科林研發 (LRCX) 2009 Q3 法說會逐字稿

Good afternoon, ladies and gentlemen.

Thank you for standing by.

Welcome to the Lam Research Corporation March 2009 quarterly conference call.

(Operator Instructions).

At this time I would like to turn the conference over to your host, Ms. Carol Raeburn, Senior Director of Investor Relations.

Please go ahead, ma'am.

Good afternoon everyone.

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 March 2009 quarter.

And Steve will share our business outlook for the June quarter, before opening up for Q&A.

A press release detailing our financial results for the quarter ended March 29, 2009 was distributed by BusinessWire shortly after one o'clock this afternoon, and is available on our website at LamResearch.com.

Today's call contains forward-looking statements, including those relating to our forecast 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 is scheduled to last until 3 PM.

And we ask that you please limit questions to one per firm.

With that, I will turn the call over to Ernie for a review of the March quarter results.

This afternoon we will discuss our March 2009 quarter financial performance.

While lower business volumes were clearly evident during the quarter, we are pleased to have met our guidance in all categories.

Shipments for the March quarter were $159 million, consistent with the midpoint of our guidance range, and down 30% from last quarter.

For the quarter 300 millimeter applications represented 87% of total system shipments.

And applications at less than or equal to the 90 nanometer technology node were 87% of system shipments.

Memory segment customers in the quarter comprised about 56% of total system shipments.

And as a subset the NAND component represents at approximately 22% of total memory.

Foundry customers were 16% of system shipments, with logic and other at 28%.

March quarter revenue was at the midpoint of our guidance range at $174 million, and down 30% -- 38% from the prior quarter.

Both our systems and installed base business units performed at expected levels.

During the period of reduced capital spending on equipment, our installed base business unit continues to be an important contributor to our revenues.

March quarter ongoing gross margin was at the high end of guidance at 26.8%, and lower than the 38.5% gross margin for the December quarter.

Approximately half of the margin difference is due to lower factory and field absorption, and the remainder is due to product and business unit mix.

Turning now to review our quarterly expenses, ongoing operating expenses were $129 million in the March quarter, which is up $2 million from the December quarter.

As you may recall, our December quarter operating expenses benefited from one-time credits related to accrued compensation reversals and a reduction in executive deferred compensation liabilities resulting from stock market declines.

In the March quarter operating expenses we saw the expected savings from the salary reductions that were announced on our last call.

As anticipated though, these compensation savings were offset by an increase in employer payroll taxes, which are typically higher in the first quarter of a new calendar year.

Additionally, in response to recent financial announcements from our customer base, we recorded within operating expenses a charge of $7 million to increase the reserves against our receivables balance.

Our ongoing operating loss comes in at the midrange of guidance at $82 million.

In the March quarter ongoing other income was $8 million, an increase of $8 million from the prior quarter, as the December quarter was negatively impacted by a $4 million ineffective revenue hedge.

For the March quarter half of the other income is from net interest income and the other half is from foreign exchange hedge gains.

We recorded non-ongoing expenses totaling $108 million for the quarter.

This consists of a $7 million gain on currency fluctuations related to our accelerated tax planning strategy, and a $23 million charge associated with our restructuring plan, as outlined in our 8-K filing of March 19, 2009.

The majority of the restructuring charge was for severance payments related to headcount reductions, as well as asset impairments.

The facilities component of that restructuring plan and the resultant savings are still being determined, and will be communicated as soon as we have the information available.

In addition, we have recorded a non-cash expense of $89 million for the impairment of goodwill related to our Clean Product Group.

We will provide more detail on this goodwill impairment in a few minutes.

For the March quarter we had an ongoing tax expense of $16 million, which is an effective tax rate of negative 21.6%, outside the expected range of negative 7% to negative 12% of our operating loss.

The change in tax rate relative to our forecast range is primarily the result of a change in California tax law that reduces the estimated 2009 benefit of R&D tax credits by approximately $5 million.

While these changes will provide a long-term reduction in Lam's state tax obligation, the fiscal year 2009 impact is negative.

For the June quarter we anticipate a tax expense of approximately 10% of our operating loss.

GAAP loss per share was $1.58, and the ongoing loss was $0.71 per share, consistent with our guidance range, and based on a shared count of 125.6 million shares.

During the course of the last two quarters we have announced cost reductions totaling $40 million to $45 million per quarter versus our baseline quarter of September 2008, with approximately 60% to 70% of these reductions realized in operating expenses and the remainder in gross margin.

