are the big users of new technology type chips. So with this, Peter and I would like now to take your questions, if you please.

[Craig DeYoung:]

Ladies and gentlemen, the operator will instruct you momentarily on the protocol for the Q&A session but beforehand, I would like to ask as I always do that you kindly limit yourself to one question with one short follow-up if necessary. This will allow us to get as many callers as possible in. Now operator, if we could have your instructions and the first question please.

[Operator:]

Yes, thank you sir. (Operator instructions) Today’s first question is from Mr. Nicolas Gaudois. Please state your company name followed by your question sir.

[Nicolas Gaudois:]

Yes, hi. Nicolas Gaudois, UBS. Just first question on the immersion units for this year, I mean, 3 months you were saying you will be up from I guess the 55 units, usually it is for new tools in ’08. Obviously which is not how things are looking and also, maybe if you could help us trying to find a base for immersion units for this year and by implication of how would you quantify the pent-up demand towards the difference versus maybe 3, 4 months ago and now which is represented by companies just coming to a halt in terms of a capacity to buy tools offerings [ph], because clearly now you have got to be a moving part to factor in into the next year hopefully when things normalize and this pent-up demand comes back. Thank you.

[Eric Meurice:]

Nicolas, thank you. In fact, it is a very insightful question particularly on the pent-up demand because that is the real question. So, first is at this moment we have clearly no visibility. We made it – we wrote it in the press release. We must say that we have been surprised. We used to say that we are protected by the fact that the age of significant technology transition business are there and therefore we will have a minimum of x numbers of units and this is why we called the number of immersion units for last call. What happened, which was a bit of a puzzle to us is that the recession was so steep in Q4 that even those technology transitions did not happen. They did not happen for certain numbers of reasons. The first is that the application drive that is the numbers of new products, new telephone, new MP3 players, new hard disk, new et cetera basically this is the application drive disappeared as much as the rest of the demand. So the drop has been significant, which means the customers were starting to ramp their 45 nano or the 35 nano found themselves with a no demand type of activity. So you do have an application impact. The second impact was or is the issue of financing certain numbers of tools, in fact significant number of tools where it is supposed to be purchased by people who are at this moment in restructuring phase or looking for financing or just waiting in fact for the consolidation that we are talking about of the DRAM industry, which is a good thing, but it takes a bit of time to negotiate. As you know the Taiwanese Government and we can talk about this later but this – all the companies involved in this restructuring are in the wait and see attitude just because there is no so much of financing but who is going to buy or what it is, what is the new consortium et cetera. The third aspect, which has completely changed the cards is that in view of lack of demand and potentially lack of financing and difficulty to finance or to justify profitability of all that our customers played smart or are playing smart. That is they are trying to build up new technology nodes by mix and match of what they already have. So in other terms, if a customer had one immersion unit and 5 ArF dry units, they are playing with a bit of Double Pertaining dry and less immersion layers on the existing installed base and they can generate a mid-node or node, which are a bit less aggressive or a node with less yield or et cetera, et cetera. But basically they are making use of what they have and that was rendered possible or that was rendered necessary because they didn’t have cash and that was rendered possible because the demand has gone down. So they don’t need the machines at question that become idle. So that puts a bit of a twist to the short-term environment which explains why we come up with a very low guidance for Q1 of 180 and 200. But definitely which is to your question of pent-up demand, this 200 million is – run rate is absolutely unsustainable. These are smart ways of working these solutions is not solving the mid-term and the application curve is going to start hitting the customers. In other terms, yes the demand is low but slowly the demand is converting to the newer products which have to have more integration of 35 nano flash cards or 40 or 55 nano DRAM et cetera, et cetera or DVR 3 et cetera. So, you are – even if the demand volume doesn’t pick up the mix is going to require capacity. And Nick, I would like – Nicolas I would like at this point to leave it there and stay qualitative. At this moment, we are relooking at this pent-up demand number to see when those numbers, 55 or 60 will reoccur and obviously as a bit of a mixture between total market units and application mix and I don’t think we are at this moment in any position to give you, I would say, a simulated satisfactory answer.

[Nicolas Gaudois:]

Okay, great. But just a quick follow up on the (inaudible). Maybe if we just say we have Taiwan’s arm [ph] consolidation came, you know, if something was to happen in refinancing and leads with two many players and their respective JVs we might (inaudible). What kind of number of units were mentioned could be added, what kind of franchise. I mean, I understand some of that depends on what exactly process will be and therefore a number of units you need, but just to give a range of what is basically locked up here because of these negotiations ongoing?

