美光科技 (MU) 2007 Q4 法說會逐字稿

Good afternoon.

My name is Anthony and I will be your conference facilitator today.

At this time I would like to welcome everyone to the Micron Technology fourth quarter and fiscal year-end 2007 financial release conference call.

(OPERATOR INSTRUCTIONS).

It is now my pleasure to turn the floor over to your host, Mr.

Kipp Bedard.

Sir, you may begin your conference.

Welcome to Micron Technology's fourth quarter and fiscal year-end 2007 financial release conference call.

On the call today is Steve Appleton, Chairman and CEO; Mark Durcan, President and Chief Operating Officer; Bill Stover, Vice President of Finance and CFO; and Mike Sadler, Vice President of Worldwide Sales.

This conference call, including audio and slides, is also available on Micron's website at Micron.com.

If you have not had an opportunity to review the fourth quarter and fiscal year-end 2007 financial press release, it is also available on our website, again at Micron.com.

Our call will be approximately 60 minutes in length.

There will be a taped audio replay of this call available later this evening at 5.30 PM Mountain time.

You can reach that by dialing 973-341-3080, a confirmation code of 9266785.

This replay will run through Tuesday, October 9, 2007 at 5.30 PM Mountain time.

A webcast replay will be available on the Company's website until October 2, 2008.

We encourage you to monitor our website at Micron.com throughout the quarter for the most current information on the Company, including information on the various financial conferences that we will be attending.

Please note the following Safe Harbor statement.

During the course of this meeting we may make projections or other forward-looking statements regarding future events or the future financial performance of the Company and the industry.

We wish to caution you that such statements are predictions and that actual events or results may differ materially.

We refer you to the documents the Company files on a consolidated basis from time to time with the Securities and Exchange Commission, specifically the Company's most recent Form 10-K and Form 10-Q.

These documents contain and identify important factors that could cause the actual results for the Company on a consolidated basis to differ materially from those contained in our projections or forward-looking statements.

These certain factors can be found in the Investor Relations section of Micron's website.

Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance or achievements.

We are under no duty to update any of the forward-looking statements after the date of the presentation to conform these statements to actual results.

With that, I would now like to turn the call over to Mr.

Bill Stover.

For those listeners may not yet have access to our press release, let me go over a summary of financial results for the fourth quarter and fiscal year which ended August 30.

For the fourth quarter net sales totaled $1.44 billion.

That is an 11% increase over the preceding quarter.

And the Company recorded a net loss of $158 million or $0.21 per diluted share.

In spite of 15% declines in quarter over quarter DRAM and NAND pricing, the Company achieved a 4 percentage point improvement in overall gross margins, largely as a result of the cost reductions being realized on NAND Flash memory.

We're very pleased with our NAND progress and note that NAND Flash realized a positive gross margin in the fourth quarter.

For the fiscal year, net sales totaled $5.7 billion, and the Company recorded a net loss of $320 million, or $0.42 per diluted share.

Gross margin for the year came in at 19%, reflective of the ASP environment, and for the bulk of the year the early stage of ramp of NAND devices.

DRAM gross margins improved several percentage points comparing 2007 to 2006.

Our cost reduction efforts in SG&A and in research and development are beginning to be reflected in a lower cost structure.

We expect the quarterly runrate in 2008 for SG&A to be between $125 million and $135 million.

SG&A in the fourth quarter of 2007 reflected a higher level of legal expenses in preparation for several upcoming trials.

Stock-related non-cash expenses totaled $44 million in 2007, of which approximately $21 million are reflected in SG&A.

A strategic move to partner for NAND development has been noticeably beneficial in containing R&D expense.

We expect R&D to run between $165 million and $175 million per quarter for 2008.

Our forecast for capital spending in 2008 is $2.5 billion, with the vast majority of the dollars being spent on 300 mm wafer fab tooling in support of NAND and DRAM processing.

Approximately $750 million of that '08 capital spending is expected to come from partner contribution.

The growth across the balance sheet is reflective of the expansion of operations at our IM Flash ventures and TECH Semiconductor.

The Company is well positioned with $2.6 billion of cash and investments.

I will close there and turn commentary over to Mike.

With solid demand for computers, mobile phones and a variety of consumer electronics products, we had strong quarter over quarter revenue growth in excess of 10%.

Contributing to the topline growth were a 20% plus NAND Flash trade ASP increase, and as referenced in the press release, strong sequential bit growth in NAND and DRAM production and shipments.

Our capacity and R&D investments are resulting in further output growth, and we are in the mode of taking share in the memory market.