As of the March quarter, Lam has delivered approximately $25 million in operating expense savings.

And we are on track to deliver on the remainder of our overall commitment during the June quarter.

We are also on target to achieve the cost reductions committed in the gross margin line.

Substantively all of these cost reductions are manifest as a reduction of cash outflows.

Turning now to the March quarter balance sheet, cash and short-term investments, including restricted cash, totaled $806 million.

This balance represents a reduction of $312 million versus the December quarter, and is comprised of three key elements -- debt repayment, nonoperating cash impacts, and net cash used by operations.

Debt repayment of $240 million was paid in anticipation of not meeting certain loan covenants.

Due to the current credit environment, the refinancing options available at the time were unattractive from a cost perspective.

And given the strength of our balance sheet, we made the decision to pay off the loan, reduce our interest expense, and avoid higher interest payments and other costs related to refinancing.

Nonoperating cash impacts of $48 million included investments related to the acquisition of worldwide exclusive sales distribution and manufacturing rights for the dry bevel productline of SEZ, Inc.

And net cash used by operations of $24 million consist of $52 million in cash losses from ongoing operations, other operational-related cast disbursements of $59 million, both offset by a reduction in Accounts Receivable of $87 million.

Looking to June, we expect reduced cash operating losses from ongoing operations.

And based on current expense levels, we expect that Lam will be neutral on an operating cash basis at revenue levels of approximately $250 million, depending upon the product and business mix and business cycle timing.

Accounts receivable days outstanding increased to 103 days, up from 93 days in the December quarter, predominately as a result of revenues declining more quickly than receivables.

As we discussed in last quarter's call, we expect that higher DSO levels will continue for the next few quarters.

Inventory was down 3% for the March quarter.

Mainstream manufacturing and spares inventory performed as expected; however, several components of our inventory do not vary directly with business levels.

Lam's balance sheet strength allows us to make critical inventory investments despite the business cycle, which help us in maintaining customer trust, securing marketshare, and responding quickly to short-term customer requirements.

These investments include spares inventory placed at customer sites to enable rapid response to machine down conditions, evaluation and joint development project units placed with customers to promote increases in our marketshare, and flexible factory inventory to allow us to respond to short lead time system and upgrade opportunities.

These investments, coupled with lower overall business volumes, contributed to lower inventory performance of 2.1 turns in the March quarter versus a December quarter performance of 2.7 turns.

At the end of March Lam's deferred revenue balance was $44 million.

And this amount does not include shipments to Japanese customers of $8 million that will revenue in future quarters.

Returning now to the goodwill impairment, as a result of a combination of factors, including the current economic environment, a sustained decline in the market valuation for semiconductor capital equipment companies, and a decline in Lam's operating results, we have concluded that the fair value of our Clean Product Group is below its carrying value.

As a result, the Company has recorded a non-cash goodwill impairment charge of approximately $89 million in the March quarter.

The goodwill impairment charge is based on our current best estimate, and will be finalized in the 10-Q report.

Total capital expenditures were approximately $11 million for the quarter, and depreciation and amortization increased slightly to $20 million.

Employee headcount declined to approximately 2,950 from 3,300 employees, and reflects the additional headcount reductions announced last month.

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

Now to Steve's comments.

Good afternoon everyone.

Clearly this past quarter was a challenging period for the Company, as spending on semiconductor capital equipment continued to decline sharply.

As Ernie outlined in his comments, there are many moving parts to our P&L and balance sheet this quarter.

Bottom line is our actions in the March quarter are consistent with maintaining our strong balance sheet, applying appropriate conservative financial decisions to valuations and reserves, and aggressively reducing costs, while deploying our working capital in a variety of ways that strengthen our position for the future.

I think at this point in the cycle we remain in very good position with our customers and our balance sheet.

As many of you have likely observed and heard over the past several weeks, the semiconductor industry is starting to see signs that, as Intel said recently, the worst is behind us, as memory price is improving, and utilization at the foundries is increasing, particularly at the leading-edge 65 and 45 nanometer nodes.

However, visibility into shipment requests over the next few quarters is still very limited.

And as a result, we remain cost is about when any significant resumption in equipment spending will occur.

We continue to expect that most tool purchases for 2009 will be limited to select technology buys, and as such, wafer fab equipment spending is likely to be in the range of $10 billion for the year, consistent with the outlook we shared at our last conference call.

Going forward, we expect that several memory customers will upgrade DRAM lines, and put in place new wafer start capacity at the 5X technology node.

