艾司摩爾 (ASML) 2008 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by.

Welcome to the ASML 2008 third quarter results conference call on October 15, 2008.

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 now like to turn the conference over to Mr.

Craig DeYoung.

Please go ahead, sir.

Thank you, operator.

Good afternoon and good morning, ladies and gentlemen.

This is Craig DeYoung, Vice President of Investor Relations and Corporate Communications at ASML.

I'd like to welcome you to our investor call and webcast.

As the operator has mentioned, the subject of today's call is ASML's 2008 third quarter results.

Hosting the call today are Eric Meurice, ASML's Chief Executive Officer, Peter Wennink, ASML's Chief Financial Officer.

Eric's joining us today from Tokyo, Japan.

And I mention that just in case there's any technical difficulties where he gets separated from the call, we'll make sure that he rejoins it as soon as possible.

I'd like to draw your attention now to the Safe Harbor statement contained in today's press release and third quarter results presentation, both of which you can find on our website at www.asml.com.

The length of the call will be 60 minutes.

And now, at this point, I'd like to turn the call over to Eric Meurice for a brief introduction.

Yes, thank you, Craig.

Good afternoon, good morning, and thank you for attending our Q3 2008 results conference call.

Peter and I, as usual, 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 the Q3 financial performance, with added comments on our short-term outlook.

I will complete the introduction with some further details on our longer-term view.

Following the introduction, we will open up for questions.

Peter, please.

Thank you, Eric, and welcome to everyone.

First of all, like Eric said, I would like to review our Q3 result highlights.

The [third] quarter ended, clearly, in line with our guidance with revenues of EUR696m on the shipments of 37 systems, 26 of which were new and 11 were Used.

Amongst the new systems, 16 were the Immersion systems supporting the strong and also stable new systems ASPs of EUR21.6m, with the ASPs of all systems being EUR16m, which is the result of a larger portion of new systems sold in the third quarter.

Our gross margin of 38.1% was consistent with the lower sales in this quarter versus the previous quarter's.

And, along with our reduced operating expenses of [EUR182m], our operating margin for Q3 was 12%, allowing for an operating margin year-to-date of close to 17%.

We maintain a strong cash balance.

At quarter-end, it stands at EUR1.3b.

We have debt of EUR600m due in 2017 and we have a standby facility of EUR500m.

Our bookings were nearly flat in units quarter on quarter, with 31 system orders in Q3, 18 of which were new and 16 of which were Immersion.

Although we booked a larger number of new systems in the quarter, those systems did carry a high ASP of about EUR7.7m.

This high used system ASP was due to the fact that we booked four Used XT1700 Immersion systems in the quarter, leading to a total booking value of close to EUR500m in Q3.

This leaves us with a system backlog value at the quarter end of over EUR1b.

Now, the current financial turmoil and the significant overall economic uncertainty creates a situation in which our customers are reassessing their own capacity ramp-up plans, and those of their strategic alliances.

Furthermore, our customers are confronted with lower-than-expected demand causing over-capacity, low memory product pricing and, important, a restricted access to capital which limits their ability to ramp-up to latest technology.

Now, due to this wait-and-see attitude of our customers, we are unable to guide on short-term bookings or to give any type of mid-term sales forecast.

As mentioned before, a significant appetite for leading-edge lithography systems still exists as memory makers have designs ready on further cost reductions through the shrink while reducing old line capacity at the same time, as we have seen announcement of 200mm capacity taken out.

Also, logic makers push for greater device performance through shrink of their feature sizes.

We do expect that for the foreseeable future virtually all customers will continue to postpone non-leading edge capacity additions.

However, our current view is that the need for continued Immersion investments amongst those customers that do have the capital structure to invest should enable us to secure positive operating margins in the quarters to come.

As such, we currently maintain a 20 -- a 27 Immersion system order backlog, with a value of [EUR750m].

In view of these customer issues I've mentioned before, we plan to ship only 27 -- sorry, 26 systems in Q4, with an average selling price of EUR16.5m.

On this lower shipment volume our gross margin for the quarter will be approximately 36%.

And this will allow us to end the year at the previously guided approximate 20% decline in ASML's full-year revenues which, by comparison, will prove to show out-performance versus many of our peers, by the way.

On the costs side, we are determined to improve the efficiency of our R&D efforts.

We're not cutting on R&D programs.

And, as such, we expect our R&D expenditures will drop to about EUR124m in the fourth quarter, but further cost efficiencies are, indeed, expected.

And, as seen in Q3, we have further trimmed our SG&A cost significantly, I would say, and will cut costs further to EUR47m in Q4.

This represents a 19% reduction in two quarters, which I think is a clear evidence of our flexibility model.

Also, in the area of production, we will reduce cost through supply chain management and through the use of flexible labor and contracted activities.

I would like to remind you that also, as per our last call, discussions concluded with the Dutch tax authorities regarding the treatment of taxable income related to our patents, have resulted in structurally lower effective corporate income tax rate in the coming years, and which will be at an estimate 20% annual tax charge.