As we layer in additional capacity and technology advancements, we will be in this mode for the foreseeable future.

Much of the output growth is attributed to productivity increases.

In fiscal Q4 we again reduced costs at a faster pace than market prices declined.

Due to industrywide oversupply in the memory markets however, the absolute price levels have been under pressure.

Market demand for computing products continues to be robust amidst typical seasonal demand strength heading out of calendar Q3 and into calendar Q4.

We have seen a moderate flattening in memory content per system growth from the PC OEMs encountered -- PC OEMs encountered Q3.

I attribute this primarily to concerns from our customers that memory prices were increasing early in the summer.

As a consequence, OEMs took steps to maintain reasonable control of system build and material costs, and used memory content as a lever to achieve that.

End customers then turned to third-party module makers and spot players to fill memory needs, resulting in extraordinary amount of buying and selling activity in the spot markets.

With memory market prices coming back down in late summer, our OEM customers are adopting a more aggressive posture towards system memory content.

On the DRAM product front we continue to bolster the product portfolio with new offerings in the low power and low voltage arenas, as well as new high-density and high-performance DRAM products.

These products, including a newly introduced memory interface known as DDR 3, are aimed at increasing Micron's exposure to the network and communications infrastructure, mobile phone, automotive, server and PC gaming market segments.

We have internal infrastructure to support customers in these segments, and it is only natural for us to strengthen the portfolio so we can bring more value to customers in these spaces.

Our 78 nm 1 gb DDR2 chip continues to be the high-volume product for the desktop and notebook market segments.

This device has reached yield maturity at our 300 mm fab in Virginia, and we are commencing the production ramp of the same chip on 300 mm wafers in our Singapore facility.

Much of our output growth in fiscal year '08 is coming from the 300 mm production ramp in Singapore, and as a result we expect further improvements in the DRAM cost per bit profile.

In the NAND Flash arena the vast majority of our IMFT production output today is centered on our 8 gb device running on a 72 nm process on 300 mm wafers in Virginia.

This device is being marketed to a variety of consumer electronics and memory card providers, and is in fact used to populate our own branded Lexar Flash cards and drives.

We're in the early stages of volume production of the 50 nm 16 gb MLC device in Virginia.

For the same die size we get twice as many bits relative to the 72 nm version, resulting in a substantial manufacturing cost reduction.

On the heels of the Virginia 50 nm transition is the initial production ramp of the new 300 mm facility in Lehi, Utah.

With these lease and scale and technology deployment we couldn't be happier about our improvements in competitiveness, as evidenced by the 40% sequential cost per bit reduction in fiscal Q4.

Going forward we are well-equipped to be a leader in the high volume, high density MLC markets.

We continue with development and production of a variety of SLC devices at the Boise fab for high-performance requirements in computing, automotive, communications and other specialty markets.

Among those specialty opportunities today is the solid-state drive market.

We are rolling out a family of SATA interface solid-state drives later this year.

This is a logical extension to our existing portfolio as the targeted customer base and channels are already in existence here at Micron.

There are compelling advantages associated with solid-state storage, such as performance, ruggedness and power conservation.

As we continue to implement cost reductions through both scale and technology advancements, solid-state drives will reach mass market penetration and become a tremendous driver of demand for NAND Flash devices.

The mobile phone market is growing nicely in 2007, and from a market segment standpoint continues to be the primary driver of demand for our CMOS Image Sensors.

As I have mentioned in past calls, we pride ourselves on having the highest performance pixels in the industry.

As a result, we are taking share in the high end of the market.

This was evidenced in fiscal Q4 with the quarter over quarter increase in imaging revenues despite a moderate unit shipment decrease.

We have worked diligently at establishing a balanced customer base in the mobile phones space for our image sensors.

The results are encouraging as the revenue pipeline is loaded with design wins at multiple customers.

Beyond the mobile phone market, our 8 megapixel sensors are now featured in a variety of digital still cameras that are now available to consumers.

This is a space formally locked up with DTV technology, and we believe today's CMOS digital still cameras are just the tip of the iceberg in terms of further CMOS market penetration.

We're at the forefront of this effort.

Without question, the market environment is challenging, and this is attributed to the industrywide supply growth that is outstripping demand.

With recently announced cutbacks in capital spending and anecdotal input from capital equipment providers, this cycle will be self-correcting, just as has been the case in the past.

Fundamentally demand is quite strong for our products.

Customers like our product portfolio and roadmap.

And we're becoming much more cost-efficient in manufacturing these products.

With that I will turn it back to Kipp for a Q&A.

We would now like to take questions from callers.