NAND Flash companies will move forward with their conversion to the 3X node.

And advanced logic and foundry capacity is slowly being put in place at the 4X node.

Microprocessor companies will move to 3X over the course of 2009 and early 2010.

All of this activity should result in sustainable shipment levels higher than the March quarter, on a going forward basis, but as to how much higher is still unclear beyond the June quarter.

At this point in the cycle we are sharply focused on the development of next-generation technology solutions, working side-by-side with our customers.

We have deployed working capital and placed a significant number of etch and clean joint development projects and evaluation units at our customers' locations to enable successful penetration of new applications, and defend and grow our existing marketshare.

In etch we believe there are a number of areas of opportunity for marketshare growth in both conductor and dielectric.

In conductor these opportunities include critical front end DRAM applications at the 5X node and below, double patterning where NAND customers are evaluating patterning schemes at the 3X node, and high-k/metal gate in advanced logic and foundry for the 3X nodes and below.

In dielectric etch we expect to see increased opportunity for marketshare growth as the transition of memory from aluminum to copper is completed at 5X DRAM for most memory manufacturers.

In 2008 we won and successfully defended several back end of the line coppers selections in both DRAM and NAND, as well as some high aspect ratio contact selections for the leading-edge technology nodes in both of these memory segments.

As the remaining conversion to copper at 5X occurs, it remains a sizable revenue and important marketshare growth opportunity for us.

With new application wins in both dielectric and conductor, we expect our application-based marketshare to increase by a few more percentage points in 2009 over 2008.

In the clean market we see growth opportunities due to the continued adoption of single wafer clean in front end of the line applications in 5X and 4X DRAM and NAND, 4X and 3X logic in foundry.

And of particular note is the introduction of single wafer clean tools in both front and back end of the line for NAND manufacturing at the 4X and 3X technology node.

We are focused on penetrating these new applications in memory and logic, as well as defending and growing our strong back end of the line and backside clean positions in our key foundry and logic customers.

Some foundry and logic customers will make tool selection decisions in 2009 for 4X and 3X technology nodes, and we are working closely with most of them to win P2R positions in high-k/metal gate, front end of the line post-ash and post-etch clean applications.

We are positioning the spin-based DV-Prime to penetrate post-ash and post-edge cleans in the front end of the line, as well as defend and expand our back end of the line backside clean market position through the DV-Prime's ability to efficiently recycle and deliver superior residue, particle removal and drying performance in a high productivity, low-cost tool.

This capability delivers the customer an excellent value proposition relative to traditional batch and other single wafer clean solutions.

And we have a number of JDPs in evals with major customers focused on these opportunities.

The traditional strength of our spin clean has been in the copper back end of the line and backside cleans for foundry and logic.

And we continue to work with our key customers to defend and grow these positions, and win new applications at advanced technology nodes.

Our linear clean technology is gaining momentum in front end of the line memory logic and foundry, demonstrating strength specifically in front end of the line post-edge cleans where problems of material selectivity, critical dimension control, and damage control continues to demand technically superior solutions.

We are developing with our customers innovative new technologies in particle removal, drying and multistep processes that enhance device yield at 4X and 3X nodes and below.

We continue to make progress winning qualifications for new applications of most memory and logic customers.

dry bevel cleaning continues to be an important area for us in the single wafer clean market.

And we see relative strength in this area for our clean product portfolio during this downturn.

These tools offer yield improvements to our customers by removing defect sources at the wafer edge, as well as increasing the number of good die at the wafer edge.

Our bevel clean technology is significantly penetrated at the leading-edge in all IC areas, memory, foundry, logic and microprocessor.

And we think that metal hard mask structures in the future will demand more bevel cleaning steps at each successive node.

As we move to next generation nodes in all these device categories, we expect the dry bevel cleaning market to grow from an estimated SAM of approximately $50 million today to approximately $150 million over the next few years.

And we expect to have a very significant marketshare in this segment.

With our spin, linear and dry bevel clean products, we provide our customers the broadest, most versatile, high productivity yield-enhancing clean solutions in the industry.

Relative to the near-term environment our June guidance is as follows.

Shipments between $200 million and $230 million.

Revenues between $180 million and $210 million.

Gross margin at 30%, plus or minus 2 percentage points.

Operating loss of $60 million, plus or minus $10 million.

And a loss per share of $0.50, plus or minus $0.10.

As we carefully manage the P&L expense structure of the Company during this downturn, we are even more focused on managing our cash flow.

