美光科技 (MU) 2006 Q4 法說會逐字稿

Good afternoon.

My name is Lindsey, and I'll be your conference operator today.

At this time, I would like to welcome everyone to the fourth quarter fiscal year end financial release conference call.

All lines have been placed on mute, to prevent any background noise.

After the speakers' remarks, there will be a question and answer period. [OPERATOR INSTRUCTIONS] Thank you.

It is now my pleasure to turn the floor over to your host, Mr. Kipp Bedard, Vice President of Investor Relations.

Sir, you may begin your conference.

Thank you, Lindsey.

I'd like to welcome everyone to Micron Technology's fiscal fourth quarter and fiscal year end 2006 financial release conference call.

On the call today remotely is Mr. Steve Appleton, Chairman, CEO, and President.

With me here in Boise is Mr. Mark Durcan, Chief Operating Officer, Bill Stover, Vice President of Finance and Chief Financial Officer, 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 2006 financial press release, it is also available on our website 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 p.m.

Mountain Time.

You may reach that by dialing 973-341-3080, with a confirmation code of 7930225.

This replay will run through Thursday, October 12th, 2006 at 5:30 p.m.

Mountain Time.

A webcast replay will be available on the Company's website until October 5th, 2007.

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.

During the course of this call, 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 to you 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 on the Company'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'd now like to turn the call over to Mr. Bill Stover.

Thanks, Kipp.

The results we'll be going over, just for confirmation, are the consolidated results, including our IM Flash Technologies joint venture with Intel, our TECH Semiconductor joint venture wafer fab in Singapore, and Lexar Media for the 9-week period subsequent to our acquisition of Lexar.

For those listeners who may not yet have access to our press release, let me go over our summary financial results for the fiscal year and the fourth quarter which ended August 31st.

For the fiscal year, net sales totalled $5.3 billion.

That's the second highest total in our Company history, and the Company recorded net income of 408 million, or $0.57 per diluted share.

We're certainly pleased with the third consecutive year of increasing net income.

The major themes benefiting this year's results are one, a strategic shift in product mix to take advantage of our CMOS image sensor business, which became the #1 supplier in 2006.

Two, acceleration of our NAND flash efforts through partnering with Intel, and our acquisition of Lexar Media, which positioned us with scale and expertise to be a significant competitor in the NAND market.

And, three, the strengthening of our overall financial condition and balance sheet, which in part is reflected in the net interest income reported for the year of $76 million.

As was mentioned on our second quarter conference call, operating income in 2006 benefited from the technology sale to Intel, which realized $230 million for the Company.

For the fourth quarter, net sales totalled 1.37 billion, and net income was 64 million, or $0.08 per diluted share.

You will notice that the fourth quarter results reflect a 24 million amount, or $0.03 per diluted share, for the noncontrolling interest in TECH Semiconductor.

The product mix which TECH runs has been primarily PCB run devices, which have enjoyed favorable market conditions recently.

The noncontrolling interest line reflects 57% of TECH's profitability for the quarter.

Gross margins for the year came in at 23%.

As we noted in our press release, the fourth quarter gross margin was impacted by legal settlement costs.

Absent the effects of the settlements, gross margin would have been approximately 3 percentage points higher for the fourth quarter, a slightly higher level in the third quarter, primarily on improved PC DRAM margins.

With regard to the legal settlements, I want to acknowledge that the $45 million amount includes our previously announced Tessera agreement, and the remainder all relate to contingency matters previously disclosed within the Company's filings.

As these are active matters, there's no further detail that we can provide.

Selling general and administrative expenses in most regards have increased proportional to the activities associated with IM Flash, TECH Semiconductor, and Lexar.

Lexar itself is a step-function increase of about $20 million per quarter.

Additionally, the fourth quarter of 2006 incurred a heavier than normal legal cost.

We expect the quarterly run rate in '07 to approximate $140 million.

Stock-related noncash expenses total 26 million in 2006, of which approximately $11 million are reflected in SG&A.

Total stock-related noncash expenses in 2007 are expected to approximately double the 2006 amount.

Unlike SG&A, which grew consistent with an increasing scale of operations, R&D expense stayed quite flat, even while the scope of R&D activities increased quite significantly.

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

We expect R&D to run at approximately $175 million per quarter in 2007.

It was a very strong year for the Company in generating cash flow from operations, in excess of $2 billion.

The improvement over 2005 reflects our successful diversification into specialty DRAM and CMOS image sensors, and the alliances we struck in the year.

Capital spending for 2006 came in at about 1.6 billion.

