艾司摩爾 (ASML) 2012 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by.

Welcome to the ASML 2012 third-quarter results conference call on October 17, 2012.

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.

Go ahead please, sir.

Thank you, Operator, and good afternoon and good morning, ladies and gentlemen.

This is Craig DeYoung, Vice President of Investor Relations here at ASML, and I'd like to welcome you to our investor call and webcast.

Joining me today from our headquarters here in Veldhoven, The Netherlands, is Eric Meurice, ASML's CEO; and Peter Wennink, ASML's CFO.

As the Operator mentioned, today's call subject is ASML's third-quarter 2012 results.

However, as you now know, we've also announced our intent to acquire Cymer today, and therefore, we'd be happy to answer any questions you might have on either subject, as we proceed with the call.

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

This Safe Harbor statement will apply to this call and all associated presentation materials.

Let me remind you that the length of today's call is 60 minutes.

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

Thank you, Craig.

Good afternoon, good morning.

Thank you for attending our third-quarter results conference call.

Before we begin, as usual, in this session, Peter and I would like to provide an overview and some commentary on our third-quarter results, and our view going forward, as well as providing some commentary on the Cymer acquisition proposal.

As usual, Peter will start with the review of Q3, adding comments on the short-term outlook, brief outlook on overview of the Cymer deal.

I will complete the introduction, with some further comments and update on the EUV program, and more details on the intent of the Cymer deal.

So, Peter, please.

Thank you, Eric, and welcome to everyone.

As mentioned by Eric, I will focus on a review of our third-quarter results, which are very much in line with expectations.

And I will close off with a brief overview of the other announcements we made jointly today with Cymer.

A quick look at our third-quarter sales results show us coming in at about EUR1.23 billion, just above our guided level.

This is very much in line with the previous quarter.

This quarter sales remained largely skewed towards the foundry IDM sectors, which accounted for about 70%, including non-critical KrF systems, which afforded the capacity additions.

Memory combined represented a balance of 30%.

This percentage seems high when looking at the state of the memory market, but we recognized a few leading-edge evaluation systems as sales in the third quarter that were shipped in prior quarters.

This issue also affected the memory bookings in the quarter, as these are recognized as turns business.

In addition, there is increasing uncertainty in the last few quarters as to the application for which these systems are used, which lead us to combine memory sales and bookings data [bigger than that] in our presentation materials this quarter.

The ASP of all systems recognized in Q3 was EUR25 million, which is an increase of about 10% from the previous quarter.

Service and fuel option sales remained at a healthy level of around EUR230 million.

Q3 net bookings came in at EUR830 million for 33 systems excluding EUV, with booked average selling prices at around EUR25 million versus EUR22 million in the second quarter.

The quarter's bookings profile was skewed by the turns business of the evaluation systems, as mentioned previously.

Our order backlog at the end of Q3 was EUR1.34 billion, excluding EUV, totaling 48 systems.

The backlog profile at quarter's end remained very similar to that at the end of the prior quarter.

Regarding our share buyback program, as of July 9, ASML had to suspend the current program for regulatory reasons in connection with the customer co-investment program.

And in Q4, we plan to reinitiate and complete the previously announced buyback program of EUR1.13 billion.

As to the outlook, we anticipate fourth-quarter sales coming in at about EUR1 billion, putting 2012 annual revenues at about EUR4.7 billion.

A gross margin of about 41% is expected from Q4 sales.

R&D and SG&A expenses will be about EUR155 million for R&D and EUR64 million for SG&A.

The increase in R&D as a result of the initial ramp of the 450 millimeter program for which the co-investment program has been initiated.

Customer funding of this program will start, by the way, in 2013.

The increase in SG&A is due to a one-time Dutch austerity taxation for the year 2012 on high income individuals, which taxation is fully payable by the Company.

This charge will be booked fully in the fourth quarter.

We currently recognize our customers' uncertainty regarding underlying semiconductor demand in the tablet and smartphone segments as well, as the fact that the PC sector is not yet accelerated by Windows 8 and the ultrabook form factor.

The uncertainty around broad-based semiconductor demand, which is especially relevant in the memory sector where spending is expected to remain subdued for the next two quarters, is currently limiting our view of 2013, rendering us unable to be specific with a reasonable level of certainty about the coming quarters at this time.

We do see, however, sustained demand for the logic sector, as the planned 28 nanometer node strategic buildup to a worldwide capacity of about 300,000 wafer starts per month is expected by the middle of next year, mid-2013.

And as the 22 nanometer logic ramp will start in the second half of next year.