While we have increased our focus on limiting cash outlays to maintain a strong balance sheet, we are using, and will continue to use, our balance sheet to make strategic investments in product development and solution support at the customer interface, and to make expenditures in other areas that strengthen our position in the downturn, so we can optimize our profitable growth in the upturn.

With that, Ernie and I will now take your questions.

(Operator Instructions).

Gary Hsueh, Oppenheimer World Markets.

If you look at your WFE and you are kind of maintaining that $10 billion number for 2009, and so the gist of it, I guess, is over the last quarter in terms of spending the environment has stabilized.

Now that $10 billion coming down from roughly $22 billion, implies a decline of 57%.

Would it be reasonable to start modeling shipment numbers somewhere down 50% to 55%?

Or is there a little bit more upside in the second half that we are somehow missing that might be baking in 2010 CapEx?

Can you just help me out with the trajectory of the shipment number coming off a pretty good guidance here for the June quarter?

I think that -- our view is that wafer fab equipment was probably somewhere around $20 billion, maybe $21 billion.

So we're down 50%, 52% as opposed to 57%.

But having said that, I think that when you look at the industry, and we are no exception, our shipments were down significantly in the March quarter.

And our guidance is for shipments to be up at the midpoint about 35% in June.

I think if you look at where we are going to have to be, if the wafer fab equipment environment is somewhere around $10 billion -- and again, I would emphasize around $10 billion, because it is a very fluid environment -- that you would expect that shipments would begin to stabilize somewhere in that vicinity as we go forward for the rest of the year, if in fact it ends up being a $10 billion wafer fab equipment market.

Jim Covello, Goldman Sachs.

Thanks so much for taking the questions.

Steve, a couple of questions.

First, just relative to one of the concerns that folks have had about the memory segment in particular -- just the idea that there is a whole bunch of dormant or latent capacity out there, that as prices improve than memory, all this latent capacity, would come flooding back online and prevent anybody from making orders.

Given the increase in shipments that you're talking about, and the increase in orders that Novell has talked about tonight, I would imagine that doesn't seem to be the case.

But can you address that a little bit?

I think that when we look at the latent memory capacity, it is all sitting at 8X and 7X.

And even with memory pricing stabilizing, I think you're right, that if those that have shut down the 7X and 8X turn those lines back on, and brought those 512 megabit devices into the market, you could probably see that pricing environment tank.

I think that where we look at the investments is that in DRAM on the leading-edge the spending is going to be on 5X.

And the effort is going to be on getting 2 gigabit DRAMs up and in production.

I think that part of what is in our wafer fab equipment forecast is an expectation that some of that 300 millimeter tool sets that are idled in these 7X and 8X lines are going to come into the market in some of the noncritical areas for etch.

And would serve as somewhat of a damper in terms of how much spending you would normally have to spend to create a 5X line.

We are spending a lot of time with our customers looking at, where do they want to upgrade, where do they want to refurb and have us take older equipment and upgrade it, where do they want to just buy upgrade chambers, and where are they going to buy new systems.

So there is quite a bit of activity, but it is centered around the customers who have the money with which to go and make the 5X investment.

I think we all saw recently where -- just today I think Hynix talked about raising $900 million or something with an equity play.

So I think companies are trying to figure out how to get access to capital, so they can make the conversion to 5X.

But I think that the speed at which that is going to occur is going to be muted by what is the demand environment.

And that is kind of the control point for everything is that if the general economic situation keeps demand suppressed, we are not going to see significant amounts of investment at 5X, even though people -- even if they have the money and they have the technology capability to make the move.

Brett Hodess, Banc of America Merrill Lynch.

Steve, you just articulated that some of the areas you're working with the customers specifically on the technology upgrade.

I am wondering if you could talk specifically about for Lam how much dollar etch opportunity is there if a company wants to grade a fab from 7X to 5X on the DRAMs side, or if you are upgrading a 6X to a 4X on the foundry side, what -- can you size for us what those opportunities look like in dollars?

We have been doing some work in that area.

When you go from -- when you want to upgrade, you need to -- it depends on how much equipment is available.

But I think that if we took a 10,000 wafer start 5X line, and you are going to convert and reuse some 7X, I think you can -- you could probably spend about -- let me check here.

I am looking at some data right now.

Relative to what you would have to spend, if you bought all new equipment, you would have to spend -- for 10,000 wafers starts, you would have to spend $95 million, $90 million in wafer fab equipment.