Our forecast for capital spending in 2007 is $4 billion.

That appears to be an increase from our last forecast for 2007, but merely reflects that 2006 came in under our last estimate.

Approximately 1.5 billion of that 2007 capital spending total is expected to come from partner contributions.

The second half of fiscal 2007 will really be the first period to see the noticeable benefit of increased volume coming from Manassas and Lehi ramps associated with our capital spend.

The size and scope of our operations expanded greatly in 2006, as we positioned the Company for market opportunities in NAND, and capitalized on our industry-leading position in CMOS image sensors.

Cash and investment balances increased 1.8 billion to 3.1 billion at year-end.

Total debt declined to less than $600 million.

The growth across the balance sheet is reflective of the consolidation of IM Flash, TECH Semiconductor, and Lexar.

The approximate $890 million purchase price for Lexar was allocated over the net assets acquired and liabilities assumed, and resulted in the recognition of approximately 490 million of goodwill.

I'll close there and turn our commentary over to Mike.

Thanks, Bill.

We couldn't ask for much more with respect to the strength of demand in the mobile communications, consumer electronics, and computing markets.

These markets account for the vast majority of demand for Micron semiconductor, memory, and imaging products.

As our customers roll into their peak selling seasons for cellular phones, digital TVs, and personal computers, we are braced for even stronger demand as we move through our fiscal Q1.

We have spoken extensively about our strategy to continue broadening the product portfolio and create exposure in a variety of markets.

This strategy is playing out nicely, as we make further inroads into these newer markets.

We have made a conscious effort to reduce exposure in the desktop and notebook markets with our commodity DRAM offering.

This journey continues but by no means are we withdrawing from these markets, as they continue to be the single largest consumers of DRAM.

At this point in time, it is gratifying to see the market performing so well in terms of DRAM consumption, particularly when many industry experts were predicting a sluggish environment due to Vista Operating System rollout delays from this year to next.

DDR2 is now entrenched as a mainstream memory solution for the PC market.

With strong demand, we have seen average selling prices increase over 20% from the lulls in mid-summer.

Prices are continuing to rise in the near term, and we are optimistic that this environment will sustain right into what is expected to be one of next year's major demand catalysts, the Vista Operating System.

The minimum effective DRAM configuration for Vista will be 1 gigabyte, with optimal performance in most applications achieved at 2 to 4 gigabytes.

This represents a significant step-up in memory content with typical PCs today configured with about 800 megabytes.

The server market is at the heart of our DRAM strategy, and we are holding a leadership position with a full line of high density chips and modules.

We are not sitting still here as evidenced by our continued early enablement of fully buffered DIMMs, and the recent introduction of the world's first 1-gigabyte DDR3 component.

In the current quarter, we are commencing volume production of our mainstream 1 gigabyte DDR2 components on 78-nanometer technology in the Virginia plant.

This is the DRAM industry's smallest 1-gigabyte die, and the best cost/performance solution in the market for high density applications.

We've seen an acceleration in NAND flash demand over the past 60 days.

Portable audio players are a primary driver, and as we all know there is a seasonal aspect to the timing of consumer electronics demand in any given calendar year.

NAND pricing has been on a roller coaster this year, and the recent trend has resulted in about a 30% average selling price bounce off the bottom, seen earlier this summer.

Beyond Christmas season, I would not be surprised to see things slow down, and we may see price pressure again in the early part of next year.

Over the intermediate to long-term we are bullish on the demand for NAND, particularly from the mobile phone and personal computer segments.

These markets are negligible drivers of today's demand in markets in which Micron is deeply entrenched with our complimentary products in our portfolio.

To that end, we are aggressively deploying new capital and advanced technology to extend our NAND flash market presence over the course of the next couple of years.

The mobile phone space has been the primary demand driver for our explosive image sensor business over the past couple of years.

Our mobile phone imaging business had another quarter of growth, with high pixel density imagers of 1 megapixel and greater, now accounting for well over one-half of our imagery revenues.

We believe this is a good indicator validating Micron's position as the recognized leader in image quality and in CMOS imaging technology.

In addition to imagers, the current offering of low-powered DRAM and cellular RAM are vital pieces of our mobile phone success.

With these products, we believe that today's mobile memory architecture has legs well into the future, with the market eventually migrating to a NAND plus low-powered DRAM architecture.

As one of only a few manufacturers with the capability to develop and manufacture these fully integrated solutions, our value proposition to the handset makers increases as this evolution continues.

We are helping our customers succeed, as opposed to competing with them, and believe that this is a significant differentiator for the company.