In summary, we remain cautious in an environment of a global financial crisis and uncertain economic future.

Fighting against product oversupply and falling product prices, our customers are struggling with their own business uncertainties.

We are prepared to adapt our Organization and cost structures such that we can weather these storms.

But in this environment we will keep a sharp focus on our technology programs by balancing efficiency and execution to make sure that our technology leadership is reinforced.

I now turn back to Eric, who will give an outline on our technology progress, which, I said, is a key behind ASML's clear robustness.

Eric.

Thank you, Peter.

So Peter has highlighted the fact that in these very uncertain times ASML is arguably doing better than competitors or peers, in terms of yearly sales decline in a very challenging environment.

And this is a proof, a testament, to ASML's strong product positioning in support of customer technology transitions, which result in high system average selling prices and continuous market share growth.

Now that's a statement.

We've been doing, in fact, fairly good if you add up the fourth quarter of this year which will be remembered in the industry as a bad year.

But during this downturn which is starting, we will continue, as Peter say, our strong investment in R&D.

But we'll strive to be as efficient in that spend as possible, with a bit of adjustments and trimming being made so that not to impact any strategic programs but, I would say, get us to be more efficient.

We will not touch any strategic programs, in particular Immersion double patterning and EUV.

In Immersion, our Immersion business continues to show significant strengths, as we again shipped 16 systems this quarter.

2 -- 16 additional Immersion orders, so it's pretty nice to have a book-to-bill of one in this situation, leaving 27 Immersion systems in our backlog.

We've shipped 46 Immersion systems year to date and we now have more than 110 Immersion systems delivered to over 20 individual customers worldwide so, clearly, a leadership position.

We expect that our Immersion market leadership will even strengthen as we go into 2009.

We are introducing our new TWINSCAN XT1950 Immersion system which, thanks to a faster stage and an improved lens, has a performance increase of about 25% compared to its predecessor, the XT1900.

This makes this tool, first, increasingly affordable for 55 nanometer and 45 nanometer processing, which is what the predecessor was capable of doing but, in addition, extend the performance of this machine to the 32 nanometer node.

It is clear that this machine, with its unique lens summarized by the numerical aperture of 135, and the overlay of 4 nanometer specification, with this, this tool has no competitive tool match in existence, or even announced by our competition.

In addition, we have now, today, formally introduced our next generation multi-purpose platform, the NXT.

It will have unprecedented speed of above 200 wafer per hour and overlay accuracy in the 2 nanometer range.

And we'll begin to ship in mid 2009.

It will be positioned first as a very advanced tool for double patterning, but will soon become the future cost-effective Immersion platform of choice.

Beyond Immersion, and beyond double patterning and, as reporting in our recent EUV conference in Lake Tahoe in California, we have achieved significant breakthroughs in our Extreme UV program, as the productivity of the EUV Alpha-Demo Tool which we have shipped, as you know, in Albany in New York and IMAC, this tool is now reaching initially targeted throughputs, and as light source suppliers are now confirming road maps which enable system performance well beyond 100 wafers per hour.

Starting in 2010, we will ship 60 wafer per hour production EUV system upgradeable to 100 wafer per hour.

System with even higher speed performance growth of more than 140 wafers per hour at resolution of 22 nanometer and less are on the road map for 2012 production.

EUV will provide game-changing lithography cost dynamics which will lead to widespread adoption of this technology by 2011, 2012.

We will be well ahead when this market transitions takes place.

In parallel to our focus on R&D execution, and as hinted by Peter, we plan to accelerate our operational efficiency programs for us to become an even stronger Company, capable of weathering the current crisis, but deliver -- and deliver profit.

And even though I would say be more aggressive price-wise.

We have developed structural manufacturing and supply chain improvements which are already starting to yield cycle time and cost improvement, as we make short-term adjustment to our cost base using our embedded labor and contractor flexibility.

And Peter gave you two or three statistics.

So, although it is much too early to show optimism as to the economic environment, it is, however, important to note a certain number of fairly positive developments.

The first one is the euro weakening versus the yen and the dollar.

And, as you know, we are a euro Company, our competitors are yen based and our customers are buying in dollars.

So, being or becoming a price aggressive Company due to the weakening of the yen is a good thing.

I remind everybody that euro has appreciated versus the other currencies by a factor 60% or 80% in the past four years.

We have weathered this appreciation and still gained market share and increased profitability.

The fact that the tables are turning is in fact a very, very good encouraging possibility for us to proceed giving even more value to our customers and probably even reinforcing our market share gain and protecting our margin.

That is good.

The accelerated retirement of FLASH and DRAM 200mm capacity is also a good thing.

You have heard a significant numbers of our customers have now announced the retiring of those Fabs.

You have DRAM Fabs in China, in Korea, in Taiwan, in America, which have been retired.

You have even FLASH Fabs in Korea and in America which are being retired.

This is unprecedented in the industry.

This is, in some ways, bad news for our customer in the sense that the price pressure forced them to do so.