If you upgrade and you have a significant reuse, you might only have to spend $10 million.

And so one of the things that -- and maybe it is a little bit more than that -- but I think that -- we look at it from an etch standpoint.

So I can't speak to what it means for Lippo and CBD and PVD and all those tools.

But the etchers that are in a 7X line or in a 6X line going to convert to 5X or 4X, there is a fairly good reuse capability on the noncritical steps.

There is also an ability to upgrade a chamber so that, while it may have been running a critical 7X application, you could actually upgrade that chamber for a reasonable price to operate at 4X.

I think that what it does is that it can significantly reduce, if the equipment is available, how much expenditures go into that 5X line.

But I don't think that there is that much conversion that will occur for too long, because we are in one of those periods right now that, because the overall demand is low, companies can actually take a 7X line that would normally be running full out, and they could go and shut it down and convert it.

Or they could reduce the wafer starts through that and convert some of those pieces of equipment.

But as we see demand start to pick up, the ability to convert some of those tools to 5X or to 4X NAND is going to be reduced.

Tim Arcuri, Citigroup.

First, I just wanted to get stock-based compensation by line item.

And then second of all, I had a question about the structural margin profile.

I am just looking back to the last big downturn.

I am looking back to like 2003.

At a pretty similar revenue level, the margins are bottoming about 1,000 basis points lower this time.

Granted, you have a clean business this time.

But now you are back up to the $200 million revenue range, and the margins are again about 1,000 basis points below where they were back then as well.

Certainly that gap is going to close going forward, but I am wondering relative to what you did, you know, you look at the old Company versus the new Company, what sort of structural margin degradation, if any, should we think about as the revenue grows in the next few years?

This is Ernie.

The total equity-based compensation for the March quarter was about $13 million.

And relative to the question on margin, we will turn it over to Steve.

I will answer part of it, and then Ernie will fill in some were details.

I think there is a couple of important points to make.

In the last downturn in 2001 to 2003, systems business fell about 70% to 80%.

Where we are right now in the cycle is the systems business for Lam is down closer to 85% to 90%.

And it is probably worse for us in this quarter because, with no revenues out of Intel, and Intel being a spender at this part of the cycle, it is exacerbated a little bit.

Our model was built on the ability to handle a peak to trough decline of 40%, and with the ability to very rapidly shed volume-based variable cost.

And with still running 60% of the volume of where we were, the absorption on the fixed cost components is still significant.

When you drop to 80% to 90% down in terms of your system shipments, then the absorption factor is a very small component of fixed infrastructure, make up a very high percentage of what your cost structure is.

So we have that issue in the factory.

And Ernie can detail some of that out, if we want to.

But we also have it in the field, because we charge all of our field service costs to above the line.

So as we have made specific decisions to, in essence, treat certain aspects of our field service headcount as strategic, we are running underabsorbed relative to the amount of installation warranty and service contract work that would normally have them fully occupied.

That impacts our margin as well.

And in this downturn, again, because of the magnitude of the current hole in system shipments, we are having an overstated short-term effect of underabsorption in the field that contributes to the margin numbers that you are seeing.

Relative to the field and factory-based absorption, as Steve said, the ongoing decline, in particular in the systems business, in this cycle versus the 2001 cycle is putting a strain, or underabsorbing the factory to a far more significant degree on a percentage basis.

I would be happy to go through some of the specific numbers in a follow-up call.

But I believe that part of the structural margin change that you have articulated here is clearly the result of this unprecedented low level of system shipments relative to even a greatly reduced level of factory spending.

And in fact the same phenomenon in the field with far less systems available for warranty and installation, and our decision to keep that strategic resource in the field to remain able to provide critical service needs to our customers.

Satya Kumar, Credit Suisse.

Steve, just a little bit of math.

If wafer fab equipment stayed flat at $10 billion, I think you commented your shipment would stay flat at these levels.

If equipment spending is flat at these levels by our math, you could upgrade the memory factories over the next two years, especially looking at some of the commentary from the NAND guys saying the shrink to 3X will not be as capital intensive.

Right now some of these NAND factories are running at 70%, and several DRAM factories are running at 50%.

So the question is, if you spend $10 billion, by my math you can get 25% bit supply growth in DRAM, and about 50% in NAND Flash.

Do these numbers sound about right to you?

And do we have a structural need for a CapEx higher than that through the next two years, unless demand has to be higher than that?

Let me comment on a couple of things.

We are saying that we still believe that the wafer fab equipment environment is likely to be around $10 billion.