We couldn't be more pleased with our position in the mobile phone value chain given Micron's technology and product portfolio.

I'll also point out that our fiscal Q1 will be the first opportunity to fully incorporate Lexar's products and markets into the mix.

With over 70,000 retail outlets, top-performing products and leading controller technology and the intellectual property, we are excited for the prospects of this new segment, in both the retail space, as well as placement with our OEM customer base.

To wrap things up from my perspective everywhere I look in the market I see robust demand for Micron products in all end markets.

Our near-term efforts are focused on balancing product mix, supply, and allocations, to optimize customer satisfaction and profitability.

Our intermediate to longer term efforts are focused on technology development and capacity expansion, to bring even more value to the customer base.

As always, we appreciate your continued support and interest in the Company.

I'll turn it back over to Kipp.

Thank you, Mike.

With that, we'd like to take questions from callers. [OPERATOR INSTRUCTIONS] With that, please open up the phone lines.

At this time, I would like to remind everyone [OPERATOR INSTRUCTIONS] Your first question is coming from Glen Yeung from Citigroup.

You may go ahead.

Thanks.

Just a couple of questions, first one is if you look at gross margins in the quarter up slightly from the previous quarter, particularly if we take out the charges, can you talk about as we look into the November quarter, how that gross margin may trend?

What I'm specifically looking for is, are there changes in how you're going to break down your wafer starts, to optimize gross margins for the next quarter?

Let me break that up into a couple of questions, Glen.

We'll start with Mark and he can address some of the cost programs we have in play for next quarter and throughout the year.

And then we'll hand the second part of that over to Mike, to address perhaps what his expectations are for pricing.

So, Mark?

Thanks, Glen.

On the cost front, for DRAM we see relatively flat bit growth over the next couple of quarters, we should see that accelerate towards the second half of the year, driving some significant cost reductions in the core DRAM arena, maybe in the 30 to 40% regime.

As we look at NAND, obviously, we're pretty early into our NAND ramp.

And while we're pretty pleased with the way the technology is rolling out, we expect that it's going to take another few quarters, before we really start to see the benefit of the capacity we're putting in place.

So strong bit growth through the the year, leading to significant cost reductions throughout the year, that we will realize a much better position then as we roll out late into the fiscal year.

On the product mix front, Glen, if you look at our DRAM portfolio, we've got a pretty stable platform of business in the synchronous DRAM area,, and in the mobile DRAM and the other RAM products for mobile phones, industrial equipment, networking products, and so forth.

And that business is highly predictable.

So generally speaking that business is going to be stable to slightly up, as we roll into Q1 and into Q2.

The real challenge for from us a mix standpoint in the short term is on what we call the commodity DRAM front, and that's targeting the proper mix of DDR versus DDR2.

There are a variety of transitions going on in the marketplace, with respect to, in the computing area, that are driving DDR to DDR2 transitions from a demand standpoint, and of course the pricing is bouncing around, it's been quite active actually in the last month or so, and it looks like it's probably going to continue to be, certainly for the rest of the calendar year.

So what we're trying to do there is to optimize both customer satisfaction and probability, and the bias at this point, is probably towards a richer mix of DDR2 versus DDR, which is a little bit different than has been the case looking backward over the past couple of quarters.

A follow-up to your comments, could you comment a little bit on what you see in the next quarter for image sensors, and also what you think PC demand is going to look like in the fourth quarter, largely because we're in front of this -- fourth quarter?

Glen, we're having a little bit of a problem hearing you, are you asking for Mike's interpretation of the image sensor business in Q1?

Yes.

And also any thoughts on PC demand trends for the November quarter, given we're one quarter ahead of Vista?

First of all in the image sensor front, we're continuing to have a richer mix of I'll call them high pixel density imagers in the mobile phone space.

I mentioned that over half our business in Q4, in our Q4, was for one megapixel and greater sensors.

Actually, we're heading towards being the 2-megapixel sensor really being the sweet spot sensor, for I'll call it the midrange mobile phones.

And we're continuing to grow our, you know, our high pixel density chips as a percentage of our overall revenues.

VGA for us is still pretty substantial, but it's, now it's in the neighborhood of a quarter of our business, and most of the growth for us really is occurring in 2 and 3, and even 5 megapixel area.

On the PC DRAM demand front, things are on fire right now.

And we can't even come close to meeting the needs of our customers.

And the real hot product today is DDR2, and I presume that's primarily tied towards consumer demand for both desktop and notebook PCs.

So if there was any concern or worry about a lull or a lag in PC demand leading into the Vista introductions next year, it's not happening, at least it's not happening from our view.