But, clearly, we are now witnessing a significant capacity shrink in this business which is, again, a very good thing for the need for further CapEx.

And then the third point, as we have mentioned it on a regular basis, the virtual stability of 300mm capacity during the year 2008.

In other terms, we and our competitors have not shipped in 2008 a lot of capacity build in FLASH, DRAM or Logic.

We have really much concentrated in technology transition capacity but not so much in real square meter capacity increase.

So, in other terms, if you add my point number two, which is retirement of 200mm, and my point number three, which is the fact that we haven't added significant 300mm capacity, you can consider that this all will contribute to a very healthy restart of the business when electronics and demand strengthen.

That's a problem of the day.

The end demand has [tanked] for the past, say, two months or so.

This has created a reaction in terms of CapEx build, but this reaction is healthy in a sense that it doesn't create over-capacity of any sort and allows us, I would say, a springboard of restart.

So, in conclusion, it is natural that a CapEx industry like ours is impacted by a short term -- in the short term by the uncertainty of the overall semiconductor end demand.

It is natural in a wait-and-see attitude, or in a slowness of end products, CapEx is squeezed.

Customers can always wait a certain numbers of weeks and months.

And this is what we are having today, so very natural position.

However, we're confident that in the mid-term future, due to our strong technology and market position which I talked about, and due to the current rationalization of the Fab capacity worldwide or, in fact, the acceleration of this rationalization, we're fairly confident about our mid-term future.

So, on this, I would like to open this to questions.

Thank you.

(Technical difficulty) Thanks, Eric.

Ladies and gentlemen, the Operator will instruct you momentarily on the protocol for the Q&A session.

But beforehand, I'd like to ask you to kindly limit yourself to one question with one short follow up if necessary.

This will allow us to get to as many callers as possible.

Now, operator, if we could have your instructions and then the first question, please.

Thank you.

Ladies and gentlemen, this start -- at this time we will begin the question and answer session.

(Operator Instructions).

One moment, please, for the first question.

The first question comes from Mr.

Menon.

Please state your Company name followed by your questions.

Hi, thanks.

It's Janardan Menon from Dresdner Kleinwort.

It's a question on a couple of statements you've made.

One is saying that you have some degree of confidence that your -- you will be bale to have a positive operating margin in coming quarters.

And on the back of the envelope I calculate that you probably need about EUR450m or so of revenue to reach a break even.

You've also said that 62% of your backlog will ship in the next two quarters, of which about EUR430m is shipping in the current quarter.

So that leaves you only about EUR213m of revenue from your backlog to be shipped in Q1.

I was just wondering do you have enough visibility on your order intake in Q4 to ensure that you do get the remaining EUR240m at least of revenue -- of orders with shipment dates of Q1'09?

Eric, you want to take that?

Sorry, I was muted and I was starting to talk.

So I think Peter should take the questions on your break even etc.

That's a good question.

I can take the question on the bookings.

So, go ahead, Peter.

Yes, on your first -- your question, I'd like to refer to our guidance on Q4, which is just above EUR500m sales level with a 4% operating profit.

[We have there between 3 and 5].

So, yes, I mean if you do your math and you include some of the cost reductions that I talked about, I think your assessment is a reasonable one.

Okay.

So you do have enough visibility that you will get that much of orders released?

I mean you (multiple speakers).

Okay, so --.

That's what Eric will answer.

(Multiple speakers).

Yes.

At this moment, we -- again, the key message that we're giving to you today is we have a serious uncertainty regarding timing of all these projects which we've been working on with the different partners in the industry.

So the timing is the issue.

But the size of the business in 2009 we are comfortable that at least, for instance, Immersion is going to strengthen and probably grow compared to this year.

So in Q1 achieving an additional EUR240m, so booked into Q4, is clearly within reach.

Okay, and just a follow up.

You took in quite a lot of orders from a major US IDM, as per your backlog statements in your presentation.

How do you see your market share development at that IDM?

And is it something that makes you confident about 2010 at this point in time, the kind of orders you've got?

So certainly this is a very large IDM, as you can imagine.

And this is an area which we -- a customer that we've said we are developing a potential entry into a very large node which will ramp into 2010.

So the fact that we got, I would say, the first order for this process gives us a confidence that we have a high chance of winning a large part here, yes.

So you will regain market share that you may be losing in 2009?

Absolutely.

Thanks a lot.

Next question, Mr.

Gaudois.

Please state your Company name followed by your question.

Yes, hi there, Nic Gaudois, UBS.

Just a first question, coming back on the Immersion probably growing basically next year versus this year.

And what gives you this level of confidence, Eric?

I guess the two questions in my mind is that, with where we are now, Immersion is partly dependent on capacity requirements, not only shrinks from existing capacity.

But also, importantly, and some of your reasonably large customers may actually struggle to have a balance sheet to necessarily do the shrinks they would like to do.

And, in other words, might actually see some delay in their shrinks road map.

So maybe if you could help us a little bit on that, it would be useful.

And I've got a more specific follow up.

Yes.

