科林研發 (LRCX) 2011 Q3 法說會逐字稿

Good afternoon, ladies and gentlemen.

Thank you for standing by.

Welcome to the Lam Research Corporation March 2011 quarterly results conference call.

During today's presentation, all parties will be in a listen-only mode.

Following the presentation, the conference will be opened for questions.

(Operator Instructions).

I would now like to turn the conference over to Shanye Hudson.

Please go ahead.

Thank you, Brandy.

Good afternoon, everyone, and welcome to Lam Research Corporation's quarterly conference call.

With me here today are Steve Newberry, Chief Executive Officer and Vice Chairman of the Board, and Ernie Maddock, Senior Vice President and Chief Financial Officer.

Shortly, Ernie will discuss financial results for the March 2011 quarter.

Steve will then share Lam's business outlook for the June 2011 quarter before opening up the call for Q&A.

The press release detailing our financial results was distributed by Business Wire shortly after 1 PM this afternoon, and is available on our website at lamresearch.com.

Today's call contains certain forward-looking statements, including those related to our forecasts of market share, shipments, revenues, expenses, margins, earnings per share, cash generation, and free cash flow, as well as other statements of the Company's expectations, beliefs, and plans.

There are important factors that could cause actual results to differ materially from those described in these forward-looking statements, and a list of these factors can be found in the slide package accompanying this conference call and on our most recent Form 10-K filed with the Securities and Exchange Commission.

All forward-looking statements are based on current information, and the Company assumes no obligation to update any of them.

This call is scheduled to last until 3 PM, and we ask that you please limit questions to one per firm with a brief follow-up.

With that, I will turn the call over to Ernie.

Thank you, Shanye.

I'm pleased to report very good overall performance for the March quarter, resulting in earnings per share above the high end of our guidance.

For the quarter, shipments were within our guidance range, at $813 million, a sequential decrease of 9%.

Application and market segment breakdown for the quarter were as follows.

Applications at 65-nanometer and below represented 92% of overall system shipments, and 79% of overall system shipments were for applications at 45-nanometer and below.

System shipments for NAND were 23% of overall system shipments, followed by DRAM at 15% and other memory at 4%, making the total memory segment 42% of overall system shipments.

Logic and other accounted for 33% of overall system shipments.

And, the remaining 25% were to foundry customers.

Revenues for the March quarter were above the mid-point of our guidance range, at approximately $809 million, down 7% sequentially.

Ongoing gross margin was near the high end of our guidance range, at 46.2%, and was down from 46.8% in the December quarter, due to reduced factory absorption and a less favorable customer mix.

Consistent with the plans we shared with you last quarter, we've increased our investment in customer-facing productivity enhancement and research and development activities.

These customer-facing activities are supported with selected incremental investments in core R&D activities, and were the primary drivers of the sequential operating expense increase of $11 million, bringing total OpEx to $177 million.

As we discussed last quarter, we expect another quarter or so of incremental investment levels, and would then expect to see these investments stabilizing after that time.

We believe that these ongoing investments are important to enable to us maintain and enhance our competitiveness and achieve continued market share gains.

Ongoing operating income was $197 million, which resulted in an ongoing operating margin of 24.3%.

During the March quarter, our ongoing tax rate was 8.3%, a few percentage points lower than anticipated, primarily due to certain favorable discrete tax items.

For fiscal year 2011, we continue to expect our tax rate to be in the low teens.

And, based upon the current business outlook, anticipate a similar low-teens rate for the remainder of the calendar year.

Based on a share count of approximately 125 million shares, earnings per share was $1.45, exceeding the mid-point of our guidance range by $0.12.

Approximately half of this delta was due to the more favorable tax rate, and the remainder was the result of our stronger operating performance Moving to the balance sheet, we ended the March quarter with cash and short-term investments, including restricted cash, of $1.4 billion.

DSO for the March quarter was 72 days, flat from the December quarter, and inventory turns were 4.9, down from 5.6 days in the December quarter.

Our DSO and inventory performance both reflect the timing of customer shipments during the quarter.

Deferred revenue at the end of the quarter increased to $247 million, and excludes $36 million of shipments to Japanese customers that will revenue in future quarters.

Equity compensation expense came in at $12 million.

Depreciation and amortization was $18 million, and capital expenditures were $36 million.

We ended the quarter with about 3500 employees, and there were no share repurchases during the quarter.

During the March quarter, our cash flows from operations were $242 million, representing 30% of total revenues.

And, we are on track to deliver similar or higher levels of performance in the June quarter.

Over the last four quarters, Lam's cash flows from operations were approximately 27% of total revenues.

And, during the same period, our free cash flow was approximately 23% total revenue.