So things are really looking positive on the PC front for us.

And when we get beyond the hot seasonal season in the January/February timeframe, our expectations and our customers' expectations that are Vista intro is going to be a real good demand kicker for them, for system shipments, for us both for system demand, as well as a big boost in terms of memory content.

So we are quite bullish on the prospects for our DRAM business in the computing environment.

Thanks a lot.

Your next question is coming from Mike Masdea from Credit Suisse.

You may go ahead.

You have some commentary on the profitability on the DRAM side, could you help us out a little bit, at least directionally what's going on with the CMOS and NAND side?

And Mike is that again from a cost perspective, market perspective?

Gross margin perspective, kind of overall.

So were you able to given the mix comments you had, and also given that you are covering more of your fixed costs on the NAND side, are you able to improve your profitability in both of those, or were you able to improve your profitability in both those segments?

Well, this is Mark, let me take the NAND piece of that.

Clearly, we're still building out our NAND product portfolio.

So if the you look at where we are today, profitability-wise we're underwater.

Now, as we move throughout the year and richen up the product mix, move more and more of our products into SLC, from SLC into MLC, we'll see some significant improvement there.

We're also putting a lot of new capital in place, and generally there's a lag as you bring the new capital online, and then try and get it fully utilized.

So we're anticipating as I said earlier, we're anticipating 50% bit growth quarter over quarter, as we move through the year, and we should see significant cost reductions associated with that going bit production.

And, Mike, with regard to the imaging business, we've been very pleased with that performance in the fourth quarter.

ASPs on average were actually up slightly, and margins up slightly, so we're very pleased with that position.

Just to follow up on the comments for the NAND side, so is it my understanding is that it has gotten a little bit worse first, and then it's going to start to get better, now that you have got the mix starting to get a little bit better, and the volume starting to get better?

Yes, it's definitely true, we're not where we want to be cost-wise on NAND today.

And that's going to get better as we move through the year.

Got it.

And then just kind of your commentary on inventory both from what you guys are doing with your own inventories, which I think were up a little bit this quarter, and then what you are seeing out there in the industry?

Yes, our inventory balance quarter over quarter, we were up in terms of total megabits in finished inventory, that was really isolated to our cellular REM products, which are our mobile phone memory products, and I would attribute that to some inventory accumulation on the part of our customers.

And we were up on Flash megabits as well, in terms of the, what you would think of as our traditional DRAM product portfolio, pretty flat quarter over quarter, and you know it's basically hand to mouth today.

We still do have some inventory on the cellular RAM, and we're working through the inventory on the Flash.

I don't think too much to be concerned about.

Based on the, I'll call it the emotion in some of the phone calls that we are having with with customers today, particularly on the PC DRAM front, I think it's pretty safe to say that there's no inventory out there at all.

Just to follow that one up real quick and given the money out there on the table, potentially given that market is pretty healthy right now, any kind of change in view?

I know you said you were going to hold bits kind of flattish.

But any opportunistic potential out there to make the customers happy and take a little bit of money in the short term?

Well, with respect to our strategy, which is to strengthen our position in the high-end computing space, specifically servers, we're not deviating from that at all.

And we're not deviating at all in terms of continuing our Flash ramp, and so forth.

Now, with respect to what we do with some of those commodity DRAM chips, you know, they're suitable for desktop and notebook applications, to try and take advantage of the opportunities we have, absolutely no question.

We are trying to push things as hard as we possibly can to increase prices.

And still support the same customers that we are working with.

So we can be as opportunistic as anybody else, and we're trying to do the best we can with that commodity DRAM portfolio.

Got it very helpful.

Thanks, guys.

Thanks Michael.

Next question, please?

Your next question is comes from Shawn Webster from JPMorgan.

You may go ahead.

Yes, good afternoon.

Can you talk a little bit on the gross margins and the $45 million impact?

I couldn't quite tell from your commentary, if is that a one-time event, or are you expecting other things in the future, and then I have a follow-up, please.

Yes.

The items that we're speaking to there in the fourth quarter included the $28 million Tessera agreement that we reached, and we previously announced.

The other items as we have indicated settlement of legal matters as a general rule, we do not anticipate those to be recurring.

We deal with them as case by case.

Okay.

Thank you.

And then turning to the demand environment, can you give us your, your bit consumption expectations for the end market for PCs in Q4, calendar Q4?

Sure, I'd be happy to, this is Mike speaking.

A cross-section of our big consumers, what you can imagine as are pretty good profits for the overall industry, bit consumption up about 12 to 15% quarter over quarter, calendar Q4 over calendar Q3.