Just a reminder, if you are using a speakerphone, please pick up the handset when asking a question so we can hear you clearly.

With that we would like to open up the phone lines.

(OPERATOR INSTRUCTIONS).

John Pitzer, Credit Suisse.

I have got a couple of questions here.

I guess if both DRAM and NAND pricing were flat from current levels, what would you expect them to be for the full first quarter?

And then I have a couple of follow-ups.

Sure.

This is Mike speaking.

If both -- if pricing were flat for the balance of the quarter and the mix were relatively representative of what we already shipped for the first four weeks of the quarter, DRAM pricing would be down about 10%.

And NAND Flash trade pricing, which is essentially our selling price of Micron branded products into market, would be down about 25%.

Mike, you talked about bits per box [slowing] here in the calendar third quarter.

What is the expectation going forward?

Do you think price elasticity will kick back in and we will start to see configurations move up, or do you think PC OEMs are going to be pretty satisfied at these configs?

I think we've got a couple of things working here.

Number one, I think we are seeing -- in fact seeing some benefit of price elasticity as the OEM customers are feeling more confident about the absolute price levels, as well as the direction of DRAM pricing.

Feel comfortable about having a stronger memory content profile in their systems.

And number two, we've got the overriding effect of further penetration of Vista operating system, which tends to require more memory than its predecessors as well.

So we feel pretty comfortable about the near-term, and as well as the intermediate, the longer-term about acceleration of memory content.

Last question from me.

If you looked at the May quarter at your segment reporting, the CMOS Image Sensor business actually lost money on an operating base.

I'm kind of curious if you were able to break even this quarter there or not?

We will put that in the K, like we have done in past quarters.

So if you will bear with us on that, we will get that updated for you here in the next, I think, week or two until we get that posted.

Glen Yeung of Citigroup.

Can we just start with what your bit production was in Q3, and what you think it will be in Q4 for NAND and DRAM?

On a relative basis, we were as we said -- the growth was up about 15% on the DRAM side and -- sorry, up about -- be up low double-digits going forward on DRAM and up about 40% Q over Q on the NAND side.

That is NAND into the foreseeable future, or is that just for the next quarter?

That will be an average going forward over the next three or four quarters.

Then can you also look at the same thing just on a cost basis?

You get normally cost down each quarter.

How do you think that might track over the next few quarters for each of NAND and DRAM?

DRAM I would say down high single digits on a go forward basis, and NAND probably down midteens.

Thanks for that.

I wanted to ask you, given the pricing environment that we're in, what your sense as to what happens to all the 200 mm capacity that you and the industry have, and whether or not you think that is something may come off flying a little more quickly than it might otherwise have happened?

On the 200 mm side we've got a number of different dynamics going on.

First of all, we have options to move the capacity around in a couple of places where we think are more productive than potentially the DRAM is allocated to at the moment.

We have transitions going from some of the DRAM capacity into 15 nm NAND.

And also, as Mike alluded to, we have pretty strong performance on the design wins side in Imaging right now, so we anticipate that there will be some transition of DRAM wafers into Imaging going forward.

Okay.

Is that at any different rate than might normally happened?

What do you mean by that?

Are you accelerating the movement from DRAM to Image Sensor, given the fact that DRAM pricing is down so much?

I don't think that is the driving factor.

The other thing to keep in mind is we don't have that much advanced DRAM running on 200 mm.

Most of that stuff is what you would think of as legacy, which is either the synchronous or the pseudostatic, etc.

So I think is a little bit of -- almost, in fact I think all of our really advanced stuff, the 50 nm and the 78 nm is all running on 300 mm at Virginia and Utah.

And Singapore is in the conversion right now to 200 mm.

They are part way through it.

When you say are we accelerating, there's not a lot left actually.

The Imaging business though, as you know, has been down.

And what we are seeing is that should start to grow again.

Obviously it grew this quarter, and it should grow again from here as we move forward.

If that happens, yes, we will convert more of it over it to it.

Just one last quick one in I heard the comment that overall PC demand looks pretty solid.

I really want to get a sense from you, if you could just provide more color on how you see fourth quarter PC component outlook shaping up?

We have to translate that to memory demand, of course, because that is our area of expertise.

If I take a cross-section of our customers, calendar Q4 versus calendar Q3, in the PC space we're looking at a low of 5% sequential growth in bit consumption and a high of 20% growth in sequential bit consumption, and kind of a weighted average as being close to midteens, 15% or something like that, which is typical seasonal.

It is nothing extraordinary, but I think typical seasonal.

Jim Covello, Goldman Sachs.

Two quick questions.