Lam's business model enables to us to generate strong cash flows while at the same time making the investments we believe are necessary to further strengthen our market position and benefit from future growth opportunities.

With that, I'll turn it over to Steve for his comments.

Thank you, Ernie, and good day, everyone.

Thank you for joining our call today.

As Ernie shared just now, our March quarter financial performance was in line with our expectations, and reflected solid operational performance across all of our business units.

Since our last call in January, political tensions around the world, rising fuel costs, higher food prices and the tragic events in Japan have all contributed to clouding the macroeconomic outlook.

It's still too early to fully understand the impact of these events on the overall global economy and the electronics industry.

However, GDP growth projections for 2011 remain in the range of 3% to 3.5%.

And, our current assessment is similar, leading us to believe that, overall, our industry fundamentals and market drivers will remain relatively strong.

The consumer electronics environment continues to be largely driven by increased adoption of mobile products, smartphones and tablet devices.

While we expect overall IC unit growth to be 10% to 12% this year, we believe that NAND unit growth should outpace overall IC unit growth rates by approximately two times, resulting in [bit] growth in the range of 85%.

This level of bit growth should translate into NAND wafer fab equipment spending in the range of approximately $8 billion, up from 2010's baseline of approximately $6 billion.

With our belief that tablet capabilities will expand beyond a content consumption orientation to include content creation, we expect tablet growth rates to remain strong over the next 18 months to 24 months, leading to sustainable and growing NAND demand during that same period.

Looking at the DRAM market, consumer PC demand in established markets is growing at a modest 3%.

But corporations are continuing to proceed with their PC refresh cycles.

The pace of this activity has remained flat with the back half of 2010, and, as a result, we have adjusted our view for calendar year 2011 total PC unit growth to approximately 11%.

This creates a DRAM bit growth forecast that's at the lower end of an estimated range of 50% to 55%, and results in a decline in wafer fab equipment spending for DRAM in calendar year 2011 to approximately $5 billion, down from the approximately $8 billion spent in 2010.

Combined, total memory wafer fab equipment spend is projected to decline by approximately $1 billion, or about 7% year-over-year.

As we take a look at the foundry segment and advanced logic, we continue to expect a strong year in terms of WFE spend.

Consistent with our expectations, these manufacturers are investing in leading-edge capacity to position themselves to meet their customers' expected needs at the advanced nodes.

Typically, these investments will occur before actual demand materializes.

As a result, we expect 2011 WFE spend for foundry and IDM advanced logic to increase to $14 billion to $15 billion versus approximately $11 billion in 2010.

As device complexity continues to grow, the cost for our customers to add incremental new capacity, or upgrade existing capacity, continues to increase with each successive technology node.

While this trend applies to all market segments, the impact is particularly pronounced in the foundry advanced logic space, where technology node shrinks are primarily accomplished by adding new capacity rather than converting and upgrading an existing process line.

A specific example of this capacity cost increase can be seen by comparing a 65-nanometer logic process with a 28-nanometer logic process.

Adoption of new process steps such as High-k metal gate and metal hard mask schemes in back-end processing, along with continued increasing layers per device, collectively contribute to etch process times for each 28-nanometer wafer, which are more than 50% greater than that of a 65-nanometer wafer.

This increase results in more etch chambers required to output the same number of wafers.

As Lam is the acknowledged and established etch market share leader in each of these areas for [foundry] and advanced logic, we are well positioned to benefit from this increase in investment levels.

The increase in process complexity also translates into the single-wafer clean market, requiring customers to transition more wafer cleaning steps from batch to single wafer processing to effectively manage defect densities and device yields.

Single-wafer clean processes represented approximately 65% of the total clean steps for a 65-nanometer foundry logic device.

At 28-nanometer, we expect that percentage to increase to 85%.

Our differentiated spin clean technology and innovative linear clean tools designed to capitalize on this trend, and we expect to benefit from incremental share gains in 2011.

Looking beyond the 28-nanometer and 22-nanometer nodes, our customers are facing unprecedented technology challenges, including 3-D device structures such as phemphat gates and vertical NAND.

These technology inflections are driving our customers to initiate major R&D activity earlier than they have typically done in the past.

Not surprisingly, Lam must parallel our customers' investments to ensure that we are strategically positioned to deliver solutions and ultimately benefit from those future market opportunities.

The investment plans that Ernie spoke of earlier are intended to do just that.

We are committed to further build upon our success by differentiating on critical leading-edge applications and leveraging our installed base position for early learning, and partnering with our customers to deliver solutions that address their most difficult manufacturing and technical challenges.

Given this set of industry challenges and opportunities, our expectation is that overall calendar year 2011 wafer fab equipment shipments will grow by 12% to 17%, with total spend between $32 billion and $34 billion.