On a year-over-year basis, Q4 versus Q4 of '05, up about 60%.

Is that the normal seasonal for you guys, in terms of bit demand in Q4?

Normal seasonal in Q4 would probably be on the lower end of that, maybe 10 to 12%.

So it's slightly higher than normal seasonal.

Okay.

And then can you tell us what your DRAM bit shipments and ASPs did quarter-over-quarter in August?

We can tell that you the bit shipments were up low single digits.

And for ASPs, I believe we were, go ahead, Mike.

ASPs I believe quarter over quarter were relatively flat.

On the DRAM.

Flat quarter over quarter?

Yes.

Yes.

If pricing stays flat where it is now, where do you think your average pricing could land in November for DRAM?

They'll be up for sure, and I would just have to say single digits at this point, based on our visibility.

Okay.

Thank you very much.

Your next question is coming from Joseph Osha from Merrill Lynch.

You may go ahead.

Hi, folks.

I guess as I understand it, if we back out the NAND Flash, and we back out Lexar, the DRAM gross margins on an apples-to-apples basis would have been up slightly.

Is that what you said?

That is correct, Joe.

Okay.

Is that, you know, again if I go back, this is I think for most of us, the best DRAM operating environment we've seen in several years.

Is that, by my estimation, a very small improvement in gross margins driven by the fact that you're sort of mix limited, and you're limited in terms of what you can do out of your 200 mill facilities, because you're supporting IMFT, or what's the story here?

Keep in mind two things, Joe.

On is that the price increases we actually saw that benefited us in the quarter came substantially in the last month.

And so, at most a third of the sales.

And secondly, don't forget the 45 million that Bill referenced, as a cost to goods sold cost that came in during the quarter.

I can see that.

But if you go back, I mean the last time, this, you had a DRAM business that was generating at one point 40% gross margins, and even in the November '04 quarter, 34% gross margins.

So I guess I'm a little flummoxed as to how the DRAM margins can be stalled here in the sort of mid to high 20s, despite this extraordinarily good operating environment.

I think again one other piece that Mark mentioned earlier, just to reiterate, is that we have put a lot of cost in place to date in building our businesses, and we're on the early phase of wafer output.

As you can appreciate pretty substantial wafer increases helped drive the fixed portion of that overhead cost to goods allocated on a wafer basis lower quite quickly.

And as Mark mentioned earlier, we've got a pretty good cost reduction profile getting heading in front of us here.

Okay.

So let me ask you, then, and I'm not going to, obviously I know you are not going to call pricing so I'll do it for you.

If pricing were to stay where it is now for the remainder of the quarter, which I believe comply to mid-single digits sequential increase per your frequent comments.

Can you get your DRAM gross margins to improve?

Definitely.

As Mike mentioned, we have got ASPs would be up high single digits, if they were flat from here, and as Mark mentioned, we're going to get increasingly positive cost reductions.

So that equates to margin expansion, you bet.

Okay.

Thank you.

You bet, Joe.

Your next question is coming from Tim Luke from Lehman Brothers.

You may go ahead.

I was wondering if could you give us a feel for how big a piece of the revenue the NAND portion may have been in the August period?

And then, going forward, it sounded from your commentary that we would wait until the middle of calendar '07, before we move into profitability in the NAND arena, or how should we view that kind of time around the transition?

With regard to the first part of the question, the NAND represented 9% of revenues in the fourth quarter.

And we also indicated that there's a quite limited amount of Lexar NAND revenue that was able to be recognized as the product that was in the channel.

By way of the purchase accounting, did not allow us to pick up the revenue associated with those final sales.

If you look at Q1 and estimate approximately $160 million of NAND Lexar-related sales, that's a fairly significant increase for the quarter.

With the incremental ramp also, both nonLexar related NAND?

I'm sorry, could you repeat that for me?

So you'll get a boost of 160 million associated with Lexar, and then you'll have a further mix increase in NAND, will be on top of that?

Without trying to call the ASPs, the bits grew over 50% last quarter, they are going to grow 50% quarter-over-quarter this quarter.

And with respect to your expense structure, how should we think about that developing going forward?

It looked like SG&A was somewhat higher than what one might have modeled this quarter in August.

Yes, you should, I think as Bill mentioned in his comments, you should model SG&A 170 to 175 range, or excuse me, R&D, 170 to 175 range.

And the SG&A, I believe he mentioned the 140 to 150 range.

140 looks like where it's going to run for quarters in '07.

And do you think this other income level of around 15 level, down from sort of 50 in the prior quarter, is where it's likely to be through fiscal '07, or how should we think about that?