[Eric Meurice:]

Yes, I can guess the number of about 15 to 20 for 6 months. That would be the base critical mass, I will call it critical mass restart of this environment.

[Nicolas Gaudois:]

Great. Perfect, thank you very much.

[Eric Meurice:]

Yes.

[Operator:]

Next question is from Mr. Simon Schafer. Please state your company name followed by your question.

[Simon Schafer:]

Yes, thanks so much. It is Goldman Sachs. I wanted to ask about not necessarily the fourth quarter outflow in terms of working capital. I realize that can be revised probably it sounds like in the first quarter. But broadly speaking ASML is finding itself in a situation whereby you have a very well capitalized balance sheet in comparison to really all of your suppliers as well as of course the majority of the customer base, I think, Nick alluded to some of that especially in Taiwan. So what are the risks that not the same as 2002 the company gets a lot more aggressive in terms of offering options, or really financing other people’s upturn, if that makes sense?

[Peter Wennink:]

Yes, I think that is – that would be something that you would think of and clearly I think the whole customer base is thinking of that and – but we make it very clear that we are not the central bank of the industry. So, this is something that we are highly careful with for the simple reason that you know. It is something when you – if you start doing this you probably have to start doing this for the whole customer base because they are all in the same position and they are all competing with each other. So that is something that is difficult to apply in a – on a broader scale. With respect to the suppliers, yes like we said we did pay our suppliers by the end of the year nicely on time for the inventory that they shipped us which customers started to push out, which actually settles that – basically settles us with them on the outstanding balances. Now we won’t need a lot – we need very, very little over the next two quarters because what we anticipate that we will sell we basically have in inventory. So that will come back in terms of cash. That means that you are drying up your suppliers. Now the question is how bad is that? If we compare this to the 2001, 2002 downturn, our supply base has actually changed in the sense that we have fewer very large suppliers that we depend on and those suppliers are bigger. We’re seeing also the consolidation in the supply base and also many of those bigger suppliers have learned and have diversified much more than they have ever did before. So in the scan that we have done through the supply chain we have not identified any major supplier that will call that risk at this moment largely because they’re bigger and they’re more diversified than they have been before. So from that point I don’t think we would see a cash drain and on the customer side I think I was clear on – what our policy is.

[Simon Schafer:]

Absolutely, that is very helpful Peter, thank you. My second question would be and I realize it is very difficult and my question isn’t necessarily related to anything demand related, but just if we just assume for a second that companies shrink roadmap is on track and that is sustainable for the next 3 years or so, what you think is a normal sort of run rate in terms of the requirement of annualized number of machines that you customers may have to take, assuming that they are not expanding any capacity of course. So, just technology purchases, what would be a normalized type of run rate assuming that customers want to stay on as well.

[Eric Meurice:]

We have estimated this about 6 months ago and the number was about $2 billion of sales for us, which is – which includes mostly immersion tools and a bit of service to it.

[Simon Schafer:]

Okay, thank you so much.

[Operator:]

The next question is from Mr. Cluner Flugger [ph]. Please state your company name followed by your question.

[Cluner Flugger:]

Hi, good afternoon. It is Cluner Flugger at Timura [ph]. I was wondering on the immersion demand if you could differentiate the different level of confidence, the different customer groups even on the main 3 groups. Is that a different level that you feel?

[Eric Meurice:]

Yes, that is also an interesting question. I think we all believe that the DRAM industry was in dire situation by not having a sustainable business model and I would say at this moment probably most of us start to feel extremely comfortable that the required consolidation by the Taiwanese government is going to make a bit of a difference. Of course we haven’t seen the results yet, but it is encouraging that the Taiwanese government has said we will help you and support you but we want to see a businesses trend that works, which is obviously a one that says you cannot over-invest and be two many of you in the pie. So there would be a restructuring and retiring of capacity. We can trust I suppose the Taiwanese government. They have very well prepared – developed the foundry business as we know. And we think that very good thing. Also you have seen that now the big players in DRAM has – have retired capacity and are playing to be a bit more smarter in the negotiation of contract pricing to avoid getting into the current situation. So, I would feel at this moment a fairly – I would feel very good that the DRAM players will be the ones which will get us out of the crisis. The flash people are good but they over dreamt new applications as we all know from last year. Now you will hear them say that the engine of growth is the solid state drives and the add-on cards. This is good to hear. This is an obvious area of growth and this is an obvious area where they need if one achieved cost levels and obviously because we are offering customers latest cost level, cost points and we are giving potentially one or two more node opportunities. We feel very good that they will be sustained by this newer growth engine. So in other terms now that the dreams are corrected and the new engines are very clear and they are based on technology, we also feel comfortable that we will expect this to be coming a bit after the DRAM. Logic, our logic has been slow for lithography for a long time. Now they are hit by demand at the moment, where they ran 45 and also when the people who have gone foundry like ST-NXP, Freescale et cetera are starting to run these new technologies or will be. So, I expect that to also happen, but I will expect this in 2010.