While this represents meaningful improvement from 2010, it's important to recognize that in the consolidated industry environment in which we now operate, where 10 customers comprise approximately 85% of WFE spend, it is, and will be, increasingly common to see quarter-to-quarter shipment variability.

While being mindful of the short-term financial impacts of this variability, Lam Research will remain focused on ensuring we meet our customers' productivity and technology solution needs as they respond to the growing demand trend for advanced semiconductor devices.

With these factors in mind, our June quarter guidance is as follows.

Shipments of $780 million, plus or minus; $25 million, revenues of $745 million, plus or minus $20 million; gross margin at 45%, plus or minus 1%; operating profit at 20.5%, plus or minus 1%, and earnings per share of $1.07, plus or minus $0.07.

In summary, as we look at 2011, despite short-term push-outs from recent events, we think current prospects for the industry remain healthy.

This environment provides Lam with the opportunity to solidify and enhance the value that we bring to our customers, and we expect it to the result in a further strengthening of our market position going forward.

With that, Ernie and I will take your questions.

Thank you, sir.

We will now begin the question-and-answer session.

(Operator Instructions) Our first question comes from the line of Stephen Chin with UBS.

Hi, Steve and Ernie.

Nice [quarter].

Just a follow-up question, Steve, on the shipment guidance.

Could you share more color on the customer type that is responsible for the push-out?

Is it mostly memory customers?

And does the guidance in the June quarter assume any push-out because of the shortage of raw wafers that might be available in the industry?

Thanks.

Yes, the activity for us, as it relates to requested shipments in the June quarter, are really primarily a function of both some memory push-outs, most of it was DRAM-related.

Some of that was offset by some pull-ins of NAND.

But really, kind of the biggest net push-out effect was really related to foundry and other types of logic activities.

Obviously, when we see that kind of activity, we have extensive conversations with our customers as to what the reason is behind the pushes and what their expectation is in terms of when they're going to want those deliveries.

And I think the good news is that customers' intentions, relative to when they want them is for the most part, to take those deliveries in the September quarter, with a few of the pushes ending up in December.

But primarily they want delivery in September.

As it relates to material shortages, none of our customers have indicated that currently they're experiencing any wafer start, other types of material shortages.

I think if we think about the inventories that these customers normally carry, if we look at the public statements as well as the private comments that they make to us, I think that most customers, if not all of them, really expect that they're going to be able to work through what potential issues there may be, because most of them carry 2 to 3 months worth of wafer inventory.

And the expectation is that wafer supply will be back in good shape in that timeframe.

Next question, please.

Thank you.

Our next question comes from the line of Patrick Ho with Stifel Nicolaus.

Thanks a lot.

Nice work on the quarter.

Steve, could you just give us a quick update in terms of the application wins that you saw in the etch marketplace?

And the follow-up question is, do you feel that you are still on target for some of the shared gains or the share position for your Company for 2011 for both etch and clean?

Yes, I think we're 1 quarter into a lot of the activity that we're working with customers as it relates to those applications, that they're looking for qualification.

I think to date, we're on track with our expectations in terms of the progress that we're making in terms of the etch penetrations as well as the clean penetrations.

I think probably at SEMICON West, we'll probably provide a more comprehensive update because we'll have more information at that time.

But clearly, we had a lot of momentum, a lot of wins in the second half of 2010 that will manifest themselves into ship market share gains in 2011.

Then, of course, when we look at the makeup of wafer fab equipment spending in 2011, it is going to be materially different, given Intel's statements as recently as yesterday, or the last day or 2, that they're going to increase their capital spending even more than what they had previously announced.

And so clearly, as a function of us not participating in none of it on the etch side, that will be some market share issues that we'll have to kind of offset with the application wins and the shipments that we do in 2011.

So, we'll see how that plays out over the year.

We'll update that at SEMICON West, and as it stands right now, we're in good shape.

Next question, please.

Our next question comes from the line of Atif Malik with Morgan Stanley.

Thanks for taking my question.

Good quarter.

Steve, if the push-outs are not macro-driven or Japan earthquake or wafer shortages, or indirect shipment delays in Japan, and the memory pricing hag been going up year to date, I just want to understand what's driving those push-outs.

Is it that some of the spending that was going on was for market share reasons on the logic side, and they're not materializing?

And those customers are changing their plans?

I think that there's a little bit of activity that we saw post-the earthquake/ tsunami where I think a lot of customers kind of froze in place, so to speak.

And what they really wanted to understand was, will this or could this have an impact in terms of demand in the marketplace?

Would this or could this cause issues relative to supply chain?

And so, there's definitely some aspect of that.

But, I think we're going to move beyond that, if not already have moved beyond that.

When you talk about pricing, certainly the NAND pricing environment is very favorable right now.