On the DRAM supply, can you guys give us some kind of estimate of either how much you think DRAM CapEx needs to come down to fix the excess supply situation, and/or when that might occur?

And then my follow-up would just be on the solid-state drives for NAND, do you have any forecast for 2008 of what you might expect for some unit penetration on the notebook side, as some kind of example?

Thanks so much.

I will take the first one in terms of CapEx coming down.

As you know, it is always a combination of supply and demand that has its impact on pricing.

And so you are really asking us to do both.

I think with your background in capital equipment you have got a pretty good handle on how that is starting to shape up for '08.

And it is way too early for us to try and estimate what bit growth is going to be in '08 from the demand side.

So I'm going to defer that one for awhile.

I think Mike is getting ready to handle the SoC question.

I can't reference penetration percentage, but our expectations in 2008 are that the demand for solid-state drives will be somewhere between 5 and 10 million units.

And that the memory -- the storage that that represents in terms of the percentage of the NAND market, it is still in single digits.

Probably 3, 4, 5% of the total market in 2008.

We do believe that within three or four years, say by 2011, we think that the demand for solid-state drives will be upwards of 50 million units per year, probably closer to 60 million units per year.

And that the number of bits consumed in those drives will be somewhere in the neighborhood of 20 to 25% of the total NAND market.

That is why we're so bullish on this particular application segment.

Shawn Webster, JP Morgan.

Can I circle up -- backup to your fiscal Q4 DRAM production and NAND production?

I didn't catch that.

We actually ran over 20% NAND core DRAM bit production quarter to quarter, and we were well over our 50% that we guided to in the prior quarter on NAND.

I'm sorry, the 20% number, that is DRAM production?

That's correct.

Core DRAM production.

Total DRAM up around 20% and then NAND, 50 to 60%.

Both were up above our -- both NAND -- DRAM was up over 20 and NAND was up over 50.

We're not going to give specifics, but we're up in both categories above that guidance.

Then what are you seeing in terms of DRAM channel inventory now?

I don't have good visibility into the channel inventory.

I can tell you of course our own inventory is in the neighborhood of four weeks -- somewhere around the neighborhood of four weeks of supply, and plus or minus depending on the particular technology, which is about where we would like to be.

Then I don't think I caught it either.

Did you throw out what you guys saw in terms of your bit per box or an estimate of what your customers are loading for your -- for DRAM consumption?

I can prefer you to the slide deck.

We did show that number.

It is in the slides?

Yes.

Then can you -- is there any update that you can give us -- you guys have been speaking before on the desire to restructure or sell off your sensor business, is there any update there for us?

Yes, we haven't really been that specific because we're still exploring the alternatives.

I think the way to characterize it is similar to what I have characterized it in the past, which is we have every intent to continue to manufacture the product at the silicon level.

It is really what we're exploring are other -- is our [down] business model with which this product makes its way to market.

In which case, should we do a partnership of some type where we break that out?

It is a little bit -- a different characterization than selling off our Imaging business, because we intend to continue to manufacture it.

It is really a piece of it that is modeling.

Nothing is going to happen this quarter, but we will keep you updated as we move into the next quarter.

Thanks.

That makes sense.

In terms of the -- you had a very -- it seemed like a small inventory write-down.

Was most of that DRAM or what was the composition of that?

We identified that that is both DRAM and NAND.

Greater than 50% of it, the majority of it was on the DRAM side, a little bit on the NAND side, and not on the Imaging side.

And that is just mark-to-market stuff?

You got it.

Doug Freedman, AMTECH Research.

Just to start with, any guidance on what you think depreciation for '08 is going to look like?

It will run between $500 million and $540 million, increasing over the year.

Here's $2.2 billion for your '08 estimate.

So no change there.

Then you commented that you are in a position to be taking share.

Can you discuss if we are looking at -- in the past you have been trying to steer some of your share more towards the specialty market.

Any progress that you are having there, and any discussions you can have as far as where you're targeting your share gains from?

Sure.

We are doing -- our objective is to take as much share as we possibly can in what you would characterize as the specialty markets.

Specifically we would refer to those as the server market, the mobile phone market, the networking and communications infrastructure market, automotive, medical, industrial and consumer.

For us the desktop and notebook market primarily drives asset utilization.

And that is our share is lowest I think in all those market segments that I preferred to.

With respect to us growing share, our objective would be to place as many bits as we possibly can in the premium or the specialty markets, and then utilize the notebook and the desktop markets to drive the asset utilization for us.

In the next fiscal year we're probably looking at becoming more significant in the notebook and the desktop area relative to where we have been in the last year, and continuing to do the best job we can in those other market segments.