The reason for the strength of the logic business is due to the accelerated ramp times of the current advanced technologies, and to a reduced time period between the transition from 28 nanometers to 22 nanometer nodes.

As previously mentioned, this transition is very lithography-intensive, with a significant increase in immersion tool usage versus the 28 nanometer node.

In addition, we'll have the impact of market uncertainties.

We anticipate the shipments of our first 11 NXE 3300B EUV systems that will help prepare our customers for the insertion of EUV in volume production lines by 2014.

This will contribute approximately EUR700 million in revenues for next year.

Turning to the subject of our offer to acquire Cymer.

As per this morning's press release, we have entered into a definitive agreement to acquire all outstanding shares of San Diego by Cymer Incorporated.

The transaction, which was unanimously approved by the Board of Directors of ASML and Cymer, would entitle each Cymer shareholder to receive $20 in cash and a fixed ratio of 1.1502 ASML ordinary shares per Cymer shares.

The issued ASML shares will be New York shares and will be listed on NASDAQ.

As it will take some time to harvest most of the obvious synergies resulting from this merger, we believe that the first 12 months after the closing of the transaction will be somewhat dilutive to a maximum level of 5% of the earnings per share.

But 24 months after the closing, we should see the full benefits of the synergies, and that should account for up to 5% accretion of the earnings per share.

These assumptions, by the way, do not yet include the impact of the accounting of the purchase price allocations, of which the impact needs to be determined after closing.

And we expect the transaction to close in the course of 2013.

And with that, I would like to turn it back to -- over to you, Eric.

Thank you, Peter.

Let me, in fact, focus my comments on the planned acquisition of Cymer.

This move is designed to achieve, in fact, three things.

First thing is to derisk and accelerate the extreme ultraviolet, the UV technology roadmap.

Second thing is to give ASML the positive impact of a steady and growing service business.

And the third facet is to improve our financials through the benefit of vertical integration and of synergies.

So, let's go back and discuss these three points.

Regarding the objective of overall derisking and accelerating UV development, as well as creating energy synergies, we believe that this merger will enable significant execution performance improvement.

First, it will help taking duplication out of development.

Responsibility for each key module like the vessel, the CO2 laser, the seed laser, control software, algorithm, et cetera, et cetera, will be assigned to only one site without fear of IP know-how transfer control, with no need to duplicate or to distribute development ownership in a non-optimized fashion.

The development will also benefit from the flexibility of a larger pool of dedicated talents, dedicated experts.

Secondly, reducing the complexity of manufacturing and test flows, module will be manufactured in numbers of centers of excellence -- of course, in San Diego, Veldhoven, and at our key suppliers.

They will then be integrated and qualify as a system near the scanner or at the customer site.

This will save the cost and time of assembly and testing at the different locations.

Regarding the objective of improving our financial prospects, of course, part of it comes with the synergies and efficiencies I just talked about.

But also, by integrating such a high value component of the total machine, we will certainly increase our gross margin mathematically, while we will scale R&D and SG&A structure through the synergies, thus creating an EPS improvement opportunity.

Regarding the objective of getting access to a growing service business, we are very much encouraged that the source business, as per nature, includes a significant consumerable content like the other oil for printers, which is even more significant for the UV technology.

And this will ensure growing profit opportunity on a somewhat less cycle-dependent business.

This merger is also well-timed.

We now expect steady progress towards 2014 production goal of achieving 69 wafers per hour, a goal which we disclosed, I remember, about a year ago.

On the overall scanner system from the scanner system not the source, the progress has been excellent.

Customers have now exposed more than 23,000 wafers on the six NXE 3100 machines, which are currently installed.

Regarding the 11 NXE 3300 system, which is a new generation, being prepared for shipment for 2013, the platform has already shown exceptional performance, demonstrating imaging level of 16 nanometers, direct imaging of 16 nanometer, and overlay of significantly less than 2 nanometers or extremely good performance, confirming that the system itself is ready for high time.

On the source front, the progress has been made first in stabilizing the source utilization rate in the field and the sources themselves can now reliably expose.

Second, we have now been able to prove 30 watt exposure power, equivalent to about 18 or 20 wafers per hour on an NXE C3300 on a sustainable basis in different lab experiments at Cymer and ASML.

We have -- we would have hoped to have shown about 40 wafers per hour, not 18 -- 20 equivalent by now, but it was more difficult than expected to obtain beam control stability during the summer as we worked on.

The achieved 30 watt plateau, however, is a very good basis from which to move up.

Hopefully, stably towards our 2014 goal of 105 watts or 69 wafers per hour equivalent, as I mentioned at the beginning.