And that's why what we saw in terms of what customers want in the June quarter is more NAND shipments to take advantage of the supply and demand situation that exists there.

Whereas memory, the pricing is tighter, the demand is a little bit lower than what people thought because of the consumer PC growth not being as strong.

Ultimately, what we have is the foundry logic segment, which was pretty bullish, I think, earlier in the year for a variety of reasons.

Some may be related to how quickly they can absorb a lot of what they've already taken, what are their customers doing relative to their demand profiles, what's happening as customers move from running product on 65-nanometer to run it on 40-nanometer or 45-nanometer?

What's the pace at which customers are wanting to come up and take 28-nanometer type ICs?

I think they're in kind of a period of evaluating and assessing all of that.

And so, they've kind of put some of the things on hold for a number of weeks or a couple of months.

But, I do think that as the year plays out, we're going to see that foundry logic will be very strong in the second half.

Just a follow-up for Ernie.

Ernie, assuming is revenues of $800 million exiting this year, how should we think about the OpEx exiting this year?

As we indicated, based upon the guidance Steve provided, for the OpEx that you would infer from the guidance that was provided, we'll probably see another $2 million variation in that as we move forward.

But then we expect that to level off.

Okay, thanks.

Thank you.

Our next question comes from the line of Jim Covello with Goldman Sachs.

Maybe first, if I could start out.

Steve, what's the thought process behind the buybacks at this point?

Obviously, continued real significant accumulation of cash.

Jim, I think one of the things that's important to remember is, we do have a significant amount of offshore cash.

And so, if you look, for example, at where we would be at the end of March, and you exclude the Company's restricted cash balances, you would be at 50/50 onshore/offshore.

If you include that restricted cash, you're closer to 40/60.

So, the overall cash balance that's available to the Company relative to any share repurchases in the short term might not be as great it would at first appear.

And as we look forward into the end of this year, we would expect to see 70% to 80% of the cash accumulation of the Company occur offshore.

So, I think our philosophy remains fundamentally unchanged, which is that we are open to share repurchases as we see the appropriate opportunities come along.

And when those do, you will see us take action under the authorization that we currently have available to us from the Board.

And then, maybe if I could just ask 1 follow-up on that, then I will throw in my final question.

On the buyback, would you ever consider -- I know some companies that have an offshore cash issue have issued some debt against the offshore cash and used that to buy back some stock.

Is that ever anything that you would consider?

And then, secondly, with the disruptions in Japan and some of the manufacturing facilities at a place like TEL, is that just too short term in nature for Lam to potentially pick up any share?

Or, is there potential share gain opportunities as they're unfortunately disrupted in their manufacturing facilities?

Thanks very much.

Sure, Jim.

I'll take your question about the borrowing, then turn over to Steve for comments on TEL.

We're always looking at the markets that are available to us relative to options that we have.

So, yes, we are looking at that idea of borrowing, secured by the offshore cash.

We're looking at other forms of potential liquidity for the Company that would enable us to be even better positioned to do some sort of activity, whether that be something strategic or a share repurchase.

As those circumstances we feel are favorable to the Company, we will act upon them.

Jim, I think Tokyo Electron has communicated that their new plant where they were going to produce etchers, is going to be delayed coming online.

I think the reality is that they were still in production with their etch tools in another plant, in a different location.

And I think that from that standpoint then, I think they have been able to make the deliveries that are consistent with their customers' needs.

And so, I don't expect that we will see market share gains as a function of the plant issues that they have in Japan.

Thanks very much.

Thank you.

Our next question comes from the line of Satya Kumar with Credit Suisse.

Hi, thanks.

Steve, I think earlier you had said that the first half this year was looking approximately similar to the second half last year, and June shipments could have been up.

Obviously, with the push-outs they are looking down.

But you've maintained, or perhaps even taken up, the expectation on CapEx for the year.

I guess most of that is from Intel.

Does that mean you're looking at a second half that has to be higher than the current level in the first half?

And specifically, looking to the September quarter, are you expecting that shipments are directionally up or down from here?

My comments, a couple calls ago and last call, were that I expected the first half would be relatively flat to the second half.

And with the recent activity that we've had, we're going to be, at least from the standpoint of Lam, which may be different than other companies as a function of who their customer base is, we'll probably be down in the first half approximately $100 million relative to the second half of 2010.

Then, consistent with your perspective that if the total WFE environment is somewhere between [$32 million to $34 million], the second half will have to come up.

When we look at what our customers are asking us for, in terms of second-half delivery, they are requesting a higher number of shipments in that second half, somewhere in the 10% range, potentially more, second half of '11 versus first half.

We'll have to see how that plays out.

But if the industry is going to spend $33 [million] or potentially $34 million, at least from the profile for Lam, our shipments would need to come up in the second half.