The reason why we believe that there is growth in Immersion next year is because the numbers of engines where you need Immersions is much bigger than -- in 2009 than it is in 2008.

So you are right to say, however, if 2009 is economically challenging these engines will be tamed a bit, but there are so many of them that we expect this to -- the numbers of opportunities to win against the, I would say, the difficult economic situation.

So let me give you more color.

Obviously, if you talk about the first tier DRAM customers, mainly in Korea, they have to run the new technologies in DRAM.

And at this moment it is a key set of projects to move beyond 5x in first tier DRAM.

So that's an engine that we did not expect, and it somehow is triggered by the recession, the need to in fact show even faster transitions.

The second set of customers, which you could call -- you can call second tier DRAMS, those guys have not invested anything in Immersion this year, in particular because of the difficulty of accessing cash, as Peter say, as you say.

But it is now vital for them to do so because they cannot let the first tier take two years ahead of them and potentially even get into another node while they haven't yet transitioned at all into Immersion.

So are they going to get cash?

Well, you may have heard in Japan that Peter has now secured a $1.7b or $1.6b of cash from a set of banks here.

So this, I think, is resolved.

You have heard Micron has been securing cash with its partner in Taiwan, Nanya, and they have also resolved the strategic questions about where they are going to put all those tools.

As you have heard, they have now at last taken a decision on the acquiring of the Inotera share they didn't own, and that gives them a significant springboard.

By the way, as you know, we have 100% share at Inotera at this moment, so we're very satisfied with this development.

And, again, this environment, Nanya, Inotera and Micron has no Immersion of volume, I would say, at this moment and these are leaders in the -- in what they do and they have secured cash.

So you know that something is cooking.

You're talking about Qimonda.

Qimonda is obviously in a weak environment at this moment.

They are looking at their options in terms of partnership, but they just got $350m of cash due to the sale of Inotera.

And they have potentially already got some Immersion tools from us at this moment and the question is how much in fact -- how many tools can they really afford?

But they have to move.

So you've got that segment which, I would add, will include one more -- two more Taiwanese who are being financed by first tier players so that, I would say, will give us a second engine.

And the third engine is the Logic engine.

You've obviously -- in fact, Janardan mentioned the significant order of Immersion tools that we received from this American IDM, and that will be shipped into next year.

And you have -- you can guess that we are shipping effectively to the companies doing what is now called computational scaling.

It's another big word to say lots of Immersion tools.

And that has to start, and will start, in Q1 and Q2 of next year.

You add to this Foundry which, at this very moment, is limited by end demand.

But, as you know, this is -- 2009 is a year of transition to Immersion and I estimated the volume to be one tool a month.

I don't think we will be very far from this.

Maybe I make a mistake in the Q1 timeframe again, because my timing may be impacted by the recession.

So all together, if you sum these opportunities, you should be above year 2009.

Thanks.

[2010].

Great.

That's very comprehensive.

Now on more of a detailed question, if we look at the revenues you got from Immersion in the quarter divided by the number of tools shipped, we get to an ASP of about EUR28.5m.

That's down from EUR29.3m recorded before [in first year recorded before that].

Is there anything mix related which explains that?

And how should we look at the ASPs next year with the introduction, of course, of the 1950?

Yes.

So, at this moment, one of the reasons of ASP deterioration, I would say in Immersion, would be mixed with Used Tool, because we are starting to have a fairly interesting and lucrative Used Tool business.

The other part is that we are now, I would say, customizing some of the tools for different nodes.

And you know that in the Logic arena they don't make full use of the 1.35NA lens and we are offering some special price for degraded lenses, again, to maintain the margin.

So ASP-wise it would be natural to cover now a bigger range, a bigger mix of products, Used plus, I would say, customized Logic products, and then 1950s which cost more than 30.

And, with this, come up with, depending on the quarter, an ASP which will be more or less high.

But, again, this is done so that we would try to maintain the margin.

Although the margin is of course under pressure, not so much because of that, but because of the overall situation where, if you have a minimum demand, of course, the customers -- it's a customer-driven market more than a supplier-driven market, obviously.

Thank you.

Next question, Mr.

Schafer.

Please state your company name followed by your question.

Yes, thanks a lot.

It's Goldman Sachs.

I wanted to ask about really more the costs side of the equation.

I think you mentioned, Peter, that perhaps you could expect some type of further cost efficiencies.

I think you're more talking about gross margin.

But if you were to find out that perhaps the cycle, as we see it today, is lasting around current levels or low levels for some time, how much aggressiveness could you show on the operating expense side?

Yes.

To give you an indication we -- in fact, we currently still have 20% of the workforce on contract, so that's called -- that's basically contracted activities.

So that is, in principle, if it lasts longer we need to go to lower business volume levels.

That is the first thing where we can go.

We have, in SG&A, of course we are [training] but we are close to 20% in two quarters, so that is where we probably could do a few million.

But that's not going to add a major amount of money to the bottom line.

In the supply chain management, we are -- we have started since the beginning of the year quite an extensive and comprehensive program on reducing cost.