Last quarter, I also mentioned that we had received customer commitments to purchase four additional NXE 3300 for delivery in 2014, and we're now expecting another 4 to 8 commitments to be received in the next six months from multiple customers.

And as the industry prepares for first semiconductor device production on these UV systems by 2014, as again, we planned since about a year.

One last word regarding the status of our customer co--investment program.

As you know, we received shareholders approval for execution of this program in September.

This program allows investment by Intel, TSMC, and Samsung, in 23% of our equity, and provides also for the commitment of these three customers of EUR1.4 billion -- nearly EUR1.4 billion of R&D funding over the next five years.

The share issuance to those customers and the related synthetic share buyback, which avoids any dilution, will be completed in Q4 as planned.

The proposed merger with Cymer is very, very much in keeping with the objective of this customer co-investment program, as it enables the next step in technology, which was an objective of the co-investment program.

The merger itself will also be significantly facilitated by the customer R&D financing pledge, the EUR1.38 billion I talked about, and by the customer backing of our equity through the fact that they will own soon 23% of our shares.

So, with that, Peter and I are pleased to take your questions.

Thank you, Eric.

Ladies and gentlemen, the Operator will instruct you momentarily on the protocol for the Q&A session; but as always -- and I know there's a lot of questions about both the results and the Cymer acquisition details, but I would like to ask you, beg you, to kindly limit yourself to one question with one short follow-up, so we can get as many in.

And I'm sure we'll have the opportunity to answer a vast majority of your questions.

So, now, Operator, if we could have your instructions and then the first question, please.

(Operator Instructions) The first question comes from Mr. Didier Scemama.

Please state your company name, followed by your question.

Yes, it's Merrill Lynch.

Good afternoon, gentlemen.

Thanks for taking my question.

What I would like to understand, and I think you've sort of alluded to it earlier on, Eric, is -- you've been working with Cymer for some time on solving those light source issues.

So can you just try to be a bit more explicit as to why you think that bringing Cymer in-house is going to help you reach the 60 or 70 wafer per hour throughput you've got as a target?

Thanks.

We think that we would have progressed to a result anyway with the current system of very deep cooperation.

But we also think that being merged will allow it to do this faster and with less risk, and with less potential conflict.

It is clear that this is a very difficult project.

If one of us would -- one of the two partners would see things differently, like one of us would protect IP, one of us would try to do something in parallel, et cetera, you can create a risk of a fight, a risk of economic negotiation, which would be of no value for both.

And also the human factor, to get engineers to work together, it's always better to put them into the same aquarium as opposed to different aquariums.

So it's an efficiency question.

It's also a business question.

We are reaching the point where being together will reduce the risk of not achieving.

And we felt, on both sides of the companies, that this was the right time to do so.

Great.

And then, Peter, just quickly did you say that the maximum dilution is going to be 5% in year one?

Yes.

That's what we currently expect.

That is to be year one after the closing.

And so after the closing, the next 12 months after the closing.

Because it will take some time to harvest the benefits, as Eric alluded to in his opening speech.

It will take some time to fully get those benefits onboard.

So we believe that after 24 months after the closing, that will be the case.

So, then you will see a 5% accretion.

Thank you.

[Amit Ahamdani].

I'm [Huduj Anani] from Citigroup.

Thanks for taking my questions.

A main one and a follow-up, if I may.

My first main question is, could you please talk about CapEx trends in each of your end markets?

And particularly if you're seeing any early signs of CapEx lines at your customers being impacted by potential changes in manufacturing strategy by one of their customers?

And as a follow-up, we have heard this talk about some of these technological factors such as DLC and 3D stacking, impacting on NAND demands.

Could you maybe share your thoughts on that?

Thank you.

Well, the CapEx trend, I think Peter was clear in mentioning that, at this moment, the logic business is clearly on a roll.

There is indeed discussion within the market about where the different players are manufacturing the project.

And nobody knows that is a discussion where Apple would have their different logic chip manufacturers, et cetera, et cetera.

All this is happening, however.

We see a steady request for our machine for capacity, because 28 nanometer has not yet reached the level of capacity required necessary for just basic demand of these products.

That will go up to June and we think these are strategic buys.

They won't move with markets short-term fluctuations.

And we have seen that 22 nanometer, which goes immediately after 28 nanometer, is in fact accelerated because you -- there is a war out there between the different architectures, arm versus microprocessor architectures, which pushes everybody to be even more aggressive in the logic arena in their shrink strategy.

So we're going to have a significant demand on 22 nanometer second half of the year.

And as I think we mentioned now for two or three sessions in a row, 22 nanometers is a huge load in terms of numbers of lithography layers.