And then relative to some of the questions about what does that mean for shipments, what I said was clearly shipments have to come up from the levels that we saw in March.

We have clearly indicated that they will be a 35% in June.

Where they go in the second half, if it is going to be a $10 billion wafer fab equipment year, would have to in that vicinity or even higher.

Now having said that, when we look at 25% bit growth in DRAM, which is in the vicinity of what we think, 50% bit growth in NAND in the near term, then it is consistent with that kind of spending.

It is also consistent with some assumptions that there is upgrades and refurbished equipment that go into the 5X lines, particularly in DRAM.

In NAND not as high of a reuse orientation.

But as to what that means out in the future, I think it all depends on, one, where are we economically?

Where is GDP growth?

What is happening with new products, new applications that have what degree of memory intensity?

What kind of progress do we make for a 64 gig NAND Flash devices, and what is their costs?

And therefore what is the SSD penetration?

And the reality is, my ability to forecast that is no better, or perhaps significantly worse, than anybody else.

So we are not worried, and we are not focused on what is going to happen or not happen in 2010.

Our focus is staying close to our customers, helping them develop the solutions that they need in etch and clean at 5 and 4 and 3X.

Winning those marketshare opportunities.

Managing the cash flow, and being prepared that if this lasts for while, we can handle it.

If it comes back sooner, we will be able to respond to that as well.

Wes Twigg, Pacific Crest Securities.

Just a couple of quick questions.

The first one actually may not be quite so quick, but it is on double patterning.

You mentioned that as an opportunity in NAND, the ramp in double patterning processes this year.

I am just wondering, most of them are using a spacer-based process.

I am wondering if you could give us an idea of what the edge opportunity, since that is a more etch intensive process?

There is both what we call noncritical double patterning opportunities, and there are critical opportunities in the double patterning arena.

How much that grows the etch opportunity in a NAND 3X fab versus a 4X, we will have to go and get that data for you, because I don't have it off the top of my head.

Clearly, it represents a number of additional wafer passes.

And so we are looking at how we can win a significant marketshare position in doing that.

It looks like maybe three critical layers, six noncritical layers.

And then depending upon the throughputs associated with that.

We can get you what the sizing is.

It will vary by customer, because each customer has a slightly different approach, but we can get back to you with that.

Atif Malik, Morgan Stanley.

My first question is, Steve, thanks for providing the math on the incremental DRAM WFE spending of $10 million per wafer starts per month from 6, 7 to 5X.

Our analysis suggests that Lippo is about 60% of that incremental WFE spending.

And I just want to know what you think is the etch opportunity for you guys?

My second question is, if I look at the Gartner data on marketshare for single wafer processors, Lam [SSE] marketshare was 31% in '07.

It dropped to 17% in '08.

And DNS increased by 20% points in '08.

I just want to know what is driving that?

Has that been a mix factor, a customer spending factor?

And what is it going to take to reverse that marketshare trend in 2009?

Let's talk about etch first, and then will we talk about clean.

So relative to spending, we think that wafer fab equipment spending in 2009 for memory -- this is both NAND and RAM -- it is going to be somewhere around $4 billion to $4.5 billion.

So etch is going to be somewhere around 10% of that.

Because in this environment instead of being its typical 12% or 13%, it will be somewhat lower than that.

Then when you end up with reuse, etc., you probably end up in a situation where you can end up with $300 million and something spent in etch.

So we would expect that we would have more than 50% of that, because that is what our marketshare is in both DRAM and NAND.

And so it all depends on really how much of a reuse strategy is, but I guess the total is really, what, 400 -- maybe it is closer to $400 million to $450 million in total spending for DRAM and NAND in 2009.

Figure us for about 50%.

So relative to clean, the clean market is moving in many, many different directions, and it extremely confusing, even for those of us that are in it, let alone those who are trying to figure out what is going on.

Let me give you a little bit of a perspective.

One, is what is classified at single wafer clean?

In the Gartner Dataquest numbers they included in single wafer clean scrubber cleans, which are going through a conversion process.

And we don't include that in the single wafer clean, because that is not in our served market.

We are not in the scrubber business.

So that is one.

Two, the big spenders in 2008 were Samsung and Intel.

And the third big spender was Toshiba.

Our historical marketshare position has been in foundry and logic, and largely back end of the line, and there was very little spending in those market segments.

If you take those customers, it is no secret, we have no position at Intel.

We have a position at Samsung.