So that's now running for almost eight months.

We will see the first benefits of that coming in in the first quarter of next year.

So that's basically the cost of goods, what we basically purchase.

And so that's where a level of flexibility is.

And we have about 50% of our R&D cost is flexible.

But like I said, currently, and what Eric also said on the expectation of our Immersion business next year, we do not anticipate to make major cuts in our R&D program.

So, although we do have that flexibility, we are not planning to basically use it as of today.

But it -- and it means that it is there.

Understood.

I mean (multiple speakers).

So, if we would see a major prolonged downturn in the industry and our customers would be cutting their R&D, or at least pushing their R&D programs back because they need to do something, we will be doing the same.

And we will be able to adjust our cost levels.

So on the operating expense side in the factory we still have about a 20% in the workforce, we have extensive and comprehensive cost reduction programs in the supply chain and we still have that famous 50% flex in R&D, which is probably the last item we will ever touch.

Understood.

Thank you.

And my second question -- and I realize you don't want to talk about customer-specific projects, but AMD obviously got a very significant outside investment.

And it looks as if really is going to be some type of additional foundry player in the field, which is changing the dynamics a little bit.

When you look at that does that become, or when would that conceivably become impactful, given that, historically of course, you've had a pretty tight relationship with AMD?

Eric, you want to take that?

Yes, certainly.

I think the first news that we have to take, which is a good news, is that AMD will be able to do more in their business before we talk about foundry.

So that is the first good news and this is part of my optimism about 2009.

They clearly will have to build an Immersion capacity for next year for their own needs.

And that's again, as I say, a fairly good business.

I will not comment about whether their coming into the foundry business will develop an opportunity.

It's, first, a question about what it is, what's their business model, what product are they going to deliver, what's the alliances that they are going to create to get some customers, etc., etc.

This, again, at this moment I cannot comment.

I would say the way we simulate the business for us is to say that to try to see -- to make a scenario as to how much silicon will be consumed, and then we do -- would not be concerned about who manufactured the silicon so, in other terms, if AMD takes market share out of TSMC or vice versa, we probably will be very much not impacted.

We will supply either player.

So at this moment we are very satisfied about AMD itself capable of investing in their future and developing a good technology, a good new process for the raw materials.

Understood.

Thanks so much.

Next question, Mr.

Timothy Arcuri.

Please state your company name followed by your question.

Mr.

Arcuri, your line is open.

Please go ahead.

Hi.

Thank you, Citi, couple of things.

First of all, if I take and I look at the number of litho systems that you're shipping relative to the number of units being produced out there, and if I do some normalization for some wafer sizes and some increases in throughput on the tools, you see that you're shipping roughly 50% less systems per chip unit being produced than even in the big downturn during '01.

So I'm wondering, a, if that would argue that there's going to be a pretty robust snap back here off the bottom.

But, b, I'm wondering -- part of that's obviously cyclical, but is any of that secular in that there just is not going to be the same number of systems ordered per unit even after you make all those adjustments because of customer consolidation, or some secular issue?

Or do you, in fact, think that it's cyclical and that there's going to be a huge recovery off the bottom as a result?

In fact, we vote for the snap back.

We have a simulation system, as you know, which so far has been pretty good at telling us when we have over-capacity or under-capacity.

And that our system basically says that in view of the consistent shrink road map, if the customer uses it that there's is going to be a need for machines at a much, much higher rate.

In particular because you are going to start hitting this correction of lack of capacity, I would say, and the need for double patterning which multiply the machine to two -- by two.

And the fact that you may have heard the industry starting to be worried that shrinking will create yield issues.

It's going to be pretty hard to get the same amount of dies out of each of the wafers.

And that also goes in the favor of having more tools.

So the secular curve is still showing us an opportunity to reach our famous EUR5b sales.

But at this moment I know, in view of the fear of recession, we are now talking about a break even and levels of EUR2b or EUR2.5b or EUR3b.

But the numbers -- the theoretical number is still towards EUR5b within the next two to three years from now.

Okay, great.

And then, I guess, just as a follow up, pretty much everybody on the sell side seems to be very, very negative on your stock.

And yet, if this is as bad as it gets, it's -- you're probably not going to lose money.

If you do, you obviously wouldn't lose very much.

So you sitting on this fairly significant cash balance.

So I'm wondering how do you think about a buy back from here, given how negative the street is, given that the stock doesn't seem to react any more to bad news?

Thanks.

I have to repeat the answer that we gave three months ago.

We have a clear financial policy that says if the cash balance of the Company goes outside the bandwidth that we have given, then that excess cash will be returned to the shareholders.

It's nothing new.

So that will happen.

Now, the issue is that can we do regular share buy backs.

I said it last time also, we're still -- from a different withholding tax point of view we are not allowed to do regular share buy backs, so we need to do that through a synthetic structure like we had in 2007, which actually is surrounded with a lot of logistical and legal issues that you don't want to do that for a small number.

So if we're outside the bandwidth of our cash balance, and it's a reasonable sized number, then we could do this same level of synthetic buy back as we did last year and we will do so.

