科林研發 (LRCX) 2009 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by.

Welcome to the Lam Research Corporation December 2008 quarterly financial 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).

This conference is being recorded today, January 28, 2009.

This call is scheduled to conclude at 3:00 PM.

I would now like to turn the conference over to Carol Raeburn, Senior Director of Investor Relations.

Please go ahead, ma'am.

Thank you.

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

Here today are Steve Newberry, President and Chief Executive Officer; and Ernie Maddock, Lam's Chief Financial Officer.

Today we will discuss the financial results for the December 2008 quarter, and Steve will share our business outlook for the March 2009 quarter before opening up for Q&A.

A press release detailing our financial results for the quarter ended December 28, 2008, was distributed by Business Wire shortly after 1:00 this afternoon and is available on our website at lamresearch.com.

Today's call contains forward-looking statements including those related to our forecasts of shipments, revenues, expenses, margins, and earnings per share, 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, which can be found in the slide package accompanying this conference call and on our most recently filed Form 10-K.

All forward-looking statements are based on information as of today's date and the Company assumes no obligation to update any of them.

This call was scheduled to last until 3:00 PM and we ask that you please limit questions to one per firm.

With that, I'll turn the call over to Ernie for a review of the December quarter results.

Thank you, Carol.

This afternoon, we will discuss our December 2008 quarter financial performance as well as our 2008 calendar year results.

As expected, our quarterly results are a reflection of the challenging business climate.

Shipments for the quarter were $226 million, down 35% from last quarter and consistent with the high end of our revised guidance range.

For application and market share, market segment breakout for the quarter, 300-millimeter applications represent 85% of total system shipments.

Applications at less than or equal to the 90-nanometer technology node were 85% of system shipments.

Memory segment customers in the quarter comprise about 48% of the total system shipments, and as a subset, the NAND components represent approximately 42% of total memory.

Foundry customers were 28% of system shipments with logic and other at 24%.

December quarter revenue of $283 million is consistent with the high end of our revised guidance and is down 36% from the prior quarter.

2008 revenue was down 27% from record levels achieved in 2007.

During the second half of 2008, we saw declines in systems as well as spares and service revenue relative to the first half of the year, and for calendar year 2008, overall spares and service revenue posted modest growth.

Ongoing gross margin was 38.5% for the December quarter, consistent with the guidance provided in our last call, and is lower than the 42.3% for the September quarter due to product mix and reduced factory absorption related to business volume.

Turning now to review our quarterly expenses, ongoing operating expenses were $126 million in the December quarter, which is down $23 million from the September quarter.

Lower expense levels reflect aggressive action to contain expenses and initial benefits from our previously announced restructuring activities.

In addition, our operating expenses benefited from two significant discrete compensation adjustments.

The first of these was a reduction in variable compensation liabilities and the second was a reduction in the Company's liability to participants in the executive deferred compensation program resulting from recent stock market declines.

Our ongoing operating loss was $17 million or 6% of December quarter's revenue, better than the operating loss of 11% provided in our revised guidance.

In the December quarter, ongoing other income was $300,000, a decrease from $9 million in the September quarter.

Our interest income and interest expense were similar to the prior quarter, and as a result of declines in forecast revenue, we incurred a foreign exchange loss related to our hedging program.

In addition, the remeasurement of our non-US balance sheets resulted in foreign exchange losses.

Non-ongoing items were $27.6 million for the quarter and consisted of $17.8 million for restructuring expenses, a $7.6 million loss on currency fluctuations related to our tax restructuring, and $2.1 million for other non-ongoing items.

Moving on to taxes, we received a tax benefit on our quarterly operating loss as a result of an ongoing tax rate of 31.2%.

The benefit was significantly lower than what would have been generated by the 70% to 80% tax rate forecasted for the December quarter and reflects tax decisions made during the quarter as Lam's business outlook including total fiscal year results, became more clear.

As many of you know, our tax strategy centers on maximizing earnings in low tax jurisdictions and reducing earnings in high tax jurisdictions.

In periods of losses, this dynamic reverses, accumulating most of the losses in low tax jurisdictions and generating small profits in others.

While we remain confident that our tax strategy will yield a long term tax rate of approximately 25%, results generated during the next few quarters will be highly variable and non-intuitive.

For example, our current view for the March quarter is a tax expense of approximately 7% to 12% of our operating loss.

We continue to avail ourselves of tax opportunities that will generate long term benefit for the Company.

The lower tax benefit created by this quarters tax rate differential generated an ongoing operating loss per share of $0.09 versus our revised guidance of $0.04 to $0.05 loss per share.

Turning now to the balance sheet, cash and short-term investments, including restricted cash, total $1.1 billion, representing a decline of about 8% from September.

Cash from operations was a negative $39 million.

Helped by our strong cash performance in the first half of the year, our cash from operations for the full year was 18% of revenue and within our targeted range of 15% to 20%.