Okay, that's helpful.

Just color on September, if you could.

A follow-up on memory.

Thanks for giving the WFE numbers on NAND and DRAM.

I was wondering if you have any different perspective on the total wafer start numbers that you had previously given out, which I think for NAND was 200 new wafer starts and 400 upgrades, and DRAM, 125 new wafer and 450 upgrades.

Have those numbers changed at all?

We could talk about those a little bit.

As it relates to September, I'm not going to comment specifically, because the reality is, we are sitting here at the start of the June quarter, which has clearly changed pretty significantly from what we thought would occur when we were on this conference call 3 months ago.

And I think that consistent with my prepared comments that with such a concentrated customer base, all it really takes is 1 major customer who decides that they want to pull something in or push something out, and you can swing the shipment activity for us $50 million to $100 million.

And by the end of the year, it may not change the equation, but it could certainly change how things play out in a given quarter.

And so, I think that if you just think in terms of there was a fair amount of push-out activity, from the second quarter, and as I commented earlier, most of that push-out is residing in September.

In terms of current requests, we'll see how that plays out.

It could decide to spread out a little bit more.

But I think the important message is that as it stands right now, the year is holding consistent with what we thought, just a different distribution by quarter.

So, relative to wafer starts, we think that there's been probably more conversions, at least in DRAM, where the bit growth is a little bit slower.

And so, we think that probably somewhere around 450,000 wafer starts per month will be conversion-oriented with only 65,000 to 70,000 wafer starts per month being new wafer start capacity.

And that's important, because the conversions only cost a company somewhere between $65 million and $70 million for 10,000 wafer starts.

But if you want to put new 10,000 wafer starts, it will cost you $300 million.

In NAND, it's a little bit different story.

NAND, with the higher bit growth, can't satisfy that with just conversions.

And even though the conversions are still high, we think that they probably increased from around 400,000 conversions to about 500,000 wafer starts per month conversion, but that they're still going to have to add slightly more than 200,000 wafer starts per month of new capacity output.

And so, when you look at those ratios, that's why you kind of end up with DRAM spending about $5 billion and NAND spending is probably going to be about $8 billion.

The delta is largely driven by the fact that there's a lot more new wafer starts being added in NAND in order to meet the bit growth.

Thank you.

Next question comes from the line of C.J.

Muse of Barclays Capital.

Good afternoon.

Thank you for take my question.

First question, with the decline in shipments, the revenues guided down worse.

I guess the question is, I would have thought that perhaps deferred revenues and the Japanese revenues might have sheltered that a little bit better.

So, I guess if you can comment on that.

Then, would love to hear your thoughts on if we see that trajectory of shipments rising in the second half, how we should think revenues should track that rise.

Yes, I think, C.J., that one of the things for us, being a revenue-on-acceptance Company, is that the ability to keep shipments and revenue pretty closely in line really has to do with as you ramp up what's the linearity of that.

And as shipments ramp down, as we've seen, what's the linearity of that.

What we saw in the March quarter is that we had a non-linear distribution with a lot more systems shipping in March that didn't revenue in the March quarter.

But more importantly, when we look at our June, our June is non-linear again, with a lot more of the shipments in the last 4 to 6 weeks of the quarter, than there are in the first 6 or 7 weeks of the quarter.

Eventually, those things kind of even out, but the reality for us is that in December of 2010, we were [revenuing] $870 million.

We're forecasting that we'll revenue $745 million.

So, in the course of a couple quarters, we're going drop $125 million.

And I guess on one hand, people could look at the fact that our operating income is coming down from $27.7 million to a forecast of $20.5 million.

But if I look historically, in the past, when we've seen these kinds of revenue drops, we've actually dropped our financial performance significantly greater.

If we go back to the last time this happened, when we looked at some of the higher revenues in 2007, a couple quarters later dropping $125 million in revenue, our profitability dropped 14 percentage points as opposed to the 7 percentage points that we dropped here.

Obviously, we could mitigate that if we were to choose to react in a short-term way by pulling back on some of our customer facing and R&D investments.

But I think that that's not a smart move for us to do.

We've tried to articulate why we're doing what we're doing.

And at the same time, while the P&L might be looking a little bit rough, our cash generation is really strong, the 30% in March that Ernie talked about, and we expect to be at least that in the June quarter.

And so, I think 1 last aspect, which relates to the second half of your question is, if we felt that the decline in shipments and revenue that we're seeing coming off the peak in December was a trend heading south, we might be taking different actions.

Clearly, we're operating to the assumption, based on our discussions with our customers, that the second half is going to play out with stronger shipments and therefore the opportunity for higher revenues.

And so, we're going to stay the course in the short term.