[Cluner Flugger:]

Okay, and just as a follow up, is it fair to say that your views on the logic sector, this is the area where you have changed most, because I remember last quarter we talked about if that one foundry on a month [ph] we talked about AMD. Is that the area where you have changed most at the outlook?

[Peter Wennink:]

Well, in fact the outlook yes of foundry has been lowered compared to what I said, one a month – one immersion a month is not the case, but we are shipping out at this moment, say one a quarter. So, it is not good. But remember these are people who say I need the machines but I am end driven, end-market driven. So, I didn’t have Christmas season. So, I don’t need the tools. So this is regarding these ones, the 45 nanometer guys, which we have identified. They will restart at the moment and restart. Remember the demand curve went down 6% first time in the life time that the unit semiconductor within one quarter went down, went from a plus, I think it was a plus 8% on average here and then went down minus 6% compared to the run rate of the year before. So it is a tremendous hit on the logic side. So, I did change significantly the forecast on those guys, but on the other hand they have never been the bigger player. So what is impacting us in our top line is still the DRAM and the flash.

[Cluner Flugger:]

Okay, thanks. That is very helpful.

[Peter Wennink:]

Thank you.

[Operator:]

The next question is from Mr. Timothy Arcuri. Please state your company name followed by your question.

[Timothy Arcuri:]

Citi, hi. A couple of things. First of all, my question is around maintenance capex. You know, I know, you just answered a question that you think normalized revenues are kind of in the 2 billion range. I am assuming that that includes service. So, if you kind of back service out of that, you know that would imply using kind of a 16 million euro ASP that you are kind of talking about, maybe a 100 systems in kind of a normal year. And right now you are shipping you know, at run rate maybe 25 or 28 taking the Q1 number and kind of taking that across the year. So, you are shipping kind of a quarter of what you would consider normalized. And I am wondering what you are shipping relative to what you consider maintenance?

[Eric Meurice:]

Okay, well it is my fault. Usually I try to hide something, but you are too good for me. So I have to tell you the truth. When I say it will be the service I meant a bit of Brion already getting in there. That is the support behind Double Patterning for instance in the 2 billion. I didn’t mean maintenance and service business that we have on average about 100 million a quarter. So I am sorry about that. So, you should – I should rephrase in immersion business including certain numbers of RET/OPC type deals and et cetera that we added to a package, it will be about 2 billion and then you add 400 million of service and that would be the base business.

[Timothy Arcuri:]

Thanks, it is okay.

[Eric Meurice:]

Sorry about that.

[Timothy Arcuri:]

Okay. So, then when it gets back to the idea of really maintenance capex. So that would say that the normalized number of system shipments is kind of in the low 100s per year, say 125 something like that and you are shipping kind of 25 or 28 run rate in Q1. So, what do you think that the maintenance level is for the industry? So, normalize this kind of over cycle what you would expect to ship, but what is the absolute minimum that if the industry wanted to run a maintenance capex for the next 5 years what is the minimum number of systems that they would have to take per year relative to the number of that kind of 25, 28 run rate that you are shipping in Q1?