We also achieved our objective of generating positive cash from operations in the Spin Clean business.

Major non-operating cash disbursements included $23 million for stock repurchase activity, $12.5 million in debt repayment, and $9 million for the purchase of the remaining minority interest in SEZ.

Our stock repurchase activity was concentrated at the beginning of the December quarter and we have currently suspended purchases under the plan authorized by the Board of Directors.

The rapid change in business volume has contributed to our inventory performance of 2.7 turns versus a September quarter performance of 3.8 turns.

Accounts receivable days outstanding increased to 93 days, up from 64 days in the September quarter.

On a forward-looking basis, we will continue to selectively extend payment terms to certain customers and expect that the level of business that includes expected payments, extended payment terms will remain relatively constant on lower revenues, and thus expect higher DSO levels will persist over the next few quarters.

At the end of December, Lam's deferred revenue balance was $68 million.

This amount does not include shipments to Japanese customers of $9 million that will revenue in future quarters.

Total capital expenditures were $12 million.

Depreciation and amortization remained almost constant at $17 million.

Employee headcount declined to approximately 3,300 from 3,700 in the September quarter and reflects most of the headcount reduction outlined in our previously announced restructuring plan.

For more complete details of the geographic break down of shipments and revenues, please see today's press release and our website for a reconciliation of our shipments, revenue, deferred revenue, and cash.

I'll now turn it over to Steve.

Thank you, Ernie, and good afternoon, everyone.

On our last call, we reported that we were beginning to see signs of rapid deterioration in the semiconductor industry, including decelerating IC unit demand, significant reductions in fab utilization, and sharp declines in investments for new system shipments as well as the beginnings of reduced spare parts and services purchases.

In November, we implemented a cost reduction plan designed to lower our breakeven and reduce our quarterly spend rates by approximately $15 million to $20 million per quarter.

Some might characterize those actions as bracing for a Category 3 to 4 type hurricane -- major but not quite catastrophic.

As the quarter progressed, conditions continued to deteriorate to the point that now the storm has hit our shores and we are being pummeled by a full blown Category 5 hurricane of unprecedented speed and force.

As I speak today, our environment is impacted by continued decline in IC unit demand, high inventories in all segments of semiconductor ICs, along with extremely tight credit markets and the presence of huge losses in the semiconductor manufacturing segment, resulting in an almost total shutdown of purchases for new equipment and a significant reduction in purchases for spare parts, consumables, and service support.

New equipment purchases that do exist are almost entirely and strictly for specific tools needed to upgrade production lines to the next technology node.

I expect this environment to continue for as long as it takes for the global economies to stabilize and for IC unit demand to recover and then grow beyond demand levels we saw earlier in 2008.

Eventually as a function of a stronger global economy and the introduction of new applications and new electronic products, created in large part by new semiconductor devices coming out via 4X and 3X generation production fabs, we will see a return to stronger investment for capacity expansion.

How long it will take before meaningful wafer start expansion occurs is anyone's guess, but we are expecting this type of environment to persist for at least the next six to eight quarters.

We expect wafer fab equipment spending in calendar year 2009 to be in the vicinity of $10 billion, representing the decline from calendar year 2008 of approximately 50%.

This level of spending will be the lowest level of wafer fab equipment spending since 1994 and 35% to 40% lower than the spending in the middle of the last major downturn in 2002.

Reasons for those lower level of spending forecasted for calendar year 2009 is a much more consolidated semiconductor manufacturing industry operating at a more capital efficient 300-millimeter, the availability of used tools to fill 200-millimeter wafer start demands, slower adoption in the foundry and logic fabs to next generation technology nodes, limited access to debt to invest in new technology production even for those who want to and need to, and during this downturn no need for investment to position for a new wafer size that existed in 2000 and 2003 timeframe when the move occurred from 200-millimeter to 300-millimeter.

This leads us to an outlook for the March quarter.

We expect our shipments to decline to the range of $155 million plus or minus $15 million.

Revenues are expected in the range of $175 million plus or minus $15 million.

Gross margins in the range of 25% plus or minus 2 percentage points.

We anticipate an operating loss in the range of $70 million to $90 million, around the midpoint of the revenue guidance, with net income as a function of the unfavorable tax situation that Ernie described earlier expected to be a loss in the range of $0.60 to $0.80 per share.

As a function of these sharp drops in shipments, revenue, and profitability, we will be implementing an additional significant set of reductions to the cost structure of the Company, targeting additional variable cost reductions as well as this time, elimination of various aspects of our fixed cost structure that are unnecessary relative to our expectations of our future shipment and revenue levels.

This cost reduction plan is being developed by our management team as we speak and I expect we'll be communicating the details of the plan some time in the first half of the month of March.

In addition to this comprehensive cost reduction plan, we are taking two immediate actions that are targeted at reducing our expenses in the March quarter.