And we're going to continue to build on the relationships and the trust that we've built with our customers.

And I think target to be able to deliver them solutions that ultimately manifest itself in continued marketed share growth going forward.

That's helpful.

As my follow-up, can you talk a little bit about gross margins?

And I know clearly, shipments and revenues played a role in that .

But, as you look at the tailwind from what you're doing on the clean side, can you comment on a steady state revenue line item?

What kind of upside we could see over the next 6-plus months from what you're doing on the clean

Ernie may have some comments that he wants to add, but we've talked about the fact that our target as a function of our product [option] architecture redesign of our moving to a remote factory outsourced supplier model that is similar to etch, that we wanted our margins in the low to mid-40s% for our spin clean product.

And we certainly want our overall margins from our other products to be at least in that mid-40s% range.

I think we're on track relative to those outsourcing and option architecture activities.

I think that how the margins will play out will really be a function of how much of our business is coming from what you might call spin clean non-critical, or commodity, versus spin clean critical and linear critical, where we're bringing a differentiated capability to bear.

And therefore, our customers, in recognizing that critical capability, are willing to pay more for that.

So, there's going to definitely be mix issues, and there will be total volume issues.

But I believe that as we look at the second half, that we have an opportunity, if customers come through with what they said they would do, we'll see our margin contribution from our clean division be a more positive factor than what's going on right now at this time.

Yes, C.J., this is Ernie.

If you step back and look at the semi-con models and presume, based on the shipment patterns that we just talked about that we get back into that range, we're going to be back within spitting distance of those models.

It may not be precisely there as the result of product or customer mix, as Steve articulated, but that's a good long-term guide to think about relative to macro-level margin performance for the Company at those shipment levels.

Very helpful.

Thank you.

Thank you.

Our next question comes from the line of Edwin Mok with Needham & Company.

Hi, thanks for taking my questions.

First question regarding, as I look at your geographic mix of your shipments, it seems like North America picked up a lot in last quarter.

Can you help us out in terms of defining what kind of customer you're shipping there and what type of products [between etch and clean] are shipping to these North American customers?

From a detailed standpoint, we report what we do kind of where the geographic headquarters of a company is, and so clearly, if you think about who is headquartered in North America, that spends, you have some activities relative to Micron, you have some activities relative to Global Foundries.

So, I think that probably the biggest aspect of it would really be what's the activity going on for the logic activity in Austin.

And then, you've got some activity that's related to other North American operations.

Great.

So obviously not incremental [there I guess].

One small question that is more long-term for you, Steve.

The industry is talking about going to a 3-D [type] sensor.

In your commentary you talked about eventually that could be a growth driver for your etch business.

Is there a way you can kind of quantify how industry transition to 3-D could potentially benefit the [etch] market?

And any way you can quantify in terms of increased number of steps or anything like that?

I think that when we look at the timeframe, that these new 3-D architectures are going to come to market, we're talking about 3 to 5 years, depending upon whether you're talking logic or you're talking NAND.

We're not talking about things that are going to be really resulting in significant revenue gains in the short term.

I think the comment that we're making is that the difficulty in getting performance and yield at next-generation-plus semiconductor devices is quite challenging for our customers.

And so, when you look at capital intensity that has jumped up significantly from the 40-nanometer node to the 28-nanometer node, which will increase again at the 20-nanometer node, I think it's a reasonable expectation that with the complexities involved with [sim sets] and such vertical transistors and NAND, and then ultimately other new transistors and new materials being introduced in DRAM, that there's going to be some significant spending relative to the need for very critical etch capability.

We've already seen the trend where dielectric etch was moving to 60% and maybe even for some periods, it was 65% of total etch market.

We think by the second half of 2010, that silicon segment, the conductor segment, is probably now operating pretty doggone close to 50% of the total etch market.

I think that trend will probably stay pretty stable in that arena.

But there is a potential that it could continue to grow, depending upon what kinds of materials ultimately are used in terms of barriers and metal hard masks, and what are the etch rates and how many chambers on a dedicated basis will have to be used.

But, I think that we talk about what's going on at the 14-nanometer node really as a function of driving recognition of the customer facing and the R&D activity that's going on that's much greater than what would normally have occurred.

Next question, please.

Thank you.

Our next question comes from the line of Krish Sankar with Bank of America/Merrill Lynch.

Just a couple of questions.

Number 1, if you look at your NAND shipments in March, and probably what they end up being in June, if you take the [run rate], do you think that's good enough to sustain an $8 billion investment in NAND this year?

Or, do you think there should be a snap-back in the second have half?

I think that when we look at the March activity, certainly it was strong for NAND.

It will be stronger in June.

But, when you look at what's being requested, there still will be a strong September, potentially September being stronger than March, and December is looking pretty strong already.