[Eric Meurice:]

Again Tim that is very – that is a too complicated question. You have – of course, we simulate to make it simple. You these – all these things depend on the numbers of transitions. What you call maintenance capex is in fact is a way to assume how may new nodes get done. So, we have transitions of course simulated every year for flash, every 18 months for DRAM, every 2 years for logic. Now that is clear. So if you add them up and you say what is the maintenance capex for that. That replace, I would say 10% or 20% of the business with new chips versus old chips, you know that type of things. But you cannot even answer that question because then if you make a conversion from 65 to 55 or from 55 to 45 the answer is different. In other terms, the maintenance capex as you said just to make the transition of Moore’s Law, grows. When you make – if I give you the answer of 2 billion, which I made the calculation because I did it last – say 6 months ago, that 2 billion corresponds to a transition of 45 to 35 in flash, of 65 to 55 in DRAM, and half of a transition of logic to 45. No, not much. If I – I will try to answer your question with my memory of what is going to happen a year later when you go one more step in the technology, you probably will create another 500 million. But I do this by memory, so 2.5 would be the necessary transition cost.

[Peter Wennink:]

If everything of those transitions for logic and DRAM and NAND all happened in the same year, yes? And so that's where they are excess, right. There is a timing difference in when – what you call maintenance capex, we call the technology transition, which I think is a better word, just that needs to be, and that needs to be done. And that adds up to – we were in fact if you take 2008 six months ago where we said 2008, we all knew that that was not going to be different by capacity demand, but largely by technology transitions. Then we were looking at about a 2 billion sales number in terms of system shipments and add to that 400 million of service, I don’t know whether you recollect that, but we have been talking about those numbers. That is the kind of what you would call maintenance here, we would call that a technology transition here without any capacity additions. And that could be, if like I said, if those technology transitions would also include a full, let's say a logic node transition on top of DRAM and on top of flash, that could be 2.5 with another 400 million of chips. So that’s how you need to look at that.

[Timothy Arcuri:]

I see okay. And then my follow-up question would be on service, and you know we’ve heard from some other large companies that there was a pretty big fall off in service business at the end of December. You didn't see that obviously, and I'm wondering what you are allocating for your service business in March and June given that you have a significant number of capacity – significant number of shutdowns and there are some big DRAM customers stripping down all tools or spare parts, I'm wondering what your outlook is for service?

[Peter Wennink:]

Yeah, we do see some pressure on the service revenue, but also want to remind you that service revenue was never a major part of our business as compared to some of peers which is quite different. I mean if you are in the deposition area, your services is required in a large part. We have never been there. Now we do see some pressure on cost with our customers where actually they are looking at service contracts. We see pressure between 10 and 15% top line, but that is about it. But what we also see at that the other side like in Q4, we see system loops, customers saying, are you going to reallocate certain tools to other lines, are we going to rationalize our factories, which actually means it is a lot of work because we need to be installing and uninstalling tools. So that comes on top. So we haven't seen in Q4 a big drop-off because of those additional activities, but it is fair to say that cost reduction pressure from our customers takes a opposite pressure on the top line of our service contract revenue which is ours of between 10% and 15%.

[Timothy Arcuri:]

Okay, great. Thanks a lot.

[Operator:]

Next question is from Mr. C.J. Muse. Please state your company name followed by your question.

[C.J. Muse:]

Yes, it is Barclays Capital. Thank you for taking my question. I guess first question, I guess a month or two months ago, you guys were talking about 40 to 60 immersion tools as your likely shipments in 2009, and I guess I was wondering given what has transpired over the last 4, 6 weeks, how has that number changed, and if so where have you seen strength and/or weakness?

[Eric Meurice:]

We’d – again this is a bit related to what Nicholas was asking. So first of all, we have no visibility as to the year itself. Anything can happen. As we said, it could be a U curve type recession, so we have one or two extremely bad quarters and then everybody gets exciting because the Christmas season is not that bad and the world has recovered, and as I said, it'll definitely be a U curve, and we recover and it will be a lot of immersion tools in Q3, Q4. But it could be the opposite and the world collapses further because unemployment grows and et cetera and things doesn't happen this way. So I don't know. But to answer the question about the 40 to 60 which was I think 65, this deal corresponds to say end of demand or simulated demand when the world's comeback.