These two actions are an additional five days of shutdown in the month of February and the implementation of a salary reduction program for all employees ranging from 17.5% reduction for the CEO to 2.5% reduction for our lowest compensated employees.

These near term actions are already significantly embedded in our March quarter guidance.

Any the additional improvement as a function of cost reductions affecting the March quarter will be communicated if material.

In spite of the near term challenges of weathering this Category 5 storm, I remain optimistic and confident in the ability of Lam Research employees around the world to continue to deliver world class solutions to meet our customers' needs.

As a result of their previous efforts, we enter this downturn with a strong balance sheet, which we will utilize to enable us to continue our focus on the following priorities while supporting our customers current install base needs.

We will continue our investments in the etch markets with a focus on continued application share gains.

Areas of focus will be in winning the emerging applications around double patterning and other transistor related critical etches.

In Clean, we will focus on rapid introduction of our newest spin product, the DV Prime, and aggressively pursue penetration into new applications emerging in the front end of the line, as well as material and manufacturing cost reduction activities in our Spin Clean division to improve its gross margin contribution to the Company.

We will continue to invest and aggressively work to qualify our linear C3 clean tool into additional critical cleans requiring short contact time and precise CD control for yield and liability improvement.

In addition, we expect to introduce shortly a new damage-free particle removal capability with the linear tool expanding the applications where it delivers superior and differentiated clean results from the wafer.

We will also continue to focus on the development of new products targeted at new markets in need of advanced technology solutions, as well as continue our investments in globalization of our supply base in order to deliver high quality, low cost, long lasting spare parts and consumables to our customers.

The key objective for us during this downturn will be to prudently utilize our strong cash position, to make the appropriate strategic investments which will position us for strong financial and operational performance when the industry returns to investing and expanding wafer start output.

Lam Research has traditionally used downturns as a time to redefine our business model and work closely with our customers to solve their problems in anticipation of future production ramps.

We expect to operate no differently in this downturn, and as before, we expect to emerge an even stronger and better Company.

This concludes my comments, and with that we will now open the call for your questions.

Thank you.

(Operator Instructions).

Our first question comes from the line of C.J.

Muse with Barclays Capital.

Please go ahead.

Good afternoon.

Thank you for taking my question.

I guess, Steve, a question here is on the cost cutting side, and I know you were going to give more details in March, but I guess if you can't share a targeted breakeven given your call that this downturn is going to persist for another six to eight quarters, can you share what cash burn you're comfortable with given your currently strong balance sheet?

Well, I think that we have a Board of Directors meeting next week, and as I have not had an opportunity to discuss that with the Board.

We have not really looked at enough of the specifics of the significant cost reductions that we want to take.

I think it would be premature for me to communicate that because I think that that will be a decision that I and the management team will present to the Board in terms of what we think it's going to take to be able to make the investments that keep our products and our technology position strong, support our customers effectively.

And we'll come to a set of decisions with the Board as to what that cash burn is likely to be and then we will complete the specific cost reductions that will be aligned to that.

So sorry I can't answer that for you right now.

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

Please go ahead.

Hi, Steve.

I want to ask you a question about maintenance CapEx and how you think about that.

I remember Jim used to talk about the fact that there's an upgrade that chip makers have to buy that's every two years that's basically 10% of the initial tool price.

So that would imply there's 5% of the dollar install base that basically has to be turned every year.

And if you look at the trailing five year dollar install base, we're about $260 billion, and there's some that's come out of that because of capacity shutdowns and things like that.

So maybe you're at $200 billion or something a little bit higher than that.

So on that math, it would imply that at a wafer fab level, maintenance CapEx is maybe about $10 billion, say $10 billion to $12 billion, which is basically where we are to a bit higher even than what you're guiding in '09 wafer fab equipment.

So I'm wondering, how do you think about maintenance and would you agree with those numbers?

That's a good question, and we've been studying that pretty significantly trying to look at what those maintenance levels were in past significant downturns of this magnitude.

And one of the things that we looked at of course was 2002, $14 billion was spent.

But when you look at the makeup of that $14 billion, there was a couple things that were present.

And one was 1) a much greater number of customers that were involved in spending and we've calculated that at about $3 billion that was spent in 2002 by a lot of the logic companies that were still investing in 200-millimeter and technology buys.

And for this downturn, essentially with consolidation and no need to invest any significant 200-millimeter because you can get all of the 200-millimeter tools you want on the used market, you then go and look at, well, what's happening on the leading edge relative to memory and relative to foundry.

And we view that somewhere around $7.5 billion to $8 billion will be spent by the top 10 companies in the industry, and that's actually very correlated to what they spent in 2002.

Our data says they spent a little over $8 billion, maybe $8.2 billion.

And so then when you look at the next 10, which are still players that are present, still players that are investing, in 2002 they spent about $2.7 billion.

And we think in 2009, they are more likely to spend maybe $1.5 billion and then another $1 billion or so rounding for other players.