In fact, I mentioned last quarter that we think the multiplier is 1.7 -- 1.7 times more machines in 22 nanometer than 28 nanometer would require for the same amount of wafers.

We think now is going towards [2] -- 1.722.

So there is good inflation here for lithography.

The trend, however, in memories is clearly -- in fact, there is no trend at this moment; it's a dead duck business.

There is no trend.

But that doesn't mean it won't come back and this is a key message we have to give.

We don't want to say that we know something that the business in 2013 will not be sustained in the memory business.

No, I think we are saying that there is absolutely no statistics as to where the PC business will be.

And without any clarity on that number, I could understand our customers waiting to see how much DRAM capacity needs to be built, and how much NAND capacity needs to be built.

Because, as you know, the new PC form factor, which is basically a tablet with a keyboard and, in fact, more NAND and more DRAM than the tablet -- that thing is going to be a huge factor to the capacity requirements.

So there's a key question mark as to exactly what the number is and when is it going to start?

We will all feel probably the data when Windows 8 starts.

And Windows 8, as we know, is a enabler to these new form factors, and it will have this time a significant impact on PC trends.

So we have to wait, and I don't think at this moment any one of us would be able to say there is a negative trend.

There's just no trend, no position at this moment.

If you talk about the 3D structures and things, yes, that has always been the case.

You have numbers of opportunity for 3D structures, which means more traditional lithography and less state-of-the-art lithography.

But as we often say, where we simulate this impact on lithography is plus or minus 10% of the NAND business, so it's not something that you could see easily compared to other parameters.

That's very helpful, thank you.

Gareth Jenkins.

Yes, it's UBS.

I have two very quick ones, if I could.

I just wanted on the fourth weight new EUV tool orders, could you just confirm that that's coming from the same end market as your prior one or the same customer, I should say?

Or is it completely new ones?

And secondly, I just wondered whether you could talk through how the valuation of your bid for Cymer was constructed.

Did you look at [invested] capital?

Or did you just look at the strategic value?

Did you look at accretion?

What were the sort of metrics behind the valuation?

Thank you.

So, Peter is going to handle the valuation question regarding the EUV orders; DRAM is taking the first X units but logic is getting heated.

Logic UV 2014 is going to be a fairly huge R&D content and then 2015 will be the ramp.

So now in terms of the first, say, four units or so will be DRAM, and after that, we go full-blown into logic.

Peter, for the valuation of --?

Yes.

For the valuation, this was pretty much based on the fundamental analysis of the business.

We, of course, work in the same business so we understand the business of Cymer quite well.

They understand our business also very well.

As you know, with every scanner, there's one DPV light source sold and with one EUV scanner, it will be one EUV source.

So there was a lot of insight on both sides on, let's say, the fundamentals of the market that we are currently operating in.

We basically looked at it from two sides.

You have the EUV business, which is a business that is surrounded with, like Eric said, still there's a lot of execution issues.

You could call it execution risks, but there are also executional opportunities.

But that is less straightforward than the deep UV light source business with a very large surface component.

So, a significant part of the fundamental valuation can be attributed to the deep UV, where, of course, there's also a potential to see a significant part of that valuation for the EUV business.

But there are still a lot of things need to happen.

And that, of course, could be realized when we put the two companies together for the reasons that Eric just talked about.

Now and after having done that fundamental analysis, we, of course, also look at where the share prices of the two companies were over the last three to six months.

This is not a cash deal.

We believe that it's largely a deal that will be paid in shares of ASML, which makes it clear this is, from our point of view, very much a strategic deal.

So we also looked at the, let's say, three and six months premium-based.

So it would take the fundamental analysis and the valuation, what kind of premium would that in the end represent?

We came between 50% to 60%.

If you take 50% on the six months on a three-month basis, and looking at precedent transactions for strategic acquisitions, that falls right in the middle of those precedent transactions.

So from both sides, I think from a premium analysis point of view, but very much from a fundamental evaluation point of view, that's how we came to where we are today.

And I think as I mentioned briefly, to add to Peter, is we also realized on both sides, I guess that there was negative value if we didn't do it.

Yes.

That's correct.

So at the end of the day, we needed to resolve the negative value.

Just as a follow-up, I mean, you're -- when do you best guess do you think the deal closes?

And I guess is that related to anti-trust procedures that I guess will occur?

Yes, we have to go through a few closing conditions.

One of them is, of course, the shareholder approval of the Cymer shareholders.

And we have to go through the regular regulatory approvals to our significant one, will be on the anti-trust front.

And the other will be a notification to the City's Committee in the United States.

Now, those have -- all you can say in every jurisdiction almost pre-described timetables, but we would think that under normal conditions, we could have obtained those regulatory approvals within six months, also taking into account the regulatory waiting and review periods.