[Peter Wennink:]

Perhaps I can maybe try to help you there under the circumstances which of course you may understand is quite difficult to do. Good predictions, but let me give you just a bit of an idea. On the back half, we have 19 immersion systems, let’s say roughly 20, which is for customers where we have a high level of confidence that will ship those. Now we have – one of the previous questions was, what if this reorganization, this restructuring of the Taiwanese DRAM industry actually happens, and what is the pent-up demand that, which is between 15 and 22, it was originally between 35 and 40 tools? On top of that, you could think of and I won't think about logic or foundry because I guess it’s largely probably 2010. There are memory plays that we wouldn't call first year in terms of financial capability because there are a few that are actually in our backlog which we feel very pretty comfortable that they will be able to pay their bills. Now if we had to have 15 to 20 pent-up demands from the Taiwanese dealer foundry, then there are also other memory plays which could act to death, if they get access to some funding which we have seen in Korea, which is normally the order backlog, but which could transpire into some level of orders for the second half of the year. It could then come to this between 40 and 60 number, I think we can, because what we have in the order backlog plus the pent-up demand in Taiwan brings us already to 35 to 40. The other DRAM players place a few orders, could end like I said it could. Will that happen, I'm not sure, we're just talking about potential out there.

[C.J. Muse:]

That is helpful. Can you comment on whether you’ve seen any push from the NAND side of things relative to what your expectations were a couple of months ago?

[Eric Meurice:]

We have had thing to push and I wonder – I think we already said that at the last call during the warning, so yes, pushes have been happening and I even understand that some flash capacity have been transferred to be DRAM, so the flash capacity has also been scaled, which is why we say we think we have scaled down by 20% and DRAM capacity in general has been scaled down by 30%. But at this moment in the past whether the two to three weeks of January, we are starting to understand that the flash people are re-planning. That doesn't mean they are optimistic, don't misunderstand me, but there is a difference between pushing out and starting to say when do I buy back. So the only thing I can tell you is at this moment we are in process of when do I buy back type activities was most of it.

[C.J. Muse:]

I guess my question was relative to your pre-announcement, since then have thinks gotten worse or stabilized on the NAND side?

[Eric Meurice:]

Stabilized.

[C.J. Muse:]

Okay. And if I could follow up real quickly on the EUV side, I think you're planning to shift in 2010. Can you talk about your current visibility there and what your expectations are in terms of actual tools shipped in the first half of 2010?

[Eric Meurice:]

In the first half of 2010, it is probably one-ish or something; in the second half, is probably when we ship the rest, the five. As you know, we have five orders. These are all pre-production in a sense that they will be serving the purpose of building recipes for customers that is technology. They would need about and year and a half, if not two, for some customers to then be able to ramp full production. So 2010 should be five, maybe 2011 should be another whether a five or so, which is the other set of customers making technology, and 2012 is when you start production.

[C.J. Muse:]

Very helpful. Thank you.

[Operator:]

Next question is from Mr. Sandeep Deshpande. Please state your company name followed by your question.

[Sandeep Deshpande:]

Hi. JPMorgan. Just a quick word on the second half, I mean given that you have a particularly lead time, I mean I believe it’s even five to seven months on some of the highly leading edge tools that you will ship, I mean even if the order intake starts to improve in the second half, I mean can you comment Eric on how you see the ramp up of the shipments, and so is 2009 even if you see some pickup in the second half essentially a write-off in terms of actual shipments happening this year?

[Eric Meurice:]

Unfortunately I think we're going to try to tell the customers that in view of a bit of a restructuring and push back on the supply chain, this should be reasonable and give us appropriate lead time. So what is appropriate lead time? On a immersion tool, appropriate lead time would be six months, but I unfortunately believe that the curve is going to be three to six, is going to be – so you cannot – you will not be able to read anything in the first quarter in terms of order entry. The bookings number in Q1 may be very small, maybe higher, but it is still part of the activities of wait and see. I don't see any bookings from Taiwan because I know they are negotiating and what they are negotiating is good. But you can imagine that the Taiwanese type companies, when I say Taiwanese, I mean the companies involved in the Taiwanese discussions really telling us what they want in July. But this is not going to be orders.

[Peter Wennink:]

And Sandeep, clearly – I mean, the inventories, we do have significant inventories of stuff that we plan to build and ship as leading edge, so we plan to ship a lot of immersion tools. So there is an immersion inventory with us and within the chain. Now clearly that is not all being assembled and integrated into new systems, but so that's why you cannot say we have an extremely short order lead time. We're trying to educate customers also saying, listen, if you want tools, let’s say in the fourth quarter, you'd better start thinking of giving us orders three to six months before. And we would be able to respond to a certain level because, yes, like with this big down (inaudible) not only put inventory on our balance sheet but also on those of our suppliers.