And so I think that when we look at maintenance levels, it's a function of what capital asset efficiency today, which I think is clearly higher at 300-millimeter.

I think that we have a slower move to the next generation technology node so that the foundries and the logic companies, even though they are trying to move to 4X, they are ready to move to 4X, their customer demand pool at 4X is slow.

So we can see that the ramp in wafer starts for 4X is muted relative to the speed at which we were moving back in the 2002 timeframe.

And then if you couple with that there's memory companies who would like to move, whether it's 6X or 5X, because they have designs ready, they can't get access to capital, particularly the Taiwanese manufacturers who have lost literally billions of dollars in the last couple quarters.

So that's suppressing the ability to operate.

So I think your numbers that would say under maybe credit markets that were flowing, you'd probably see it more around $12 billion.

But that's one of the reasons why we're sitting at about $10 billion because of the accumulation of all of those factors that I'd mentioned.

Thank you.

Our next question comes from the line of [Vis Solari] with Credit Suisse.

Please go ahead.

Hi, actually it's Satya Kumar.

Steve, if I analyze your Q1 guidance, it seems like it will be down about 63% year on year.

Your guidance for CapEx is down 50%, and you're talking about a six to eight quarter downturn.

Is there really any meaningful inflection that you're seeing in the business?

It doesn't seem like that's the case for the rest of the year.

And the second part to that question, if I look at what's happening in DRAM right now, with the slowing growth in content per box and the slowing growth in ICs, is there a permanent reset that we're seeing with the capital intensity for the silicon industry?

Satya, can you repeat the first part of your question again?

Yes.

I'm just trying to see if you're seeing inflection in your business as you look out beyond Q1.

If I annualize the Q1 revenues it's down about 63% versus your CapEx outlook of down 50%.

So it doesn't seem like you're expecting much of a meaningful inflection in your business beyond Q1.

Is that the right math?

And the second part is on capital intensity of the industry longer term.

There's a couple things going on both in terms of the December quarter and the March quarter, and that's one is the presence of Intel spending, which was probably about $1.2 billion or $1 billion or so of wafer fab equipment, and it's not clear what Intel is going to spend in the first couple quarters.

I think it all depends on how fast they want to try to get their 32-nanometer product ramp going.

But clearly, where we are right now is in a situation for the March quarter where even those companies that clearly are planning to do technology related buys are almost on a weekly basis saying yes, then no.

Yes, then no.

And customers are clearly managing their cash extremely carefully.

They are monitoring the demand.

They are monitoring their competitors, and we have a situation where literally, just two days ago, $10 million of what we were scheduled to ship in this quarter pushed out to June.

The magnitude of what has changed in terms of what was planned to ship in the March quarter from only two months ago is about $240 million, so we're in a situation where the March quarter is hopefully the bottom of this in terms of a freefall in just stopping all spending.

And so while I think that we're going to be in a similar environment, I would expect that it's more likely that we'll see shipments in the quarters going forward being somewhat higher, although not significantly but consistent with a $10 billion level.

I think we view that etch, which is typically a 12% to 13% of wafer fab equipment, in this environment is likely to be closer to 10% or 11%, clean more likely to be -- in terms of total clean, 5% or 6%, slightly less than what it historically is because there's a lot of capacity related buys as opposed to technology buys.

So you're correct in that the March quarter if you run rate it, we would be actually operating to a lower wafer fab equipment environment.

And so I do expect that we'll see things get slightly better as the year goes forward but not dramatically so.

And then your second question was about DRAM and capital intensity.

I mean, right now, we're sitting on a shutdown of about 550,000 wafer starts per month of DRAM capacity.

That doesn't include about 70,000 of 110-nanometer production that occurs in Taiwan that's been shut down and is likely to never come back.

So DRAM capital intensity is going to drop to below 20% in 2009 and will probably come back to 25% or 30%, but that will be a function of continued DRAM IC unit growth, which we believe will occur.

But you've got to suck up a lot of idled capacity over the course of the year.

Plus you've got about 300,000 wafers worth of inventory either in finished wafers or in die banks that is a function of excess output that's occurred in the last few months or so.

So I think that when we look at the long term capital intensity for DRAM, and I'm talking about once we get to unit demands that suck up all of the excess and go forward, we're still probably looking at something in the 30% range.

Because one of the issues for memory of course is that they will be moving to the immersion nodes where 5X and 4X and DRAM and some 4X but clearly 3X and NAND Flash.

And so immersion is going to increase the capital intensity relative to what they were spending at the 6X node.

But what that means for Etch and Clean may not be as significant.

because a lot of that money obviously will go into lithography.

Thank you.

Our next question comes from the line of Gary Hsueh with Oppenheimer.

Hi, Steve.

Hi, Ernie.

You referenced how a $10 million pushout in the March quarter would come back and benefit the June quarter.

I know it's really still too early, but just brought to mind how much of a hand to mouth situation are we in?