I'm sorry, if your reference is to net interest income --

Yes, right, sorry.

It will range between 15 and 30 million per quarter over '07, with it being at the higher end during the early quarters by way of the cash balances, which we're maintaining right now, and with our capital spending profile that will come down over the year.

I would also acknowledge that that 140 million estimate on the SG&A, until we have the opportunity, which is really a 6 to 9-month period of integrating Lexar, we really don't have an opportunity to strip much cost out of where there may be some duplications, but about 6 months from now we'll get on Micron systems, and we'll have an opportunity to do some reduction of costs.

Could I just get back to the question with respect to the timeline for the transition towards profitability in NAND, just any flavor or color there, about what sort of timeline we should be thinking about?

Not really, Tim.

One of the reasons we're doing that is we're going to stay away from trying to predict ASPs for you.

We know, as Mark stated, we have a very substantial second half fiscal '07 continuing through second half calendar '07 cost reduction profile.

We've talked in prior analysts' meetings, we expect nothing short of 60 to 80% cost reductions through the the fiscal year, and we're going to leave it to you to determine what you think the ASPs will be, and when you think those two lines intersect, that's when we'll get there.

But you are [inaudible] the second half of the fiscal year, in terms of seeing the cost reductions?

Well, the more significant ones, but obviously if we're increasing bits quarter to quarter 50%, you're seeing pretty substantial cost reductions right now, especially when layered on the fact that the fixed overhead cost of adding additional wafers and equipment is there, and the wafers are just now starting to come out of the fabs in an increasing way.

Thank you, Kipp.

You bet.

In your next question is coming from Gus Richard from First Albany Capital.

You may go ahead.

Thanks for taking the question.

When you mentioned that the sequential bit consumptions would be up 12 to 15%, a little bit above normal seasonal trends, is the increase coming from box loading, or strong unit demand?

It's, I don't know, to be honest with you.

I would speculate it's probably coming more from unit demand than box loading, given where we are in the season.

But we haven't dissected the granularity of the demand to be honest with you.

Okay.

And then essentially you're looking at these percents year-over-year growth, Q4 to Q4, isn't that an acceleration of bit consumption?

60% year-over-year Q4 versus Q4 would be an acceleration of bit consumption, that's correct.

What we model in, and I believe what in general the industry has modeled in for DRAM demand growth from the computing environment, is somewhere in the range of 40 to 50%.

So, yes, things, the numbers would suggest that things are hotter than what had been expected, and I can tell you that the environment feels that way, as well.

Great.

Thanks a lot.

Your next question is coming from Titus Menzies from Jefferies and Company. you may go ahead.

You may go ahead.

Hello, can you hear me?

Yes, we can.

This is Titus Menzies and Jeff Reed, on behalf of John Lau.

I can ask maybe just two questions.

Firstly, you guys seem very upbeat about your CMOS business.

Can you give some visibility as to where you see the overall handset market sort of panning out over the next 3 to 6 months?

I'll give you my view on that, with the caveat that we're a semiconductor maker, not a handset maker, I can only tell you through kind of what we're seeing from our customers.

And I'll turn the clock back a couple months ago.

I mentioned that we had accumulated some cellular RAM inventory, which is a component that almost goes exclusively into cell phones.

We had also accumulated some VGA sensor inventory at about the same time.

Subsequently in the past, I'd say 30 days or so, we've seen kind of a resurgence in demand for those components, which has resulted in some liquidation of the inventory that we had accumulated, both on the cellular RAM, but in particular on the VGA sensors.

I interpret that as though our customers have worked off some of the inventory that they were sitting on.

So my take is essentially, with respect to our component feeds into the cell phone market were relatively flush in terms of inventories.

I don't know know what that means for the future.

If we take our customers' forecasts for the next couple of months, which we always take with a grain of salt, it would suggest a pretty robust environment running all the way through the first part of next year.

But kind of just to summarize and boil things down, I think things have strengthened substantially in the past couple of months, from where they were say in the August timeframe.

And at least from an inventory standpoint, I would think that a lot of the inventory that's been in the channel has been digested.

Okay.

Thank you.

So the second one is about the CapEx, of the $2.5 billion, is this coming from Microsoft, where is it going to be split out between in '07?

If your question is, we gave an indication of $4 billion CapEx for 2007, you can look at that as a bit more than half of that going into the IMFT operations.

And if you take the balance, split it about 50/50 between the TECH Semiconductor Singapore joint venture, and the balance of this to Micron historical operations.