But to be on the safe side, you never know when these things -- we say at six to nine months, but of course, we would be focusing on six months.

But as a nine months as a kind of worst case scenario.

Mahesh Sanganeria.

The company name is RBC Capital Markets.

Just staying on the anti-trust topic, how does your relationship change with the other source supplier like Gigaphoton?

Does that become part of a anti-trust issue?

No, in fact, I think this is the point that we are comfortable that there is no anti-trust issue at all.

Two facets to your questions.

One is, we do not choose who the source supplier is.

Our customers do.

So, for instance, Samsung or Intel or Toshiba or TSMC, would ask us which supplier they would want us to have, and then we qualify what they are requiring.

So, the future market share of Gigaphoton and/or Cymer depends, in fact, on again, their choice, not ours.

So we feel that this is a very important point to the authorities.

And the second part is, indeed we want our division Cymer to be able to be independent in their approach to customers, which also includes our own competitors, and therefore, they will be managed fairly independently.

Okay.

Then I have a follow-up on the power supply, the EUV power supply state status.

You mentioned that you are very comfortable with the 30 watt operating as an independent, and you have not integrated.

But what's -- how long can you -- how long in the terms of time can you operate that 30 watt?

And how much is a downtime after you have the issue with the 30 watt?

I'm just trying to get a sense of where are you in terms of the shipment of the 3300s with the respect to the EUV source.

Yes, this is a good question.

The 30 watt position that I made was a laboratory experiment, which is good enough because we tried it for numbers of hours in different situations.

So we're comfortable that it is replicable into the machine.

However, we need to ship the machine with a new source to be qualifying the performance in the field in real situation.

So that will happen sometime in the first or second quarter of 2013, but that issue is of much less importance.

It has logistic problem, yes.

Are we going to make it exactly on time, et cetera.

But the fact that we have now proven that we can do it is just a question of execution; it's not a question of risk of not achieving.

So we're not very worried about the issue of execution, potential execution, logistic delays.

This is not as important as proving that we can do it.

And then I'd like to add that throughput and let's say wafers per hour is not a delivery specification.

There's many others but this is not one of them.

Okay, thank you.

Sandeep Deshpande.

A couple of questions from me.

Firstly, on the Cymer acquisition, Cymer has other businesses and services, I mean, unrelated to NCD, et cetera.

I mean, does ASML intend to stay in these other businesses going forward?

And how will they service the businesses which they don't have direct equipment exposure to?

And then secondly, within the ASML business itself, I mean, one different thing that we're seeing is that NAND prices seem to be going up at a time when the orders in memory are pretty weak.

So, Eric, do you believe that NAND orders are likely to be coming in, in a couple of quarters, or is there a clear disconnect between memory pricing now and the order intake?

Yes, so very good, so on your question of Cymer as a LCD small business, we will wait to hear from Cymer's decision about whether they want to invest more in this class or stay constant.

We definitively would plan to have this group, this division as again an independent division and they would have to justify the business on their own.

So at this moment clearly there would be no decision on our side.

We will await for clues to have a proposal made by the current manager -- management of Cymer.

Regarding the NAND orders, yes, we are seeing -- in fact our famous simulation system says that we should start seeing some orders but we don't yet.

And the real point at this moment is that in a production environment, you can always squeeze a bit your environment, your production to get more out.

So we see this is happening but not yet to a level where we would get orders, because again, nobody really can understand the capacity required by the PC environment.

So this is -- hasn't triggered yet.

Okay, thank you.

Sachin Shah.

Thanks for taking my question.

Tullett Prebon.

I wanted to find out the stock consideration.

So are these shares in the US the ADRs?

What actual -- there's a couple of listings of them.

So is it ASML, US or UW?

Which actual shares are you going to be getting?

Well, it's the issued shares will be New York shares.

No ADRs, and will be -- those are listed on NASDAQ if you talk about the -- if you want to ask the abbreviation, just look at the -- one of the people here to just find you the right abbreviation.

But it will be New York shares as we are currently listed.

There are a couple of New York shares listed.

So, would you happen to know offhand what specifically it is?

There's UW (multiple speakers) and then US --?

Well, it's -- why don't you -- sorry?

It's US?

It is ASML US.

Okay.

Okay.

And on the regulatory side, I just wanted to go over that with you.

CPS is required.

US HSR is required.

And I think Germany is also required.

Can you maybe just go over some of the other regulatory approvals and just to confirm and walk through, there are no concerns that you are expecting?

Well, currently you are right.

We don't expect concerns at this moment.