And we're a long way off from December.

And so, NAND might be more balanced in terms of a 50/50-type shipment profile.

But really, at this point in the year that could change.

But that's kind of how it's playing out right now.

Got it.

You also alluded to the fact that the silicon etch market should grow at faster growth rate than the dielectric etch market this year.

I'm trying to figure out what it means for Lam Research, given the fact that you are already probably two-thirds of the silicon etch market.

How do you think of the incremental share gains from here going forward?

Thank you.

I think that clearly, our strongest market share is in the silicon-based applications.

And so, as silicon increases its growth rate, then that just contributes to an even greater overall contribution to our etch market share.

We believe we're in the very high 60% relative to silicon, and we are right in the high 30s% or 40% for dielectric.

I think both markets are very important, but clearly, we've benefited from a strategic orientation relative to capturing very high market share with a lot of the new applications in the middle of the line.

Certainly in NAND, as double-patterning has emerged and we're using our conductor tools for that, that has been beneficial to us.

And so, I think that over the next couple years, you will continue to see our etch market share increment up, certainly from an applications win basis.

Exactly what it is year over year, will sometimes fluctuate as a function of which customers are spending to what percentage of the served available market.

But at the end of the day, we can't control that.

What we can control is continuing to win new applications and grow the overall potential for our market share going forward.

Thank you.

Our next question comes from the line of Mehdi Hosseini with Susquehanna International.

Thanks for taking my question.

Steve, going back to your earlier commentary regarding the comeback in the second half, and especially considering the fact that maybe some of this push-out or marginal weakness has come from foundries, how could I reconcile that with the lead times and how foundries have to manage their capacity for the second half, assuming that they would have their most wafer shipments in Q3 in advance of the holiday season?

And as they exit the year, obviously they are going to have less wafer shipments.

And to that extent, why should they procure more equipment when their utilization rate could come under pressure exiting this year?

That's a good question, and the answer is, potentially, very complicated, because when you look at foundry and logic, you have to recognize that in the December quarter, there was a very significant output of shipments into the industry.

And that's clearly being absorbed right now and being converted to output, largely 65-nanometer and 40-nanometer-related.

I think that when you look at what the foundries are going to do later in the year, there will clearly be, I think, an increased investment at 40-nanometer, as some customers of the foundries are indicating that they would like to move perhaps a little faster, from 90-nanometer and 65-nanometer to the 40-nanometer node.

And so, the foundries have to kind of rationalize how many wafer starts am I running at which technology node, and so sometimes they have to kind of rethink and re-plan.

The other thing that I think will be a major reason for high level of spending in the second half of the year, at least in terms of dollars, is that a lot of the wafer starts that will be purchased in the second half will be targeted for 32-nanometer and 28-nanometer.

And the cost for 10,000 wafer starts at that technology node is somewhere around $1 billion dollars for 10,000 wafer starts, versus if you are adding 40-nanometer you're probably spending $850 [million] maybe $750 [million].

It depends on whether it's 45-nanometer or 40-nanometer.

But capital intensity at the 32-nanometer/28-nanometer is significant higher, and so when you look at the spending that will occur in the second half, a significant portion of that is for the 32-nanometer/28-nanometer node, and it's expensive.

And then just 1 question for Ernie.

Any particular reason why DSOs went up by 9 days?

They were essentially flat, Mehdi, and that's really a function, as Steve spoke of earlier, the shipment pattern for the quarter where many of the shipments occurred in the last 30 to 45 days in the quarter.

And anything that ships within that time is not collectible.

Got it.

Thank you.

Thank you.

Our next question comes from the line of Ben Pang with Caris & Company.

Thanks for taking the question.

First, a follow-up on the previous line of questioning regarding the foundries.

You commented that you did see some foundry push-outs, but you expect that some of them could come back in the second half of the year.

What node has pushed out right now?

And is this going to be the same technology node (inaudible) or a different node?

That clearly varies by customer.

I'm not going to address that specifically, because of customer confidentiality reasons.

So, you'd really to have ask them.

But, I mean, we've got a variety of customers.

They're running volume production on -- typically, I mean, 3 nodes.

And so, it varies across the board depending upon which customer you're talking about.

And I think really, if you go back and look at how much was shipped into foundry logic in December, it was extremely high.

For us, it was about 60% of our shipments.

There was still a significant amount of shipment activity in the foundry logic arena for March.

And it's not unusual that after a couple of quarters of really significant shipments, you kind have a quarter of, we've got to absorb all this stuff.

We've got to get it all started up.

We've got to get in production, we've got to get it yielding.

And that's certainly a part of kind of what I see going on in foundry, and it's why I think that when we talk to our customers about what they're doing and what they expect to be doing in the second half, that I'm confident that we'll see a strong foundry logic spending environment in the second half.