How much of a situation are we in today in terms of hand and mouth?

Because you don't provide orders or order backlog.

I'm just wondering in terms of the $155 million shipment guidance for the March quarter, how much of that is pulling out of backlog and how much of it is instantaneous hand to mouth in order turns?

You could probably characterize us as being synonymous with the beggar standing on the street corner with a tin cup out.

Hoping that anybody that goes by will drop a dime in our cup.

I mean, it's one of those situations where you've got only less than two handfuls of customers, some of them with plans to take deliveries to $5 million to $7 million, and whether they've got the orders in or not is irrelevant because -- that's one of the reasons we don't talk about orders because you could have the orders in backlog and it doesn't make any difference.

If they want to push it they will push it and if they want to cancel it they cancel it.

And with lead types as short as they are, typically most customers are sitting there saying okay, here is what we want, build an inventory and a production plan to support it, and then we'll let you know the day before whether we're actually going to take delivery or not.

So that's one of the reasons why you see us guiding shipments at $155 million plus or minus $15 million because we're in that kind of volatility.

And we're at the point where when it pushes into June, it might be helpful to June, but what I don't know is what's going to push out of June and go to September.

And so we're in an environment where the visibility is fundamentally whatever we see and we're told doesn't mean very much, and you don't ship until you ship and everything else is estimates of what's going to happen.

Thank you.

Our next question comes from the line of Stephen Chin with UBS.

Please go ahead.

Great, thanks.

I was wondering if you could provide, Steve, some more color on the March shipment guidance.

Is the sequential shipment decline due more so to a lower spares and services business?

Or is it really from lower equipment shipments?

Or are both segments declining at the same rate now?

And if you could share any color on how the gross margin of the spare services is going up relative to I think your corporate gross margin is 25% -- that would be helpful.

Thanks.

I'll have Ernie talk a little bit about some of the makeup of the 25% gross margin, because it's complicated, but probably worth addressing for sure.

Both -- the systems business is off dramatically.

The spares and service business is off less so.

Typically what we see is the spares business will track the decline in wafer starts, because spare parts and consumables are a function of wafer starts and the time that equipment's running.

And so we're looking at an environment where we've seen those wafer starts decline 35% to 40%, foundry utilizations running 40% to 50%.

So you can expect that the spares and service business will decline in that vicinity and that's where we are in the quarter.

The systems business is down in the 60% to 70%, again because we're in a little bit of a trench here, but clearly that's a function of the fact that you don't need any capacity.

People are wary about the rate at which they want to do technology at, and so that's the makeup of systems and service.

So Ernie could talk about margin.

Sure.

We tend to look at it relative to a pure margin on the tool set, and if we look at that, that is holding up relatively well.

Obviously at these revenue levels it is significantly impacted by the mix of which customers happen to be buying.

So on an underlying margin basis, while there is some small perturbation, we don't see any significant underlying change to the structural components of that.

I think the thing that is most impactive of the March quarter to a far more significant degree than the December quarter is the level of margin impact of the factories, which are obviously being utilized at significantly less intensity than they were during the December quarter and certainly during prior quarters.

So that is -- if I were to point to one thing that is most impacting the margin change, it would be a very significant change in the level of absorption coming from the factories, which is related to Steve's comments during his discussion about the fact that we're beginning to look at the fixed cost infrastructure of the Company and size that to the environment we see on a going forward basis to address some of those issues.

One other comment I would add to that is that from a product mix standpoint, in Clean, over the past year or so a lot of our revenues were associated with a mature released Da Vinci product that had good margins.

The environment today in terms of Clean purchases is almost exclusively for new technology penetrations and new technology driven applications, which means we're shipping our Da Vinci Prime, which is a brand new product.

It's in its early phases of material and labor manufacturing cost reduction, and if you combine that with the fact that you've got a factory absorption issue and a lot of margin pressure coming out of the Clean group as a function of product mix as well as volume.

Thank you.

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

Please go ahead.

Hi, Steve.

Thanks so much for taking my question.

You made the comment about the idled capacity in Taiwan and elsewhere that's going to have to get absorbed.

Can you walk us through the dynamics of how that could come back online and need to get absorbed versus how that would just get shut down at some point if the downturn lasted long enough?

Like at what point does that idled capacity just no longer become useful?

Well, one of the things about idle capacity, if it doesn't get used long enough it goes obsolete.

So that's one of the nice things about our industry.

We look at today -- we think that there's installed 300-millimeter qualified capacity of about 1.340 million wafer starts per month that exists in DRAM.

We think with about 550,000 of that shut down, you've got 800,000 being utilized and that there is a demand that's about a need for 1 million wafer starts.

The problem is that we entered the first quarter with about 914,000 total wafers in inventory.

So about 300,000 wafer starts per month of excess capacity.

So we see that over the course of the year, leading edge DRAM demand will grow to a need for 1.300 million wafer starts, and by the end of the year, the 1.340 million of capacity will only grow to about 1.375 million.