And it's good to know that you've done the investigation into the anti-trust filing.

This has to be in Germany because I don't know where we have come to that conclusion.

But at least we would like to -- perhaps you can call our Investor Relations to share your views on that particular point.

But we are going through the areas and the territories where we do business.

In every business, there are different thresholds of business that we need to achieve, which will trigger a regulatory filing or not.

So that is the process we are currently going through.

We haven't identify anything that causes us any concern.

And so just to confirm, the election of -- the stock component is not an election.

It's fixed.

It's already listed on the New York Stock Exchange.

Yes.

Okay, perfect, thank you.

Andrew Gardiner.

Thank you very much, it's Barclays.

My question around -- again, around the Cymer transaction, I'm just wondering if there are any sort of restrictions in your ability to continue cooperating and working so closely with Cymer, in terms of the development of the EUV source, now that you've entered into an agreement to acquire the company?

Are there any issues in terms of making sure you stay at arm's length from each other, or have you got enough, I suppose, processes in place already that you can continue the development as is?

I'm just thinking about concerns around the timeline of the EUV development.

That's all.

And also, in relation to the acquisition, how did you come to a decision around the stock versus cash split?

I will let Peter discuss stock versus cash.

The restructuring to operate, no, we're not very worried.

Of course, we understand a jump the gun situation.

We have to continue our relationship versus as a separate entity, and we are planning to do so.

But remember, this is a relationship of customer supplier, so it is natural that the supplier does some work with the customer and vice versa.

We have already [X] years of cooperation.

So all this is just the same type of behavior potentially improved, but not because of the merger; it's because we are in a situation of cooperation.

I believe the law would be a concern to us if there was a risk that either of the two companies would reduce or lose members of rights or assets during this transition period.

That's not the case.

We will be sure that Cymer's assets are respected and protected.

Same for us.

But again, we don't think there is any significant restriction of operation that would be of concern to the law.

Yes, with respect to your question, how do we come to the consideration stock and cash, basically when we looked at it, we said we were both companies, like I said, earlier work in the same segment.

We also believe after I talked to Cymer, and Cymer listened talking to us, we identified that in the shareholder base of the two companies were very much a focus of the shareholders on similar issues, like the success of the EUV program going forward, and the value that could generate to the Company going forward.

So actually it appeared that both shareholders basis institutional investors focused on the same semiconductor industry, especially equipment industry, with an understanding of where we're going with our technology choices.

That made it for us clear that that deal on the basis in our stock would actually be the most logical, because the institutional investor closure would actually stay, in fact, the same with regards to the industry that they are investing in.

One of the requests was that, of course, cash is always an objective feature, but that should not be the, let's say, the main element of this consideration.

We wanted to do that, but basically the majority we wanted to be in stock to reflect the strategic nature of this transaction.

And the cash was there too as an additional request and we were fine to do that.

Okay, thanks very much.

Janardan Menon.

It's Liberum Capital.

Just two questions.

One is on the boundary side on the logic side.

I'm just wondering, given that capacity of 28 nanometers has been ramped up as well about 200,000 wafers this year, and then the 300,000 wafers by middle of next year, plus the 22 nanometer ramp in the second half of next year, if you just compare those two ramps without taking into consideration any other factors like what the demand may be, et cetera, what would the foundry CapEx spending overall this year versus next year?

Would that be a flattish trend?

Would that be up or would that be down, just based on the up to [EUR200,000] this year and then [300,000].

Of course, I agree that the slightly unknown factor is the extent of the 22 nanometer ramp next year but I presume you have some idea there.

And the second question is on the acquisition of Cymer.

Does that preclude Michio and Gigaphoton technology in your EUV roadmap?

And if, by some chance, if that is not the case, and if one of them is successful, then where does that leave your involvement with Cymer by the end of the day?

Yes, I think it's about flat.

We expect about a flat logic business for us in lithography between 2012 and 2013, on the assumption that the 22 nanometer.

And I think, frankly, Craig would let you know for sure.

But I think it's [40,000] away for capacity, which will be planned.

And so on that assumption which is realistic, we would expect a flat logic but if it is more than [40,000] or 22 nanometers, then that could be because it could be additional customers coming in, in 22 nanometers.

Because the current plan that we have is very limited numbers of customers would go into 22 nanometer.

So if there is another one, then it would be an upside.

Okay?

The Gigaphoton issue in UV.

So like we have discussed with the Gigaphoton subject, we think that Gigaphoton business model has been to get into EUV when the EUV is more mature to get to this business as a follower.

It is our understanding that they are still expecting to do so.