It has been impacted by decisions by Samsung to buy from a Korean vendor, SMX in Korea.

And we are working closely with Samsung now to win back some of that marketshare and win some new applications in the front end of the line.

Toshiba, the other big spender, we have had some positions in the back end of the line.

We were also impacted in 2007 and 2008 with Toshiba deciding to go with the Japanese supplier, [Shiberuar] for some of the noncritical, which impacted our marketshare.

So the bottom line is, one, the size of the single wafer market from our perspective is a little over $400 million.

We have about 25% share of that market.

If I take Toshiba and Intel out of the market -- not that we aren't going to and aren't working with them, but given that they were $200 million of the market in 2008, our marketshare in the rest of the market is 50%.

So two customers are clearly having a big impact on us in an environment where the spending is very depressed and very low in logic and foundry, and in the back end of the line and in backside cleaning, which is where our strength is.

As we go forward, our focus is to make sure that we defend our back end of the line and backside, as well as look for opportunities to grow that in places like Samsung and Toshiba.

And so that when spending for backside and back end of the line cleaning resumes, we will be able to leverage that market position.

In the meantime, work with all of these customers relative to the critical front end of the line needs.

And then there is no question that with a small group of customers really dominating the amount of spending, and those customers largely being historical DNS customers, our marketshare in the short term doesn't look very good as a function of mix and as a function of the economic environment and where CapEx is being spent.

Stephen Chin, UBS.

Steve, you know the cost cutting that the Company has made, what kind of incremental gross margin do you think we should think about modeling going forward as Lam's business slowly improves here?

Will it be similar to this 60% level that we saw this quarter, and is that a sustainable ramp going forward?

This is Ernie.

About one-third, give or take, of the cost reductions that we talked about are reflected in gross margin.

And I think as you are thinking about margins going forward, you would certainly see an improvement from that, but you would also see an improvement from further absorption of the factory and field resources.

So I think that both items will be significant contributors to the gross margin improvement and provide leverage going forward.

Ben Pang, Caris & Co.

In terms of the DRAM transition, are all the design wins done for 5X at this point?

Is there any new design wins for 5X?

I think most of them are done, but there are still a number of them that are actually being finalized.

What you really have is that there is a low level of 5X production running today, maybe 20,000 or 30,000 wafer starts.

By the end of the year we expect there will be over 100,000 wafer starts running on 5X.

So you really -- you haven't won it until you have shipped it.

And so I think that, while it is fair to say that the decisions have been made, it is a competitive environment out there, and people work hard to try and change decisions and reverse decisions.

But for the most part, they are done.

And we will now begin to see in 2009 how that market [shay] plays out as the orders are placed and the products are actually shipped.

Wes Twigg, Pacific Crest Securities.

I just had a quick follow-up, two parts.

One on the spacer piece that we were talking about.

I am just wondering if you had actually won new business already related to that in the NAND segment, space or double patterning?

And then related to the copper transition, are you seeing any pushout or delay in terms of adoption of copper processes from the two big Korean DRAM customers?

Relative to double patterning wins on NAND, yes, we have won.

I'm not going to specify where and which customers, but the answer is, yes.

As it relates to copper, I am not aware of any delay or pushout as it relates to the transition from aluminum to copper that is going on with the big DRAM guys.

C.J. Muse, Barclays Capital.

I guess two-part question.

First, could you share the makeup of the June shipments in terms of NAND, DRAM, foundry and logic?

And then looking to your OpEx, it looks like implied in your guidance is roughly going down from $130 million in March to around $115 million in June.

I was hoping you could help me understand what that would look like exiting calendar of '09 in the December quarter.

Thank you.

Relative to the OpEx, we don't expect to need to increase OpEx to any significant degree, provided business volumes stay within the range that we would expect.

You will obviously have some seasonal fluctuations related to normal patterns, but relative to the overall OpEx levels, we expect to be able to largely contain them at that level throughout the rest of this year.

So relative to the breakout of our shipments in the March quarter we had -- 56% were memory.

And of that, as Ernie said, 20% something of that -- 22% of that was NAND.

And then 28%, logic and 15%, foundry.

So in logic we put logic, microprocessor and other.

And then 16% foundry.

In the June quarter, just kind of from a what is likely to occur is the mix will be pretty similar.

I think it will be a little stronger in NAND because we have got some activity from one of the big NAND players.

It will be slightly down in logic.

But it will be up significantly in foundry.

I think most people have been reading about what is happening on the foundry leading-edge where utilizations on 65 and 45 are way up.