Can you talk a little bit about what is going on the NAND in the retail channel with your Lexar brand, and what your outlook there is, how that this is running, any update you can offer?

First of all, let me may be characterize the Lexar operation I think with respect to where it was a year ago.

I think despite in contradiction to what maybe some people have said in the media, we actually like Lexar.

It has gone very well.

Remember the first objective for the year was to get it -- think of it get it focused, cleaned up and back to a level where it wasn't losing the kind of money it was when we acquired it.

That has been accomplished.

As we look forward to the market, I think we have a greater presence actually in some channels and maybe less of a presence in other channels by virtue of us targeting the bits into what we think are better opportunities for us.

So it has actually gone pretty well.

I think now we're looking at trying to retake market share in the retail channel, but in addition to that continuing to use more and more of our own output.

Remember that a year ago when we acquired them, we didn't really have any spare NAND.

Most of it was already committed to a customer base.

And as we have grown our output, we have been able to more and more allocate into that model and obviously leverage that entire relationship.

So I think you'll see a growing presence over the next year in that channel.

Just to add to that.

In our last fiscal quarter roughly 10% of our own silicon output went through the Lexar channel, and that is going to expand to about 15% in the current quarter.

And as we move forward our output ramp, the slope of our output ramp on NAND Flash is very significant and we're going to be putting more -- putting more and more of that output through the Lexar channel.

Tim Luke, Lehman Brothers.

Could you give us some sense as to clarify what you said on cost reduction going forward the expectation in the NAND side?

And maybe any expectation of timelines that you may be looking at to break even and offer you the gross margins maybe more constructionally in the NAND arena?

This is Kipp.

We have been guiding previously at around 20% sequential cost reductions on NAND.

Basically we have been giving you a 3 to 4 quarter rolling forecast.

As you have seen from the last couple of periods, we have come down significantly better than that.

For our reference in the press release, for example, 44 to 40% in the fourth fiscal quarter.

So our next update for you is we think we will bring cost down sequentially for the next three to four quarters in the midteens range on NAND.

When that crosses over in the future, is really a combination again as I mentioned to Jim of price and cost reductions.

So I will let you model the ASP reductions, along with these kind of sequential 14 -- or excuse me, sequential midteen kind of reductions.

There will be quarters where it is substantially higher than that by the way.

Just like you have seen in the last couple quarters, but that is a good average for the next four quarters you are trying to model your year.

In our opening comments, or in the opening finance comments, we acknowledged that NAND Flash had a positive gross margin for the fourth quarter.

Tristan Gerra, Robert W.

Baird.

What percentage of your total NAND Flash output Lexar could use next year?

And how do you capitalize on some of the controller IP you have acquired with Lexar.

Let me start with the IP piece first.

We have -- we obviously think that it has been of greater value to us.

Clearly we did announce the Toshiba settlement historically.

And of course it continues to be part of our portfolio.

So we have a very, very strong portfolio in this arena.

And we believe that is in part why we have been so strong in this arena.

And we haven't really been in litigation on the NAND front.

I will just leave it at that.

We think that part has been very valuable.

Obviously, it is hard to quantify in many ways.

But we're pretty happy with it.

Mike, do you want to get the other?

Sure.

I can't be too specific on the percentage of our output that we will move through Lexar.

I think it is worth mentioning that Lexar has favorable supply agreements with some other players in the NAND Flash -- from a NAND Flash supply-side, and we want to use that to complement our own silicon going forward.

It is safe to say that the percentage of our output going through Lexar will increase quarter over quarter for the next several quarters.

But I don't have a particular target in mind with respect to a percentage of our overall output that we want to move through Lexar.

Again, we want to take advantage of the favorable supply agreements that were in place when we acquired the company, and use those to complement our own flash output.

Then a quick second one.

The market share gains that you're mentioning in core tier 1, I am assuming have to do with a cost advantage that you currently have.

If you could talk about this, and how long do you think you can extend that potential advantage?

It has primarily to do with the bit growth, but I think you're right, it all comes back to a technology advantage that is giving us a strong quarter over quarter bit growth.

And the transition of -- or the completion of the ramp at [NTV], which we are now substantially through, and the transition of the TECH facility from 200 mm to 300 mm.

John Lau, Jefferies & Co.

I want to circle back on the CMOS sensors and give us another update again on the CMOS sensors.

Have you seen NAND start to pick up again for cell phones and digital cameras?

I'm missed that.

I apologize.

We are seeing some pretty strong acceleration in the current quarter on I will say unforecasted business, particularly on the low end.