[Sandeep Deshpande:]

And quickly you know a follow up, I mean given that some of these, I mean you know what you have shipped to the leading edge DRAM companies for instance without naming names. But given that some of these tier 2 and tier 3 DRAM companies are likely to be bailed out or have their consolidation plans done by the second half of the year, I mean do you have an idea of what percentage of DRAM being shipped by some of the leading companies next year will be using immersion, so which will force essentially these tier 2, tier 3 companies to speed up their move to immersion simply because otherwise they won't be able to – I mean a bailout will not be meaningless essentially without being able to be competing with the leading edge players?

[Eric Meurice:]

That is a very good question. The way – however, the marketplace is – at this moment, everybody has frozen, even the top tier types, because of the sudden drop, the low capacity needed, the loss of money et cetera. So, there is a sort of, no, we have got to wait and read the market a bit. Your questions about how do some people take advantage of the others being down or not being capable of taking decisions, I think this will be playing in the next three months. You have to give it a bit of a – bit more color into how bad the recession is.

[Peter Wennink:]

It also means that (inaudible) also in addition to that, your question saying, yes, it is indeed true that you have to ask yourself a question, what is the use of a reorganization if you don't make that new company more cost competitive, which actually means the pent-up demand in the second and third tier must happen.

[Eric Meurice:]

I think Sandeep's question was, what was going to happen to the bigger guys who don't know how to wait. And I’d say at this moment they are not taking advantage of the situation because the environment is so uncertain. But if the environment starts giving some positive news and you have heard that the price of the DRAM and the price of flash are going to go – coming up, is why I am starting to tell you that those large customers are starting to talk to us about planning. So in other terms, Sandeep, if you were in the DRAM business, you would already probably – but the others are waiting a bit and we cannot say that we will see the orders going up, or order going up in Q1 for Q2 delivery for instance. I think that was a much, much true value.

[Sandeep Deshpande:]

Thank you Peter and Eric.

[Operator:]

The next question is from Jonathan Crossfield. Please state your company name followed by your question.

[Jonathan Crossfield:]

Yes, Bank of America / Merrill Lynch. The question I had was, over the past five years, as you mentioned, Eric, we have seen transitions in NAND every 12 months, transitions in DRAM every 18 months. What is the chance do you think that over the next few years, as we come out of this recession, the rate of technology transition changes and memory maybe slows down on a longer term basis?

[Eric Meurice:]

That is an excellent question in the sense that that is the crux of our growth and profitability and market share. So I don't think the customers will change the profile that is more technology you can develop, the more differentiation you have, the more you take market share. And as volume is fundamental for profit in this business, they will jump on every opportunity. So if the equipment vendors offer them the options of doing it, they will do it. And if they have the financing to do this, they will. The only reason for a potential slowdown would be that products would not anymore make use of their chips. So in other terms, a sort of application slowdown. We used to say in the semiconductor industry that this will never happen. I think we have to be a bit less proud than that and we can – it may always happen. We have to be looking into any signs of this. You have seen bit of a sign of this in flash when everybody thought that when you had 2 gig, you can go to four, and when you have four, you can go to eight. And you have noticed that it is not as much business in the four gig than before expected it to be. Well because there was an application failure, people thought that cameras and things could carry 4 gigs, in fact they don't. Two gig is good enough or one gig is good enough et cetera. So clearly the application curve is a fundamental term. Now we are not the specialists here, so I'm not going to give you more than that, and you will be able to yourselves see if there is an application gap or not. At this moment that is why I mentioned that so to speak this can be add on cards, giving me enough of a good feeling to justify 35 nano, in fact 22 nano, to be honest. That is really where the thing is going to have to rush. After that, I have no clue, but as you know I am old now and my sons are not teenagers anymore, so I have lost complete understanding of the latest and greatest gadgets.

[Peter Wennink:]

And as also evidenced by the way Jonathan by the fact that when we talk to our leading edge memory customers about their needs, they are adamant about their EUV level shrink needs. I mean they don't back up.

[Jonathan Crossfield:]

Right, okay. And then maybe – sorry helpful. Thank you. Maybe just a follow up for you, Peter. On the balance sheet this quarter there was a $50 million increase in gross debt, now that didn't seem to come through the cash flow, so could you just explain the other side of that?

[Peter Wennink:]

Yes, the other side is on the other receivable parts. That is the fair value of the interest rate swap that we have, because we have both has a fixed interest rates which we swapped. And since the interest rates dropped, that’s a big fair value there that is on the other receivables. The offset then needs to be included in your long-term liabilities line. So it is really about $600 million of debt and the rest is the fair value balance sheet adjustment.