And so we think by the end of the year we'll have worked down the excess inventory and we'll basically be idle with maybe about 90,000 to 100,000 of idle capacity, and I would expect that that will be 8X type technology that will be idled and probably doesn't come back.

So I think by the end of the year, if we get -- we think there's 4% or 5% unit growth in DRAM.

That equates to about 600 million units of DRAM growth.

And then because the output is shifting from 8X to 7X to 6X, and we'll see the start of 5X, there will be a need for investment to upgrade the 5X, and there will be a need to invest from 8 and 7 down to the next technology node.

And so we'll see all of that go, but we only think that there will be maybe slightly less than $3 billion of wafer fab equipment investment in DRAM.

If you look at NAND, we think that NAND excess is probably only around 100,000 wafer starts per month right now and that the inventory situation is not as bad.

There's probably 350,000 wafers, so about 100,000 wafer starts per month excess.

We think that's going to get absorbed fairly quickly because we do think that NAND growth is still going to be present.

We think that you're going to end up in a situation whereabouts 700 million units of NAND is needed.

And while we don't think that's largely SSD driven, we think that they will be positioning for that, because we think that Toshiba is going to move in the middle to the later half of the year to 3X and do conversion of 5X wafer starts in Y3 to 3X.

We think Samsung will aggressively move its mix that's largely 5X and 4X related, and they will reduce the 5X and go more heavily to 4X and 3X.

And then Micron of course is in that mix as well with the ability to move from a 5X production probably directly to a 3X.

And so we think that there is an investment in NAND that's probably around -- we're targeting at about $1.7 billion to maybe $2 billion.

But we think that by the end of the year, with an expected demand environment we're talking about, that we've got both DRAM and NAND running in a pretty healthy supply and demand balance.

And then, in 2010, I think how much they invest in terms of whether it's similar or whether it's more really is what's the global economic environment, what's the recovery in terms of the consumer spending, what are businesses going to do for PC expansion, which I'm sure they are going to delay.

So you could be setting the stage where in memory, you'll see an uptick in spending because they will have to add more wafer starts than they are going to have to deal with in 2009.

So hopefully that was helpful.

Thank you.

Our next question comes from the line of Steve O'Rourke with Deutsche Bank.

Please go ahead.

Hi, thank you.

Good afternoon.

Just shifting gears when you think about this downturn and maybe some of the opportunities you might have, how are your etch prospects in segments where you've traditionally not been very strong like logic, and what applications might be the lowest hanging fruit if it's possible to win?

Well, Steve, I think for one, actually our market share of logic is very similar to what it is in DRAM and memory.

The only issue is when you look at the IDM logic players, they haven't spent very much.

If you say foundry is logic, we have over 50% market share in foundry.

Where we're being impacted is our market share in microprocessor, because we do very well at AMD, who is not spending all that aggressively, and we don't have any business that we do with Intel.

So that clearly impacts us, and in particular where you have a year where Intel's percentage of spending is up significantly from where it's been in '07 and '06, that impacts our market share.

So having said that, where we're focused is in making sure that in the dielectric space where our market share is approximately 40% globally that we are targeting a number of additional critical dielectric etch steps, both in memory -- particularly DRAM -- and foundry, foundry logic, and we believe we have opportunities to move that up.

Our market share in conductor, which is slightly more than 60%, we have more limited opportunities, but we're certainly looking at some emerging applications for the use of conductor etch tools in double patterning applications.

And we're looking at how we can expand the portfolio of our etch product to give us a more diversified set of offerings that cover both what you might call the non-critical segment to the medium critical to the critical.

But certainly, most of our effort in conductor is about strongly defending our position and winning a few application gains and most of our market share growth objectives are around dielectric, where we see opportunities to exploit our capabilities.

Thank you.

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

Please go ahead.

Hi, thanks for taking my question.

Steve, I had a question regarding used tubings sold.

There was recently some news that your Taiwanese makers are selling off some of the fab capacity to other customers.

How much do you think that will impact the market or your business this year, and of the $10 billion WFE spending you're estimating, how much do you think that money will go into buying the used tubes [or] to some company like yourself?

Well, that's a difficult question, Edwin.

Certainly over the last couple years we've seen a huge amount of 200-millimeter memory tools come off line and go into the 200-millimeter logic market and go into the 200-millimeter foundry market.

Recently, ProMOS sold approximately $20 million worth of used tools to TSMC, which went into some trailing edge technology.

We've seen trailing edge fabs in various regions of the world that fundamentally, the entire fab is either started up on used tools.

So when we look at the 200-millimeter market, historically it's been about a 10% of the total market spending, and that's in terms of what the OEM equipment suppliers provide.

Certainly when you look at the used equipment brokers, that's a back door channel that there's a lot of difficulty in getting total visibility to that.