Low end for us in the Sensor area would be in the VGA and 1 megapixel area.

And I call it spot business.

It is business that we hadn't forecasted.

I'm not sure if I could attribute it to a recovery by one particular customer or seasonal demand increasing significantly.

But we are seeing some pretty good positive movements in the current quarter, particularly in the low end, that are unforecasted.

But nothing significant in the higher 3 or 4 megapixel.

But did you mention anything about the higher than 4 megapixel in the digital camera area again?

We have a series of design wins with our 8 megapixel sensor in the stand-alone DSC market that -- actually sales are doing quite well there.

But we have done a pretty good job -- actually a very good job of forecasting the demand there in terms of working with our customers.

And we put enough work in process in the pipeline essentially we don't have any shortages.

I would suggest that that is also the case in the high-end mobile phone area.

On a five megapixel we have got a very high-profile design win with a five megapixel sensor.

And the same situation, we have done a pretty good job of assessing the demand situation and we don't have any shortages.

So we're right on top of things there on the high-end.

Aaron Husock, Morgan Stanley.

I was just -- sticking with Imaging for a minute.

I'm looking at the August quarter.

Can you talk a little bit about what you're seeing in the like-for-like pricing environment?

I'm assuming that like-for-like pricing was still down in the quarter, despite your blended ASPs being up a little bit more than 5%.

Can you quantify that a little bit, and talk about if you see those like-for-like pricing environment changing at all this quarter?

I don't have the data at my fingertips such that I can give you a meaningful -- quantify that meaningfully.

But it is safe to say that, other than the very new products, which would be the high-end probably 5 and 8 megapixel, certainly the pricing was down quarter over quarter for, as you call it, the like for like product.

Do you know -- was it down as much as we have seen historically, or are are you seeing any signs of stabilization in terms of like-for-like pricing, either in the August quarter or the current quarter?

Nothing particularly alarming.

Of course, if I compared it to the memory market on a relative basis, it didn't even move.

Then I guess just any plans to move to 300 mm on the Imaging side?

I think that -- the answer first of all is yes.

It is just a matter of when.

I don't think it is if.

And it all has to do with timing on the economic model.

We will be, I think, in a position within the next probably two years where we start pulling out 200 mm equipment out of the current fabs as it transitions to the next generation mode.

The model that we have been running is utilizing that equipment to produce Imaging, because Imaging even today, by anybody, is still running on a one or two generation lagging process technology compared to DRAM or NAND.

As a result, we of course can use that equipment, extend its life for quite some time.

We could use it for memory at a very cost-effective way.

And so we will go to 300 mm as we start transitioning the current memory fabs away from some the equipment they have now, obviously in favor of more advanced process equipment.

It is not if, it is just when.

We don't really think that happens for probably two maybe three years.

Let me just make a comment on that.

Some companies are talking about doing that today, but the reality is you can't even use the advanced process known on this type of product because it just doesn't need it.

It doesn't utilize it.

So you've got a pretty high cost for doing that.

Until you can use the process mode that is equivalent -- that is 10 million equivalent, it doesn't make a lot of sense to do that.

And for us that time period is really still a couple of years away.

[Bob Gujavardi], Deutsche Bank.

Just thinking a little more long-term, is there any update on IM Flash facility in Singapore?

When do you expect that to be a contributor to your capacity?

We are running pretty much on schedule, slightly ahead of schedule actually, so we still anticipate first silicon starts in the second half of '08, and start contributing meaningfully in calendar 09.

Great.

And just to help me size it, can we think of the capacity at IM Flash Singapore similar to Lehi, or is it a little bigger, a little smaller?

Slightly bigger.

Think of it in terms of 50,000 to 60,000 per month.

Great.

Can you talk a little bit about -- you mentioned your [FSD] efforts, is it safe to assume you are doing some collaboration there?

That segment seems to be getting a bit crowded and it seems pretty R&D intensive.

Are you -- Intel at their IDF announced a FSD offering, are you guys kind of amortizing your R&D investment there a little bit, or is it --?

Obviously we've had some own announcements around some FSD products we are either just introduced or walking towards.

I would say there are a number of players out there on both sides of the fence.

There is both the (inaudible) guys and there are the component guys.

And I think a number of us are trying to figure out if there is partnerships that will work in that space or not.

We're just trying to work through it.

That is all I can really say at this point.

And I certainly don't think we are alone in terms of others trying to look at what might be the best model in that market.

Fair enough.

Final question, you alluded to this, but it looks like your NAND cost reductions came in again really much better than expectations, above 40% versus expectation around 20%.

How should we think of that?