Okay.

And my follow-up is based on the new makeup of wafer fab equipment in 2011 that you are working off of, what's changed in your served available market for your clean products between 2010 and 2011?

I think that as the logic sector continues to move more and more wafer starts to 40-nanometer and then to 32-nanometer/28-nanometer, I think what we're seeing is we believe that the single-wafer clean market is probably going to grow 20% to 25% next year.

We expect that our clean division will grow greater than that as a reflection of the market share wins from last year and early this year that will manifest itself in shipments.

And I think that it's going to grow.

It may be the fastest segment that we'll see in terms of growth over '10 and '11 because logic is very much a single-wafer clean-oriented environment and becoming even more so.

But there are also opportunities as conversion starts to pick up in NAND and DRAM, and that contributes to the growth rate of the clean market.

thank you very much.

Thank you.

We have time for one more question.

Thank you.

Our next question comes from the line of Timothy Arcuri with Citigroup.

Hi, this [Weng Yuan] for Tim.

Thank you for taking my question.

Recently, there was a lot of discussion on the readiness of [EUV] for the next couple of nodes.

Could you comment on what does the EUV impact on the etch market and specifically, for Lam's silicon etch products?

The impact of EUV on silicon etch.

I think when people talk about EUV, it's similar to talking about new [phinphet] structures and other 3-D.

It's a significant ways out.

We've done the modeling relative to when EUV, and if EUV is introduced, how many layers is it going to be applied to.

And how many double-patterning type of solution orientations will then reduce down into 1, and the reality is, by the time EUV gets introduced, there are going to be such a significant number of additional etch steps that will have been added as a function of the metal hard masks, other structures that are going to require additional etches to be able to execute, that we believe that the impact of EUV is at worst case neutral.

And as a function of architectures and when it will be introduced, I would expect that we'll be shipping more etch dollars per 10,000 wafer starts at that time than we do today by a fair amount.

That's helpful.

Just a quick follow-up.

[TSMC] is discussing a 450-millimeter fab in 2013 timeframe.

What is Lam's plan on 450?

Thank you.

We will, and we have committed to our customers, that as they define what it is they would like to do AT 450-millimeter, that we will do what we need to do.

And we'll be in position to support them.

And so, we engage in discussions with a number of customers about it, and I think our technology road map is well aligned to what customers are asking to us do.

We'll go ahead and take 1 more call.

Thank you.

Our next question comes from the line of Jagadish Iyer with Arete Research.

Two questions, Steve.

If I look at the third-party estimates for etch market as a percentage of (inaudible), it popped up nicely to 15% last year.

So, do you think that trend could continue for this year?

And I have a brief follow-up.

I think my advice is to take the high degree of suspicion, the inputs that you get from some of the third parties that etch is 15%.

I think our analysis and others who actually are in this business would say that it's closer to 13% -- may have been 13.5% -- but for the most part 13%.

I think that percentages can also be deceiving, because as a number of logic companies have commented, that once you go to 28-nanometer and 22-nanometer, that litho becomes extremely expensive.

And so, etch as a percent may actually drop.

But, etch on an absolute basis for every 10,000 wafer starts will actually increase.

An example would be, for 10,000 wafer starts at 65-nanometer, you probably had to spend $65 million in etch.

That was 12.5% to 13% of that $600 million or so, or $550 million you spent at 65-nanometer.

If you have to spend around $1 billion at 28-nanometer, etch might fall to 12% to 12.5%, but that means you're going to have to spend $120 million to $125 million for those 10,000 wafer starts.

And so, from the standpoint of increased business for us, it's going to be clearly there, and so I think looking at percentages can be deceiving.

And just to follow up, given that you have your PC growth that you have set at 11%, do we believe that shipments in DRAM could show an uptick in current levels for you, meaningfully in the second half?

Thanks.

Yes, I think that if the corporate refresh cycle decides to get more aggressive and accelerate, if investments in enterprise come up and certainly if we see that the consumer PC segment, whether it's in the kind of the established countries or we see some acceleration in emerging, clearly there's the potential that more capacity would have to be added in DRAM.

But, we'll just have to see how that is.

We think we've kind of positioned this based on our conversation with customers that people are being relatively prudent, relatively conservative, and we'll see where it goes as the year plays out.

Thank you.

Thank you.

I would now like to turn the call back over to Management for any closing comments.

Thank you.

Thank you all for joining us here today.

As a reminder, the audio replay will be available on our website later this afternoon.

And that concludes our call.

Thank you.

Ladies and gentlemen, this concludes the Lam Research Corporation March 2011 quarterly results conference call.

You may now disconnect.

Thank you for using ACT conferencing.