And therefore, when they get into the business, they will then benefit from what I explained before, which was equal opportunity or based on the decisions made by the earned customer.

So in other terms, in this business, it's obvious that we did not expect this before the merger plan that the two would come early.

We always expected Giga to be a follower.

And this new concept doesn't change that.

We are going to make sure that our division with Cymer is indeed going to get this UV machine early enough and cost-effectively.

And then, [one], two, three, four years, whatever detail will come in when EUV is, in fact, is a big business, in which case everybody will be happy.

Okay.

Got it.

Thank you very much.

Stephane Houri.

Stephane Houri, Natixis.

I have a quick question, if I may.

I was looking at your statement about UV for next year, and the fact that it would represent around [EUR700 million], which makes an ASP of around [63].

Did you have to grant some discount because the throughput is not at the level expected?

Or is there an element which is missing for me?

There is no discount there, because the throughput, like I said, is not a delivery specification.

But we will have some accounting deferrals on that.

And there's also one of those systems is going to be an R&D system, which is going to be booked in as a credit into the R&D line.

But we will be a bit of accounting accrual as we normally have with some -- but it is largely the fact that one of those systems is not in the sales line but is in the R&D line.

And where should we put the first delivery will be in Q1 or Q2?

We are planning for Q1.

Q1, okay, thank you very much.

Mehdi Hosseini.

My first question has to do with Cymer.

I'm just curious to hear your thoughts as to what was the incremental catalyst for this transaction to happen?

Over the past several years, there have been a number of challenges in commercializing EUV.

As a matter of fact, ASML has put in place your own technologists at Cymer site to help with the migration or commercialization of EUV.

But what happened over the past couple of months that pushed you over the edge and caused this transaction to happen?

And then I have a follow-up.

Yes, this is a very good question.

So, of course, in the mind of a manager, there is always two things.

One is a market timing.

Do we buy the wrong place, wrong time?

And the other one, which is, in fact, much more important is, does it make sense businesswise?

So we, of course, looked at both.

Market timing issues and valuation and stuff, which we unfortunately had to go through sometimes we also -- we've looked at times where our share versus Cymer shares went up and down, et cetera.

And you will notice that if you do this, the ratio of share to share today is not a bad time versus a year ago versus two years ago.

So we did manage this, so that there is no bad timing versus valuation.

It's never perfect timing, but we thought we think that today is really not ridiculous.

Being, however, the master decision has been the business.

And yes, I have to be honest with you.

There are two models.

One is a model that says we have -- we are out of the source business, and we benefit from that as we benefit, for instance, to be out of the lens business.

Zeiss is out.

And we benefit from that.

There is a good P&L compatibility but sometimes they have bad time, we have bad time, et cetera.

And all this is well-organized and we also benefit from being separate.

In the case of source business, we saw that there could be some times where it's better to be out or it's better to benefit from a fight between Michio Ushio, Gigaphoton and Cymer.

It's better to also have them have different cycle than we have, et cetera, et cetera.

And then we also looked at the positive of moving together, which was access to a bigger part of the pie, because EUV source is going to be a bigger part of the pie of the total system than ever.

So today is no opportunity for us to get in.

Second is, we realize, as I explained at the beginning, that there were so much synergy of efficiency in view of our going further the more we went into the development, the more we discovered the complexity, and more we discovered the need, in fact, to merge the way how to do it.

So there was a learning slowly but surely that we gained a lot of synergies by, in fact, merging.

So on the one hand we discussed all the time here whether it's good or bad to be in that business.

And then we were forced to get into this business when we discovered that there was positive synergy to get in and there was negative synergy not to do it.

So in other terms today, I will go further, but if we were not to do it, we were at a point where we may have had to duplicate efforts to develop a number of parallel stuff, which would have not been in the interest of Cymer, nor would it have been, at the end of the day, it cost interest for us, therefore, from the industry.

So, when we made this decision, it was a natural thing that happened during the last nine months or so.

And I have heard or read things in the newspaper that we've been influenced by customers and things.

Honestly, the customers are supporting us; indeed, there is no question that that also will be helpful with the anti-trust.

We're going to be hugely supported by customers.

But really what happened is that we discovered that the best model for us for Cymer and for the industry is that we find the solutions together.

Yes, it was more than natural evolution.

That is what Eric was saying.

And finding out step-by-step that this wasn't the best solution.

This doesn't happen from one day to the other.

This is a slow, more like an evolutionary process.

Thanks for detailed color.

Just one quick follow-up, rather clarification.

You did say that you think that there would be 300,000 wafers per month additional or incremental logic capacity added by the middle of next year.

Does that mean that the logic related bookings could decline in early 2013?

Or am I missing something here?