Is that MLC penetration, just higher MLCs rate or better yields or just more aggressive down to 50 nanometer?

How is the best way to think about that?

I think the results from Q4 are primarily indicative of faster ramp and higher yields we had modeled in about the MLC penetration that we actually achieved.

About the technology ramps have been pretty strong.

Going forward we will be more dependant on the 50 nanometer ramp, and that also is proceeding as per schedule.

Can you ballpark the MLC mix for me?

I would say it is in excess of 50%, actually it is probably closer to 75 for Micron, and Intel will have to speak for themselves.

Gurinder Kalra, Bear Stearns.

Do you have any view on industrywide supply growth for NAND 2008 as compared to '07?

We have seen some reports.

We have seen anything from probably 130% to 150% as it shapes up today with current CapEx announcements.

We know that a lot of our competitors will update those CapEx projections in '08 over the next couple of months, and that will give all the experts out there trying to put these market expectations together a look at a new refreshed set.

But right now we are seeing about 130 to 150%.

Now based on your estimates do you think the supply growth in '08 is equivalent to the supply growth we saw in '07?

It is coming in very similar to '07 if it hits that rate.

There's a lot of movement now in CapEx, a lot of discussions, Mike referenced about some CapEx push outs, etc.

We will just have to see how those all shake up here over the announcements over the next couple of months.

The second question is, given your NAND supply growth, which as you say this could be (inaudible) on a quarterly basis, and given what we are seeing from others, is there any reason that Q1, or the early part of '08 we will not see very significant price declines on the NAND side?

It has always been our policy not to predict pricing, because as Mike mentioned, it is very difficult to do.

So we will leave it to you experts to try to predict the supply/demand balance.

[Manish Goyal, Crest Investments].

The question is on NAND again.

Your gross margins are significantly below your large competitors.

Roughly what scale do you need when your NAND cost structure will be comparable to the leading vendors and your gross margins are comparable?

Can you size that, like either in wafer starts or market share, what you need to be comparable?

I have recently heard Mark speak about the fact that technologically now we believe we are on particularly with the largest NAND supplier today.

We're probably a couple quarters behind catching the others.

As we have mentioned in the past, we are in a significant ramp.

I think Mark highlighted just a few minutes ago that this past quarter again up substantial -- almost double the cost reductions we expected came from faster ramp in Lehi, mature yields, and the beginning move of 50.

Going through the next year it is largely due to the 50 nm, the completion of Lehi.

And then as Mark mentioned, sometime the latter part of '08 into '09 a continuation of Singapore.

Let me just add on to what Kipp just said.

We -- first of all, let me tell you, we think we're probably there in terms of the Virginia facility.

We track this data pretty good.

It is now optimized, or pretty closely to being optimized, in its NAND production.

Obviously it is making DRAM as well.

We think we are every bit as cost competitive there.

It is just a reality that when you're in the middle of a ramp it is just not as cost competitive.

And so Utah is in the middle of a ramp right now.

And obviously as that gets ramped and optimized we think that will become the case as well.

We actually think we have advantages in many ways in terms of our fixed nanometer and the die size and so forth.

When you say that it is obvious that our gross margin in NAND are lower than our competitors, first of all we don't disclose gross margins in NAND, although Bill just told you that it was positive for us.

But even having said that, in our particular scenario where we are under the ramp, we have to look at it facility by facility, and we think Virginia is right here.

Can I ask just a couple of more follow-ups?

So 12 months out towards the end of 2008, where do you think will be your either wafer starts from NAND or maybe your aspiration for market share in NAND?

Maybe I could -- I probably won't go there in wafer starts.

But in terms of -- unless someone else here has the data.

But in terms of what we think happens on market share, the data is out there.

I think the expectation obviously is that we will grow market share.

I think most of the data today says we're in somewhere in the 5 to 7%.

And I think by the time we get into '08, which would mean Utah will have been ramped, you'll probably see that double.

Just a last question on DRAM.

When you think about increasing DRAM content per PC, are you thinking more in terms of either 2 gig box is moving to 3 gig, or are you thinking that the boxes are going to 1 gig or 1.5 gig move to 2 gigs.

You can be the leading PC vendor.

Nobody is really advertising PCs with more than 2 (inaudible), and I was just wondering how you were thinking about the increase in content?

I think the biggest driver of the increase in content for the next couple of quarters is going to be the move of boxes from less than 2 gigabytes to 2 gigabytes.

So in other words, the 1 to 1.5 gigabyte machines today moving to 2 gigabyte standard.

[Bill Buton, Induce CNL Capital].

A quick question.