But also under the current circumstances it is important to stay within our financial parameters as we have clearly outlined to the market and to the rating agencies, and basically to ourselves and to the shareholders.

So, currently, we stay within that and we stick to the quoted policy.

Okay, thanks, Peter.

Next question, Mr.

[Maurice Dolio].

Please state your company name followed by your question.

My question has been asked and answered.

Thank you.

Next question, Mr.

Ben Pang.

Please state your company name followed by your question.

Caris and Company.

Thank you for taking my question.

Regarding your engine number one for revenue driver, how many DRAM companies do you expect are moving to the 5x Immersion lithography?

And could you also provide some color on 4x?

And that's for 2009.

Thank you.

Eric?

I think the difficulty of 4x is such that I would only see one company going aggressive for it and another one following six months later.

Regarding 5x, everyone.

Could you give the number around that?

How many are not at Immersion in 2008 but are going to do Immersion in 2009?

Again, some of them are in alliances so it's pretty complicated, but I could give you the names.

I don't think it's a secret to say that Micron and Nanya, Inotera as an overall alliance will have to go through this, so that's basically three players.

At this moment, you know that [Cromos] is working on going into this, so that's one.

Winbond, doing the same.

[Keymondi] is already in prototyping so will need to run production, so that's also a clear one.

And the first tier obviously being Hynix and Samsung, they're already in business.

Did I forget somebody?

No that's FLASH, DRAM, to [give] DRAM, yes.

So that means every one of them.

Thank you very much.

Next question, Mr.

Gunnar Plagge.

Please state you company names followed by your question.

Yes, hello, good afternoon, it's Gunnar Plagge at Nomura.

You talk a lot about supplier based efficiency programs, and I think you said that they helped in Q3 to achieve the 38% gross margin.

So could you talk a little bit about how these programs work and how much further scope we can expect?

Yes.

The programs are not designed around raw cost cutting and say, we need 5% less or 10% less.

It is really the coordination between the supply, our R&D group, and the factory organization, in looking at the designs, looking at the way that they source their products and see where areas are that we can take cost out.

So it's quite an extensive and integral program that we have with R&D, our system engineering group and the factory and, of course, the purchasing group.

And that is depending on what it is, whether it's a mature module, which we have for a couple of year where of course the cost reduction capabilities are not that large.

You talk about a couple of percent, 4% 5%, 6%, to a, let's say, more new technologies where we have new modules where the cost reductions can add up to between 10% and 20%.

So it is very difficult to give you an overall number.

It's a very detailed model that we have and basically covering the whole machine.

But we have had cost reduction targets over the last year of about 10% on average, which is actually something that we, of course, strive to do every year.

But, like I said, when we have new introduction, new module introductions, it is easier to get cost out than when you try to get costs out of a more mature model that we already have for years.

Same is true for the lens with Carl Zeiss.

We're almost two companies, one business.

We're almost fully integrated.

And, there, we are constantly looking for cost reduction opportunities because the lenses in our most advanced tools are close to 50% of total cost of the tool.

So it's pretty granular.

That's why difficulty to give you as a kind of higher-level overview of what it's going to be but, clearly, it's focusing on the lens and the newest modules.

Yes, thank you.

As a follow up, could you give an update on the Elpida account situation?

Yes.

In fact, it's very important that the you ask the question.

I think I forgot to mention it.

Elpida power chip, obviously, is one of the DRAM vendors who are driving into 5x.

So, as you know, they publicly are also driving shrink of 6X, but they are also ramping 5x for next year.

So, sorry for having forgotten them.

The fact that I am in Tokyo means that I can report that we're making positive progress.

As you know, we are being evaluated for 5x for Elpida power chip and the progress are positive, going in the right direction in terms of the recreation of Elpida power chip about the performance of the machine.

So at this moment I cannot do more -- say more than that, except that reporting good progress.

Okay, thank you.

Next question is from Mr.

Lee Simpson.

Please state name followed by your question.

Hi everyone, it's Lee Simpson from Jeffries.

Maybe I could just ask a quick over-arching question on pricing.

If this uncertainty that we're seeing, the downturn as you call it, is going to set to continue, if there's any sort of consolidation or alliance rationalization and we see an extra pause in CapEx, and you guys are also -- seem to be quite fixed on where your R&D levels can go to at least on just efficiency gains, how far back can ASML cut on these unusual pricing that you could do to help customers?

What's the sort of acceptable levels for you guys going into 2009?

There is no acceptable level to sell a product below the value.

So the whole game in discussion with customers, even under pressure, is to compare our value to our competitors' value.

At this moment the players, that is, Nikon, Canon and ourselves, are clearly fighting the war of specifications.

It does happen, however, that when we get into some non-critical layer we are equivalent in terms of performance.

And on non-critical we hunger, make sure to be more aggressive than in the normal situation.

So that is where I was telling you about the price pressure.

Again, we are protected by the value of the critical layer.