No, I don't think.

So first of all, by logic, you know when you mentioned the numbers of wafers per month, we do that because the logic business goes towards a ramp-up from zero to some level which is either 200,000 wafers or 300,000 wafers.

We have done this for 15 years or something.

Every node to get to this.

But the old node remains in action.

So in other terms, an engine controller on 65 nanometer is still there.

So 65 nanometer, I think it's a [300,000] wafer per month and it will remain there.

So now you've got a new business called 28 nanometer, which is a new chip.

And in this case, as you know these chips are basically the ones that are used in mobile stuff, and that will reach a level of 300,000 wafers.

And if I'm not mistaken, [40] nanometer will only be [200] nanometer -- [40] nanometer (multiple speakers) [200,000 -- 240] nanometer is a small node.

28 nanometer is a big node at [300] nanometer.

And that will last until the middle of next year.

That's what I say -- to build up.

Yes, to build up and then that thing will stay in production for 100 years afterwards.

Now, the bookings 100 years is poetic license.

The bookings, however, will -- yes, we still need some bookings to make that up to June, but then there will be raising bookings when the 22 nanometer POPs up and that's not yet booked for us, and when demand and DRAM comes back.

So we do not know when that pickup is going to happen.

It's an obvious pickup.

It is going to happen.

Will it happen in Q4?

Will it happen in Q1?

We don't know.

Got it, thank you.

Simon Schafer.

It's Goldman Sachs.

Actually, I just wanted to go back on my question of sort of trying to get a sense as to where you think cross-cycle profitability now is?

The premise of the question is that you clearly mentioned that the proposed assignment transaction is margin accretive at the gross margin level.

I think that's obvious from a multiyear perspective.

And knowing the financials then, I think you've obviously also referred to the likelihood that the customer investment program is gross margin accretive, just because you do need to get compensation for the increased R&D level.

So my question is, cross-cycle, historically, you have, I guess, in the last five or six years, you were looking at a 30% to maybe 45% type gross margin range.

Going forward, if and when the cycle may recover, what sort of margin uplift do you think you're looking at?

And if you could split that between the customer investment program and the proposed Cymer deal, that would be very, very helpful.

Thank you.

Yes, I think you are right.

And Eric, you know, alluded to that going forward, especially when EUV kicks in, there's going to be a significant component in cost of goods, that is now going to add to the gross margin, which is clear because we don't need to buy it in.

So that is going to in the EUV area, especially in 2015 and '16, when EUV will start to ramp, will have a big impact on the gross margin, that is for sure.

And we're talking about the big impact, you could think about, let's say, a gross margins anywhere between 45% and 50%, so they will move towards that range.

Now, what the impact will be at that same time of the operating margin, that has to do with how quickly.

And I mentioned that how quickly we can get to the synergies that we see, basically some more efficient R&D, it's simplifying the manufacturing processes; simplifying the sourcing strategies.

You have to realize that a big part of the sourcing chain for Cymer EUV is similar to what the ASML sourcing chain.

So there's a lot of opportunity for also the synergies there.

Now, we made a preliminary calculation.

The only thing I say, if we put the two companies together, and you look at what we have now identified as the synergies that we can achieve, we say, well, after 24 months after closing, I'd say closing happened somewhere in 2013.

So it's at the end of 2015 area you would see, on the base of the combination of the two companies, is about a 5% accretion of the earnings per share.

I think -- but if you now ask me, Peter, what does that mean in terms of your gross margin and your SG&A, I have an idea, but I think it is not opportune right now to give you all that detail, because there is still a couple of things that are open.

I just want to give you, let's say, an impact on the EPS the way we see it today and I'd like to keep it there.

But a couple of things happen.

Gross margin will go up and also there will be more efficiency on the operating expense line.

Makes sense.

But I guess just to clarify the five points that you called out that includes both Cymer and the customer investment program.

Yes.

Got it, okay.

Thank you very much.

All right, ladies and gentlemen, I'm afraid we've run out of time.

If you were unable to get through with a question on this particular call, please feel free to contact the ASML guys here in Veldhoven and in the US.

We do ask that you, as you can imagine, there is a extraordinary number of calls that we're handling this evening here in the Netherlands, and this afternoon in the US, so if you can just be patient and bear with us a bit, we would really appreciate that.

So now, on behalf of ASML's Board of Management, I'd like to thank you for joining us on today's call.

And Operator, if you would formerly conclude the call for us, we would appreciate that, thank you.

Of course Mr. DeYoung.

Ladies and gentlemen, this concludes the ASML 2012 third-quarter results conference call.

Thank you for participating.

You may now disconnect.