So we end up in a situation where in a year like 2009, I don't expect that the trailing edge 200-millimeter fabs are going to be purchasing very much 200-millimeter equipment, because their demand is down 15% to 20%.

And so they are going to be focused on effectively utilizing their existing tool sets.

There will be some investment in 200-millimeter chamber upgrades which will enable a customer to take a [130] 200-millimeter line and make it work at 90-nanometer or sometimes going from 90 to 70, but you don't have to buy new tool sets.

You can buy upgrade kits that can make an existing 200-millimeter install base perform at the next technology node.

But at this point, can't characterize what the volume of that is, but it's not going to be a big market in 2009.

Thank you.

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

Please go ahead.

Please go ahead.

Hi.

Thanks for taking my question.

Now that you have the experience with SEZ for a couple of quarters, Steve, do you still think you can grow the margins for the SEZ spend business to the corporate level?

And then I have a quick follow-up.

Yes, I think that there's a combination of issues.

One is what, relative to the competitive capability and the competitive pricing environment and the material cost of the product, we believe that we can get the product over the next year or so competitive.

Then we have to get the labor component, the use of our outsource suppliers, and our overall fixed cost factory infrastructure more in alignment with what the volumes are going to be.

So I think in the short-term which means the six to eight period environment where volume's low, that's going to be a more difficult challenge because we're going to need some volume to absorb certain aspects of that infrastructure.

Clearly over time, we will move the Spin Clean operation from what is largely a very vertically integrated manufacturing and supply chain structure to an outsourced model that, while it most likely won't replicate the degree with which we have etch in our linear product outsourced, we will move it strongly in that direction.

And that will help its ability to be able to get to margins that are consistent with our corporate margins.

But we'll have to see how its success is in terms of penetration and volume over the next couple years before we'll know exactly where it fits in.

Okay.

[DNS Semiconductor] in Japan just mentioned they are gaining share in [wet clean] segment in 2008, and they said their sales in the wet clean segment will be down 10% 2008 versus CapEx spending that's down more than 10%.

So just wanted to get your thoughts on what's happening in the competitive landscape in the single wafer wet clean business.

I understand in this depressed spending environment how to see the market share swings but I just wanted to get your thoughts on what's happening in the market share in the single wafer wet clean business.

So Steve, can you restate the beginning part of your question?

It was hard to understand.

Right -- so DNS at Semicon Japan last year in December said that they are winning share against other competitors in single wafer wet clean business, especially against SEZ in 2008.

And they said their single wafer wet clean sales were down 10% or are going to be down 10% in 2008, versus spending which is going to be down more than 10%.

So it does indicate from the numbers that they are gaining share and I just want to get your thoughts on if they are gaining share against you guys or the Korean guys.

Okay.

I think that a lot of what DNS focuses on is they obviously have a very strong market share position in the wet bench segment, which is dominating the front end of the line.

And that is beginning a conversion process on a number of applications in the front end.

And so sometimes they win a single wafer position, but that's not really necessarily market share, because they are just swapping wet bench sales for single wafer sales.

Having said that, most of SEZ's position that they clearly were successful with in the 2006 timeframe was in back end of the line and in backside and bevel cleaning.

Over the last year or so, the amount of investment in capacity related, particularly logic back in the line investments has been significantly muted.

The battle that is occurring in the front end of the line, I think that we are in a situation that we've acknowledged where DNS started working a couple years ago on trying to get their wet bench customers converted to their single wafer tool.

I think that they've had a degree of success in getting some tools in early position.

I think fortunately for us, as that front end situation continues to emerge, the downturn slows down the ramp relative to the volume purchases that customers do, tends to continue to keep the competitive environment open.

And so we think that both with our linear clean and with the DV Prime, which is now starting to gain some penetration traction, that we're going to have a good opportunity to win our share of the emerging front end of the line single wafer applications.

And then we clearly need to be able to continue to defend our traditional back end of the line backside and wet bevel.

And the reality is that there's not much buying going on and there won't be much buying going on there in '09.

So I would say that it is true that DNS has the momentum in the marketplace right now, as a function of the timing that they've had relative to their products.

But I think that we're going to have an opportunity to compete very hard in 2009 and we'll need to see how the application wins play out over the course of this year.

Thank you.

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

Steve, in terms of the memory market and the consolidation that's likely to occur in that space, how are you positioning the Company to react to a very changed landscape as we look forward?

Well, certainly what we would expect is potential to occur in terms of some of the Taiwan DRAM companies combining with each other or with Elpita, with Micron, et cetera.

Those kind of trends actually since the last major downturn have actually been occurring pretty regularly.

So whether you talk about IBM technology alliance of IBM and AMD and Chartered and Toshiba and Sony, we're very used to building global account teams that in essence work with the key customer or the key fab that's the source of the technology.

The way we approach it is we certainly make sure that we're well penetrated, well supporting where the customer is doing their next generation development.

We target the ability to win at the so-called home fab or mother fab.