But it is true that we may be losing a certain numbers of points of margin, but never to a level where we would at this moment consider this absolutely ridiculous.

Yes.

It was interesting the continued use of Used Tools in Immersion.

We saw it in 2Q.

We're seeing it again in 3Q, I think with foundries taking it this time.

Is this an interesting development we should be mindful of as far as ASPs into next year?

Perhaps, again, a helping hand from ASML to its customer base?

And can you talk us through how that might continue?

Yes it is -- as you know, there are two things.

One is, the used model is a good model for ASML, has worked for a long time.

We are shipping about 10 Used Tool per quarter.

Some goes to semiconductors, some goes to a related type industry.

We make money on the refurbishment of those machines so it's an [appropriate] margin.

And we solve a problem at the customers.

Immersion has become an opportunity because you have different segments with different needs.

Very obviously, FLASH is driving Immersion to its max, DRAM next and Logic next.

So we're creating opportunities for Logic to buy, I would say, older type machines.

And we -- when that was discovered with foundry we were very happy to oblige.

Because it's not so much a machine that you lose, it's a machine that you place differently.

So, in other terms, you buy back a machine with only two to three years of usage in a FLASH environment or a DRAM environment, you replace this machine with the latest and greatest, and you refurbish and sell to the foundry business.

So it's not like you lose a machine.

In fact, you sell the big machine were it's most needed.

And that, by the way, is another system to avoid margin erosion.

Perfect.

And it's also, on the top, don't think that there's an abundance of supply on those tools.

We can do it from time to time, but it doesn't mean that everybody coming by wanting a new 1700 that the 1700 is indeed available.

Yes.

Maybe just one quick final question on the financial malaise.

We're hearing now that shipping is down 7% q on q.

Distribution could be an issue over the next months.

I just wonder if there's anything in 4Q, given line of credits and so forth amongst -- being an issue in some of the Chinese ports, albeit [at the] commodities just now, do we see any legal resilience in your documentation that makes you immune to any of these issues of line of credit and the distribution that we're seeing?

No, not specifically.

I must also say I'm a bit puzzled about your question because it's not something that is currently on our table.

It might be that we're not seeing something, but at least it hasn't been bought to my attention.

Right.

You're aware that a lot a ports are not actually shipping a lot of resource at this time, given that some people aren't accepting other companies' lines of credit.

And I'm sure -- I'm not being specific about your line of credit, for instance, but maybe the distributors that you use, for instance, is any of this --?

We don't use distributors.

We do direct sales.

So we haven't seen this.

Okay.

Is this -- when you say port, you mean airports?

Yes, well, harbors and such like.

We don't use harbors.

We fly the machines.

Okay.

All right, great, thanks.

Next question, Mr.

Didier Scemama.

Please state your company name followed by your question.

Good afternoon, gentlemen.

Thanks for taking my question.

Actually, my first question is on DRAM.

If I look at the balance sheet of Tier 3 DRAM vendors, and I know you've mentioned that some of them are going to move to the 5x node because they have to.

And if I look also at the spot, even today down another 5%, it looks like the elasticity of DRAM pricing is not helping any more.

And I think those guys -- those Tier 3 DRAM guys are pretty reliant on, basically, the spot market.

So I'm just wondering what your thoughts are on that.

And whether those guys are not going to run into so much financial difficulty that really the only way out is a major consolidation, which eventually would be good for the DRAM market but less good for ASML?

First of all, you are correct, the pressure is such on the profitability at every level that if they continue like this not all of them can survive.

I think they would all tell you this, with a question mark for you to answer as to who are the weaker ones and who are the ones who will sustain this and, in fact, benefit from this consolidation.

Is it a problem for ASML?

I would not think so.

This is a business where these companies, all of them, are extremely efficient in terms of CapEx usage.

So if you sell to one account or two accounts, if the one account is two times bigger than the two accounts [or whatever], it's the same need, it's the same amount of square meters of silicon with the same machine.

So if you're referring to customer power, customer power is as good as competitor power, I would guess.

And at this moment we'll not see a difference between the current situation, where we have, I would say, six customers in DRAM, and if we would have five or even four.

The fight between six or four, in terms of what you call customer power, is definitely the same.

It's really much depend on the fight that we have with our competitors.

And, as I said, at this moment the fight is definitely on spec, it's particularly in DRAM, by the way, much more than in any segment, so we are able to sustain this without a detrimental price war.

And I'd like to add to that, if may Didier and Eric.

It's also one of the reasons why you hear us saying that we keep a very sharp focus on the technology programs, whether it's the NXT program, whether it's the EUV program.

If there is a consolidation in the industry, then it's extremely important that we remain the most viable partner for those customers in the industry because there is still a lot of competition between the three or four or five remaining players.

And that means that they're all looking for the technology that gives them the best cost advantage.

And that means that we need to have a roadmap that we can execute that is better than anything else.

So, yes, you are right that in principle you could say if the market has fewer players they have a bigger buying power, but it's also something else playing.

Okay, and just a quick follow up on Immersion.