In terms of the contract price, I know you mentioned about how robust pricing has been to date.

But into the quarter pricing forward, can you get some consensus and feel us as to what you're seeing from your customers right now?

I presume you're talking about DRAM pricing in the computer environment?

Exactly.

We just went through the first half of October pricing and we, we were successful in a 3 to 5% average selling price increase on DDR2.

I can't imagine we're looking at anything other than price increases.

Certainly next time around, just based on how strong the demand is.

I mean, it is a multiple of our ability to supply right now.

So, the environment is such that price increases are pretty much locked in certainly for the next time around anyway.

Thank you very much, gentlemen.

Your next question is coming from Bill Dezellem, Tieton Capital Management, you may go ahead.

Thank you.

We had a couple of questions.

First of all, coming back to the Flash business, and if we simply assume that prices are flat, given that it does affect profitability.

Would it be reasonable to assume that the losses in that business on a dollar basis, not margins, but absolute dollars, would decline in Q1, and then versus Q4 it would decline further in Q2 versus Q1, et cetera, as you roll through the fiscal year?

I think, Bill, you need to be looking at it as a bit more flat early in the year, and a greater change throughout the second half.

Thank you.

That's helpful.

And then relative to Lexar, you had mentioned that there will be an incremental 160 million of revenues in the Q1 versus the Q4, and would it be fair to say that that 160 million of it was the sales level in the fourth quarter that was not recognized, or are you adding on some number to what was not recognized in the Q1 for the normal seasonal pickup?

Lexar has been able to reengage with a number of customers by way of them being quite receptive to the broad portfolio that Micron offers.

But we're not taking a particularly optimistic view yet of first quarter growth.

That level is really reflective of what they have been able to do in recent periods.

So, said another way, in the fourth quarter, there were roughly 160 million of sales that Lexar, or the old Lexar experienced, but due to the way the accounting works, simply were not recognized on the sales line?

Bill, I didn't go back and specifically check what they had done prior to our acquisition.

This period is a little bit of a seasonal uptick, so I would expect they had slightly less than that in that prior period.

All right.

Thank you.

That's helpful.

And one final question.

It's relative to the Toshiba $288 million settlement.

Would you walk us through, #1, how the cash is anticipated to come in?

And then, #2, how the accounting will work for that, if it's any different than the recognition in line with the cash being received?

Bill, I can take the first part of that on how we expect the cash received, we were very specific in the press release to say over multiple years, we just can't go into any more details, and I'll let Bill handle the accounting approach to it.

The 288 million, the vast majority of that is actually recognized in the purchase accounting for Lexar.

So when we get the 10-Q on file, you will see an increase in receivables that is reflective of the vast majority of that amount.

As that transaction settled a number of matters that Lexar had preexisting to our acquisition, the accounting dictates that treatment.

Certainly, we're very, very pleased with the negotiations there, and it's reflective of the combination of Micron and Lexar,, and the strength that we brought to those negotiations, and Steve may want to comment to that.

Well, let me just say that the, there was a lot of anticipation I think in the Lexar litigation as they were going through it.

But we really have a lot different relationship with Toshiba, and I think that the agreement that we struck we though was pretty good.

As Bill mentioned, whether it's fortunate or unfortunate, I don't know.

But it ends up because of the proximity to the acquisition of Lexar, and because of some of it being attributable to something that was historical, it all got wrapped up and simply flows to purchase accounting, it won't flow through the P&L.

We still of course, get the cash, but it won't flow through the P&L, Bill.

That is helpful.

And I apologize for taking so much time.

As you look back on the negotiations, were there additional benefits above and beyond simply the settlement of the litigation, that in terms of future collaboration in any way shape or form, the two companies, that is noteworthy?

Well, Bill, as you might imagine, there's a lot of strict confidentiality around the agreement.

So I can't really comment on that, other than to say, we had a pretty good relationship with Toshiba historically, and expect to have a good one going forward.

Thank you all.

Thanks, Bill.

Next question, please.

Your next question is coming from David Wong from AG Edwards, you may go ahead.

Thank you very much.

Could you give us some idea of the relative gross margins of your different products like DRAM, sensors, and then the NAND Flash that's sold by Micron, not the half that's sold to Lexar at cost.

Sure, the rating hasn't changed much.

We still, gross margins are still the highest on image sensors, which you'll see when we report our financials on the 10-K that we're about 42%.

Specialty DRAM was closely behind that.

Core DRAM as we call it, was fairly close to the printed averages, and of course if did you the calculations, as Mark said earlier, NAND Flash we're underwater on today.