What was the MLC mix the last quarter?

I don't have that right at my hands.

It is in excess of of 50%.

My second question is would the ASP drop have a lot to do with more shift to MLC versus SLC during the quarter?

No, the ASPs for NAND are a little bit complicated because when we report them as dropping in the press release, we're blending in the NAND that we sell to our partner as something that approximates cost.

So our trade ASPs are actually up and the blended is down.

I think we have time for one last call please.

Bill Dezellem, Tieton Capital Management.

We have a couple of questions.

First of all, circling back to the Image Sensor market, you had mentioned that at some point in the future it may be cost-effective for the industry to shift to 300 mm.

But not thinking about cost, but thinking about the product quality, is there any improvement, or opportunities for improving, pixel density or anything interesting on that front that may take place when 300 mm is running Image Sensors?

I don't think it is today.

I think that that has to do with primarily advancing the process technology, and actually you can advance the process technology from where Image Sensors are today equally as well on both 200 or 300 mm.

That is where you'll primarily be able to get I think advantage in the -- think of it as the qualitative measurement of the pixel performance and Image Sensor performance.

There's nothing inherent at the moment that has 300 mm over 200 mm.

But I do think eventually it will come down to a cost equation.

And as you advance the process technology itself, ultimately you won't be able to have 200 mm equipment if that won't do it, because the geometries will get below what really is out there in the mass market for the equipment.

That is helpful.

And then secondarily on an entirely different topic, the restructuring last quarter you announced that you would begin a restructuring process that, if we remembered correctly, was going to take several quarters.

Would you please discuss what it was that you did in the fourth quarter, and what you still see remaining to be done?

And maybe how much of a benefit your existing actions have had, and excluding the charge that you announced in the quarter.

And what you would anticipate future actions to benefit?

Of course, we aren't to try to quantify it in dollars for you, because we have stayed away from that in large part because it is unpredictable exactly how that shows up.

But I think you should think in terms of two different really categories, because we have to bifurcate it to some degree.

We have benefits that will show up in the cost of goods sold.

And we will have benefits that will show up in the -- think of it as the overhaul overhead expenses.

I am going to not really speak too much about the cost of goods sold, because that gets mixed in to the things that we do in terms of output, yields, etc.

And in terms of trying to streamline what we have more in the operations, and that will show up in the cost per bit measure report.

In terms of the operating expenses, we have been pretty focused on the two pieces, which is the SG&A and the R&D.

Clearly we believe, as I noted at the last analyst conference, that we have to get competitive on our total overhead, operating expenses to the benchmarks that are out there in the industry.

If you look at the benchmarks in the industry, those in combination need to be somewhere between realistically 14 to 16% of revenues.

And it depends on whether the revenues are decreasing or increasing based on prices.

So it moves around a little bit on you, but that is what needs to happen.

There are really kind of our parts to that.

And the simplest part is where you just have to simply not do things that maybe you were doing before that aren't as critical.

In other words, get focused on how you are expending that money.

But when it comes to personnel, there are really three things that happen there.

There is position eliminations, which we have already spoken about to the media.

There are things where we are trying to do outsourcing, which is in other words trying to change the business model, where someone might be more effective at than we are, more cost effective.

And another piece is continuing to move some of what we do in the U.S.

to our other operations that exist elsewhere in the world.

Realistically the position eliminations have already taken place.

We had a restructuring charge this quarter for in large part that.

There may be a little bit of that that we carry forward, but probably not a lot.

And the rest of it is really round the outsourcing, or the moved to other operations around the world.

And it is going to -- it will take a couple of quarters for us to figure out what makes the most sense in terms of optimizing the model on outsourcing.

And then actually the move to other operations around the world will probably be something more along the lines of 12 to 18 months, because as an example, we may have a particular package what we are doing in the U.S., but it is going to be and of life in six months or nine months.

So doesn't really make a heck of a lot of sense to try to move that.

You just finish it out and let that transition when it is done.

That is helpful.

Thank you.

Thanks Bill.

With that we would like to thank everyone for participating on that call today.

If you will please bear with me, I need to repeat the Safe Harbor protection language.

During the course of this call we may have made forward-looking statements regarding the Company and the industry.

These particular forward-looking statements, and all other statements that may have been made on this call that are not historical facts, are subject to a number of risks and uncertainties, and actual results may differ materially.

For information on the important factors that may cause actual results to differ materially, please refer to our filings with the SEC, including the Company's most recent 10-Q and 10-K.

Thank you.

This concludes today's Micron Technology's fourth quarter and fiscal year end 2007 financial release conference call.

You may now disconnect.