It's fascinating, obviously this transition, and you've given a lot of drivers for Immersion business to go up next year.

I'm just wondering if you would look at the refurbished Immersion business, today you talked about four unique bookings for the 1700s of refurbished, do you think that the incremental growth in '09 would primarily come from refurbished units?

Or do you think that you will get growth even in your 1900, 1950 new equipment?

Like I said earlier, I don't think -- don't exaggerate the supply of the 1700s.

We have shipped 50 1700s in total.

Don't think that they will all come available to be shipped to other places of the world.

Some of them are being run in real anger at the customer Fabs and they will not be pulled out.

So -- and I think we will see some opportunities to basically get a supply of 1700s, Used ones, which we can then resell, but I don't think that that is going to have a major impact or that's going to be the big driver why the Immersion market in 2009 would actually grow.

I think there is an underlying growth trend, and Eric talked about that I think extensively, that is much strong than just the refurbished market.

Many thanks.

Okay.

Next question is from Mr.

Mehdi Hosseini.

Please state your company name followed by your question.

Mr.

Hosseini, please go ahead.

Yes, this is Mehdi Hosseini from FBR Capital Markets.

I have a question about your activities -- R&D activities beyond Immersion.

Can you please help us understand about the competitive landscape beyond Immersion?

And to what extent do you expect competition to rise up when we get to that point in the next several years?

This is a very nice question.

Obviously, we feel very good about your question because it gives us more opportunities.

Let me explain.

Today in the market you will see the FLASH and DRAM segment pushing very, very hard for EUV.

And you will see the Logic vehicle driven by IBM in particular, and Intel but as I would say separate company pushing for computational scaling, which is another way of saying double patterning with a significant amount of lithography, call it, lithography enabling design.

These are not conflicting.

They are trying to solve two different problems.

In the DRAM/FLASH you need very quickly your solutions, you need to shrink as much as possible and there are two or three limiters in the designs that makes computational scaling not, in fact, solving your problems, so you need EUV.

In the Logic arena you could use EUV, but EUV may arrive about six months late to the timing of some customers and, therefore, there maybe no other choice but trying to prolong Immersion but computational scaling.

So ASML is thrilled because, first of all, EUV, as I said, is a game changer.

It's all in our hand.

If we succeed to deliver six months earlier and at the cost target I talked about in my little discussion, then we have the business for two years probably for us.

We will sell on technology, that is, the fact that is shrinks, we will sell on costs because I think it's much, much more cost effective than computational scaling.

But, in addition, we think that it is possible that EUV, therefore, will miss by six months or so some good designs, and we what to hit it with computational scaling.

And this is where we are going to get a very large value added through our acquisition, Brion, which allows us to do two things.

One, selling with a service to the customer, which is how to basically scale, how to do the job of design integrated to the lithography.

And, two, that allows us to integrate onto each of the machine a certain number of features of metrology and features of programmability which we are going to sell.

So it's an additional profit generator and ASP enhancer opportunity.

So that is also part of our R&D.

We're putting a significant amount of money into this.

And we also expect to be a leader.

We have not seen yet significant activity by our competitors on this one -- on this segment also.

We're doing this, and we expect to start reaping profit of these EUV and computational scaling some time by 2010.

Excuse my lack of understanding about technology, but EUV is still requiring a mask, right?

That is correct.

Now, do you see a renaissance for mask-less technology?

Yes, there are two competing -- not two competing, two potential threat to photolithography as we are executing.

One is e-beam, electron beam, driven mainly by a small start-up called Mapper in Holland, very much a neighbor of ours and sponsored and supported by GSMC.

And there has been a press release yesterday on the subject.

And the other options would be imprint all driven by a small start-up in the US mainly and sponsored by Toshiba.

Those two technologies are interesting potential technology which we have assessed, and probably most of the market have assessed the same way as we have, for a good usage in lab environment.

I don't think the industry, and even those players, can trace a roadmap towards high volume production.

There may be an opportunity and we, at ASML, would be interested to participate and would be able to integrate because, as you know, we are integrated.

But at this moment it seems that they will -- the technology and the proof of concept does not really get to be a significant threat in, I would say, mass production.

But as a lab tool it would be a good thing.

So this why you would consider, like GSMC and Toshiba does, that it's not a bad thing to have the possibility of a lab prototyping machine which allows you to test your design early enough, and at lower price obviously, than the production machine.

But at this moment, as I say, I would look at the vote by the industry about the different technologies of the future.

And this is why I was mentioning the two big discussion is around EUV and computational scaling.

These are the two, I would say, big ones on which the industry is at this moment aligning.

Great, thank you.

Operator and ladies and gentlemen, I think we've used our allotted time so I'm going to invite those people that had a question or a remaining question to feel free to call the investor relations department here in The Netherlands this evening or this afternoon, morning your time.

And now, operator, if you'd formally close the call out for us, we'd appreciate it.

Thank you very much.

Ladies and gentlemen, this concludes the ASML 2008 third quarter results conference call.

Thank you for participating and you may disconnect.