[Jonathan Crossfield:]

Very helpful, thank you.

[Operator:]

The next question is from Mr. Lee Simpson [ph]. Please state your company name followed by your question.

[Unidentified Analyst:]

Hi, good morning. It is (inaudible) from Jefferies. Most of my questions have been answered but if I could go back on some earlier questions. I mean given some of that DRAM and flash price firming that you see and a lot of the large customers come back and talk to you about new orders, what sort of confidence can you give us at this stage that 1Q09 maybe the actual order chop?

[Eric Meurice:]

Good try.

[Peter Wennink:]

Neat.

[Unidentified Analyst:]

Had to give it a go.

[Eric Meurice:]

Let me try and answer. It could or could not.

[Unidentified Analyst:]

Right, that's good. And if you had to prefer, which side of the answer would you give?

[Eric Meurice:]

No. Clearly, we don't know. So we have to wait.

[Unidentified Analyst:]

Okay, no problem. As I understand from some of the earlier answers as well, it looks like you favor the restructuring in Taiwan, let's call it restructuring, and playing out over the next three to six months and then pent-up demand in subsequent six months of something like 15 to 20 tools, is that what you're trying to tell us here?

[Eric Meurice:]

Yes.

[Peter Wennink:]

Yes. For those guys, yes.

[Unidentified Analyst:]

For those guys. And on top of that, you are seeing at this point utilization at foundries is probably so low that the recovery to sort of the reorder point is not going to happen until 2010?

[Peter Wennink:]

Correct.

[Unidentified Analyst:]

Right. That's all my questions. Thanks.

[Operator:]

Next question is from Odon de Laporte. Please state your company name followed by a question.

[Odon de Laporte:]

Yes. Thank you for taking my question. This is Odon from Cheuvreux. I was wondering in the backlog what is the weight of EUV, is there any EUV units in the backlog? Thank you.

[Peter Wennink:]

No.

[Eric Meurice:]

No EUV in the backlog. The orders have been taken before. And at this moment, as I said, we have to deliver those. Then there will be orders taken sometime in 2010, and after that you will have the ramp into 2011.

[Peter Wennink:]

The reason is, we in our backlog, we only take up orders for a specific period of time, i.e. 12 months. Beyond 12 months, orders in the semiconductor industry are generally seen as a wet dream (inaudible) that is not true. Those will happen.

[Odon de Laporte:]

Okay, thank you. I have a follow-up. If I try to estimate sales in Q2, given that 41% of your backlog is shippable in H1, can we expect a level of at least (inaudible) million euros of sales in Q2? Is that fair to say that?

[Eric Meurice:]

Yes, it is in a sense that I said that the current level of 180 to 200 is not sustainable at our customers levels. So we expect this to be an exceptional low number. Now is this going to be this for three months or six, I don't know, but lower than 180 to 200 is absolutely not in our expectations.

[Odon de Laporte:]

Okay, thank you very much.

[Operator:]

The next question is from Mr. Mehdi Hosseini. Please state your company name followed by your question.

[Mehdi Hosseini:]

It is FBR Capital Markets. Eric, we have spent the past hour trying to figure out when the inflection in the bookings would take place. I want to move a step higher and just based on what you know of your customers, if you were to look at your backlog, how confident you are in the backlog as of the close of Q4? And do you see any possibility that you may have to take some of the unit out of the backlog because there is a probability that some of your units or the shipment may extend, maybe shipped in more than 12 months? And I have one more quick question for Peter, what is the mean breakeven point after all the restructuring taking place?

[Eric Meurice:]

Okay. So, Mehdi, hi. No we don't expect any cancellation or further postponement of the backlog, maybe from one quarter to the other, but mainly due to logistics issue more than anything. We always have this type of activity, so the usual 2 to 5 per quarter is a moving target, but more than that not. Peter, what about the…

[Peter Wennink:]

Your point is on slide 29 of the presentation. It says after restructuring is approximately 450 million euros a quarter.

[Mehdi Hosseini:]

450 million euros.

[Eric Meurice:]

450.

[Peter Wennink:]

It is on slide 29.

[Mehdi Hosseini:]

Okay. Thank you.