And then with the performance of the install base in the fan out fab and with strong relationships and strong performance, we not only try to make sure that we replicate the copy exact fan-out, but we also try to use the performance of our tools to win additional application steps that we may not have been the first choice at the primary qual fab -- and actually grow our market share in the fan-out fabs above and beyond what it was at the qualification fab.

So we're quite used to doing that.

We're very well positioned with the DRAM companies.

We've been growing our share in the Elpita alliance.

We've been growing our share in the Micron [Initara] Nanya arrangement, and so we expect that we'll be in good position when we learn how this is all going to shake out in terms of who combines with whom.

Operator, we have time for two more questions.

Thank you.

Our second to last question will be from Ben Pang with Caris & Company.

Please go ahead.

Thank you for taking my question.

You mentioned in your prepared remarks you'll still be working on some of the technology transitions that your customers are interested in.

How active are these customers?

Are they actually going to make decisions in 2009 for some of the double patterning qualifications?

Are they still going to actually make decisions in 2009 or do those all get pushed further down the line?

I think that those customers who have balance sheets that can afford to make the investments to move to the next technology node will do so.

One thing about this industry is technical innovation does not slow down.

It may slow down for some who can't afford it, but it does not slow down in the industry as a whole.

Because what companies recognize is that the way that they're going to find themselves able to be the recipient of market share growth relative to IC unit demand and the segments they compete is that they need to be on the leading edge and they need to be as cost effective as possible.

So whether you're talking about 6X to 5X or even in some cases some 4X wafer starts in DRAM, certainly NAND is moving from 5 to 4 and we're going to see 3X come into play.

And then in logic and in foundry, the emphasis will be on growing needed wafer starts at 45, but early pilot lines for 32 positioning will start in the second half of the year.

And what you're seeing as a function of the consolidation of this industry is there are those who have been successful, who have balance sheets that can work through this kind of downturn.

There are those whose balance sheets are incredibly laden with debt, can't get access to capital, and they will be consolidated in and absorbed in.

But the pace of technology continues.

It's the future that I'm confident in, because I think that as we get to 32-nanometer and NAND and as we get to 64-gigabit devices and we get down the cost curve and up the yield curve, it's going to be the enabler for huge amounts of demand for SFD sometime in 2010.

And I think I'm very positive about the prospects for 2011.

I think you're going to see those who can try to accelerate the ability to move to 4X technology and 2 gigabit DRAMs and get down the cost curve relative to more density per bit per wafer.

And I think in logic, you've got a very intense competition around next generation logic at 32 and 28 nanometer that is both power and performance related -- the key being low power, high performance.

And there's a number of applications waiting for high performance, low power, logic devices that will be the next generation of smart phone enabled technology from semiconductor companies.

So those things are going to continue to be invested in.

I think we're currently in a period of a temporary lull, but I think that we will see those investments go forward, and that's why spending isn't $0.

It's why there's still going to be $10 billion worth of wafer fab equipment to support the continuing advancement of technology output.

Thank you.

Our next question comes from the line of Brett Hodess with Banc of America Merrill Lynch.

Please go ahead.

Good afternoon.

Two questions.

Quick one.

You mentioned that the DSOs are up and some of that has to do with the extended terms to some of the customers.

Can you categorize which types of customers you're extending terms to?

Are they the best quality customers or is there some additional risk in that?

And then can you comment on you were just talking about your new products and can you give us a feel if most of the spending is focused on upgrading existing fabs?

Are your etchers out there already, some of the equipment gets upgraded or replaced?

Or it will have to wait until you see more capacity adds?

I'll take the DSO question.

So we're seeing requests for extended payment terms across a pretty broad set of our customer base, and we have a very structured process that we use to assess whether or not we will extend payment terms to customers that includes things like business opportunity, the relationship we've had with the customer over the course of the last several cycles, as well as the specific need for technology or other factors.

So while we do see a very broad base for requests for extended payment terms, we're being very thoughtful and deliberate about the customers to which we extend them.

So relative to the makeup of new systems for technology upgrades versus chamber upgrades -- where possible, customers will take an existing install base.

They will upgrade the chamber if wafer starts that are running down that line are not fully outputting relative to that technology node, and assuming that the rest of the tool sets in there can handle the move.

But we do have a robust chamber upgrade business in our shipment numbers.

There are wafer start increases in memory fabs and there are wafer start increases on next generation technology lines where they can't get an existing line upgraded the way they want.

So they put a new line in and output with that, and it's different for every customer and it's different for every segment of DRAM and NAND and foundry.

But it's a combination of both new tools and upgrades.

I would like to thank you for joining us today.

Please be advised the webcast of today's call will be available on our website later this afternoon.

We hope you will join us for our next quarterly financial update in April.

Thank you for your interest in Lam Research and for participating in today's call.

Ladies and gentlemen, that does conclude our conference for today.

We thank you for your participation and you may now disconnect.