It is no secret that TSMC has announced that they are going to be expanding their 4X wafer starts.

And as a major supplier, and with high marketshare with that customer, we are participating in that.

And we will be shipping into the foundry segment -- or we anticipate shipping into the foundry segment in the June quarter.

Tim Arcuri, Citigroup.

Just to quickly follow up on my last question.

Steve, last cycle you were in the high 40%'s to even 50%, kind of at the $400 million level.

I am wondering once you get through these underabsorption period here, and you kind of absorb these costs, what can we expect margins to be at the $400 million revenue level, as revenue ramps this time versus last time?

We could probably figure that out pretty quick, but to be perfectly honest with you, given that we have just been bouncing around at $175 million, and we are deciding whether to throw a party because we are forecasting our revenue to be over $200 million -- or around 200 million, I should say.

It is like $195 million plus or minus $15 million.

I really haven't thought about what is going to happen at 400.

But we can model that, and we will get it back to you.

But the message I would say to you is we expect -- and what we are seeing in the environment right now, we have looked very hard at this.

As competitive at what is going on in etch, there is essentially very little to no pricing erosion and margin erosion at the product margin level.

There are some issues about mix in terms of which customer in which quarter, because there is a little bit of a different margin profile.

So we expect etch, once the volumes get up and the absorption is improve will be right back in the vicinity where it has been.

We look at clean, and the reality is that we think that there is a very strong likelihood, and certainly our target is, that our margins in the clean business are going to be in the 40s.

Now whether they are low 40s or mid-40s, it depends on what the timing is and when the revenues get back to that level, because we've got some work to do with the product option architecture.

We've got some work to do in terms of getting our outsourced suppliers up to speed, getting our volumes up and our costs down.

But we really expect that we are going to be largely in the same vicinity.

But we can model that out and share with you.

But for the most part, you should expect we are going to be right in the same vicinity.

Patrick Ho, Stifel Nicolaus.

Steve, in terms of the revenue uptick in June, characterize how it breaks down between a pickup in the services and spare parts of your business, as well as the pickup in, I would assume, technology by shipments during the quarter.

I think from a percentage standpoint, revenue -- and we can talk about revenue, but revenue is a lagging indicator relative to -- because we do revenue on acceptance.

So I think a better way to maybe characterize what is going on in real time is the shipments environment.

And the shipments environment for the June quarter is going to be up somewhere around 30% or 40%.

And the services, spare parts type activity is up more around 20% from where they were.

So the combined is up 35%.

But clearly from a percentage basis, most of the shipments growth is actually occurring as a function of systems shipment improvements.

Thank you.

Ladies and gentlemen, that does conclude today's question-and-answer session.

At this time I would like to turn it over to management for any closing comments.

Thank you.

I think we all recognize that over the last couple of quarters we have clearly gone through a period Of really almost unprecedented levels of drop in demand for our customers.

Clearly their profitability issues resulted in almost a near shutdown, even with the strongest and most technically advanced and cost effective customers.

And that made for a very tough December and March quarter.

What we are seeing is that the strongest customers are moving forward with what we would call the technology investments.

These are not capacity expansion investments.

They are next technology node investments.

And clearly when you look at the foundry, they are both technology node investments, but they make them as a function of they have got customer demand at the 4X that wants those wafer starts.

A memory company or a NAND company wants to move to the next technology node so they can produce the 64 gigabit in NAND, or they can get to the 2 gigabit in a DRAM, and so they have cost and density issues that are motivating them to move.

I think we are now in that phase where the spending for technology buys, which have largely been suspended for three, four, five months, have now begun.

And so that is why my comments were, there is an element of sustainability that exists once you get into that mode.

Where we go from here, when we move to stronger levels of spending, that is clearly not something that we have visibility to, and feel like we can comfortably forecast.

But I do think that it is far more likely that the worst of what we have been dealing with is behind us.

And now the focus is marketshare, penetrations, product positioning, cash management, ultimately getting to cash neutral, ultimately getting our P&L where we would like it to be.

I think, as I mentioned in my earlier comments, I think we are in good shape with the customers, good shape with our balance sheet.

And we are feeling very good about our ability to compete successfully, and make the expenditures strategically and tactically that we need.

And we expect to come out of this environment stronger and more capable than ever before.

So thank you for your attention.

And we will look forward to talking to you again in the future.

Thank you for joining us today.

Please been advised that a webcast of today's call will be available on our website late this afternoon.

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