[Craig DeYoung:]

Operator, I think we have time for one last question, and I would invite anybody of you who are unable to get through, the IR team is available to try to fill in your knowledge in understanding gaps, so if you give us a call, again feel free to call us, and operator if we may have one last call?

[Operator:]

Of course. The last question is from Jay Deahna. Please state your company name followed by your question.

[Jay Deahna:]

Thank you very much. JP Morgan. Good afternoon.

[Eric Meurice:]

Hi, Jay.

[Jay Deahna:]

Hi. The question I have relates to slide 13 on your hand out. That is the slide titled “product pipeline” and it illustrates the three different leading edge products that you have. My question is are those the products where the (inaudible) Brion holistic lithography options are available, are they are available on any other tools besides those three, and is the Brion holistic package sold as part of the core ASP, the product, or is that sold as an add-on that is in a sense an option that the customer can buy if they want?

[Eric Meurice:]

Okay. Yes, this is a good question and this is how we create value. So the holistic lithography package of product one can be sold on their own, you don't need to buy any of our machine, and clearly we have done this already at some level of sales which is several hundred million, but it is there. Second is you can buy some hardware options which are included into the machines and that would be the case of all of the products that you have on the page 11 which is XT 1950 which will be in fact the first one to carry one fundamental piece of hardware. But the NXT also will carry a piece of hardware are– sorry, I should correct myself, the NXT at this moment, there is not – EUV doesn’t carry that. How do we sell these things? We sell these things in three ways depending on customer wish. One is sometimes it is included in the price of the machine, sometimes it's included as a package, so you buy litho cruiser at the price of one or two million. And sometimes you sell a service, this is why I use the term service, and I should not, but a period of six months to one year of common activity to find solutions like an optimization, like matching of a number of scanners to a number of designs. So, we can do with three ways. So in other terms, your question is timely. We never yet – I will not yet answer the following question which is how many hundreds of millions of euro do we make with this business. It is a bit too early, but it is good to say that now we have three ways of selling and we are starting to sell.

[Jay Deahna:]

I guess another – that’s good, I appreciate that. Another way of looking at it is, let's say you take the XT 1950HI, okay, or the NXT, and you look at the kind of the base average selling price for the tool, and then you look at that tool fully configured with the holistic package with all the Brion offerings and what not, what would be the difference in the price? Do those add-ons increase it by say 10% or 15%? Or is it something more than that?

[Eric Meurice:]

It could be in the orders of magnitude of 10%, yes.

[Jay Deahna:]

Okay. And are those offerings available on any of the products prior to the 1950?

[Eric Meurice:]

As I said, yes, but it would be less than 10%. These are products which have a bit less impact, not so much because they're not linked to the product, but mainly because these products have less impact on node of lower categories. So in other terms, it is nice to use Brion when you are trying to do something difficult, but when you try to do something simple, you don't need as much of the capability of lithoplus.

[Jay Deahna:]

Okay. And absolutely lastly, do you view these offerings as enhancing your sustained competitive advantage versus your competition, because it looks like with every new generation now there is a significant change in the complexity of the machine which historically has been good for ASML, this is an add-on to that. So do you view this as kind of critical or very important to your sustained competitive advantage, these offerings?

[Eric Meurice:]

Yes, yes and yes. This is a fundamental part. We believe that with this options, we do think that the competition absolutely cannot do on two ways. One, our hardware itself is we think stretchable to a level that the competition hasn’t achieved as we know by history. We do better overlay. We do stronger, I would say, well, overall robustness that creates better CD, and of course we have a higher throughput. The Brion tools clearly enable stretching the hardware, so because you stretch your hardware, you increase the competitiveness of the tool. In addition, the software stuff are also not provided by our competition. So we have got a significant competitive advantage for those products.

[Peter Wennink:]

And also Jay, it is not just ASML marketing talk. I mean there's a big logic maker in the US, very openly talked about the device scaling through what are called source mask optimization, which is basically (inaudible) more and more embedded, need some of the features that are included in this product pipeline.

[Jay Deahna:]

Right, great. Okay thanks very much.

[Eric Meurice:]

Thank you.

[Craig DeYoung:]

Okay, on behalf of ASML, I would like to thank you all for joining our Q4 and 2008 annual results conference call. Operator, if you could close this out officially?

[Operator:]

Of course. Thank you, sir. Ladies and gentlemen, this concludes the ASML 2008 fourth quarter and annual results conference call. Thank you for participating and you may disconnect your line now. Have a nice day.