That includes NAND Flash that Micron sells, not the blended NAND flash together with what goes to Intel, is that correct?

That is blended together.

That is the way we are going to talk about these going forward.

We will report our Flash business as combined.

Okay.

Thank you.

You bet.

Your next question is coming from Jim Covello from Goldman Sachs, you may go ahead.

Good evening, guys.

Thanks so much.

I have a strategy question.

How do you guys think about the fact that some of the other players in the industry, who are kind of further along in their diversification strategy, and I'm really thinking about Samsung now, are now starting to move a little bit of capacity back from NAND to DRAM, because the DRAM margins are higher than the NAND margins?

So the real question is, by the time you guys get to where I'm sure you're ultimately going to get to on the cost side of what you're trying to do with NAND, are there going to be any profits to be had, because there's so many competitors adding so much capacity in this segment?

Well, first of all, I assume when you say, and you said other competitors, you mean other competitor, which has the kind of portfolio that we do.

I suppose you could say that Hynix has done a pretty good job, too.

There tends to be an assumption out there, that we have no ability to adjust our NAND capacity by virtue of the joint venture.

And it is true that we do have a partner for NAND Flash, and we have an obligation to ramp that product in concert with Intel.

And we're doing that.

In terms of the ability to change the mix of the product, clearly we have the capability to do that.

And in fact, it's probably not as widely known that we have adjusted memory wafers including NAND already, to some degree, to try to meet additional demands on the imaging, or things like pseudostatic RAM, et cetera.

We have some flexibility but we may have to make sure that we have the cooperation of our partner.

Now, I think there's a natural, I think there's a natural mechanism that will occur when our NAND capacity gets larger.

Because remember, we aren't producing that much NAND yet.

As Mark mentioned, the Virginia facility is starting to ramp as we speak.

And the Utah facility will start its ramp in the first part of the year.

Those are really the first two big thrusts in terms of capacity and product portfolio that we have, outside of the production that we had in the, you know, in the Boise facility that was previously running.

So we need to get to a critical mass first, before we truly start thinking about making these tradeoffs.

But I wanted to point out that we will be able to make those tradeoffs, we believe, because first of all, we will have the ability to do that.

There's no question the facilities can crossover and you have a 90, 95% equipment crossover.

The second thing that is worth nothing is when the NAND market is poor, obviously our partner will have less interest in getting a lot of it.

And when the NAND market is very strong, they'll have an interest in getting more of it.

So there's a natural mechanism that leads us both in the right direction, to utilize the significant capacity that we're putting into place.

It's just that right now, we don't have that much capacity to allocate around, because we're in the early stages of the ramp.

And I understand the quarter to quarter nuances, and I understand the fact that we're in a ramp.

And ultimately you're doing the ramp because you think NAND is going to be more profitable one day.

But what happens to the extent it's not?

What happens to the extent that the real winners in this, are the folks who are modernizing their DRAM facilities now, because there's just too much competition in NAND, regardless of the great demand drivers of NAND, there's just a [expletive] of a lot of supply out there?

Well, keep in mind that we are also modernizing our DRAM facility.

In fact, we think we're the first one to really ramp 78-nanometer DRAM technology right now as we speak in Virginia.

And that's on 300-millimeter.

And then TECH is starting its conversion to 300-millimeter.

And then essentially we don't really produce any think of it as, computing DRAM, or you would think of as the high density DRAM on 200-millimeter anymore.

That capacity is essentially all consumed by either Specialty or Imaging today.

So, we don't have a lot of capacity on DRAM that runs on older technology anyhow.

And we too, are what you called modernizing or advancing the process on 300 millimeter for the product that we do build DRAM on.

There's always this question when should Micron build the next DRAM fab?

And we consistently look at that.

So that I would negate that, we're not ready to announce anything today.

But the other thing I'd point out is that the Virginia facility runs half DRAM, and is ramping on now to be half and half.

So that would actually be a relatively easy conversion, because it's really just in the same facility.

Perfect.

Two other real quick questions, if I could.

What should we think about depreciation being for this year, for this coming year?

2000 should come in 1.6 billion.

Okay.

And then final question on the Vista, you referenced it earlier as a driver.

What I wasn't clear on, is that going to be a driver you think of incremental box loading, or incremental box sales, if you will?

And I guess the real debate that's going on in the industry, that we've seen the boxes kind of preloaded with DRAM, which is necessary, or DRAM considering that the OEMs thought they were going to have to be Vista-enabled already.

I think the significance in terms of demand creation from Vista is probably going to be more weighted towards box loading than system demand.

And you know, our own internal studies would indicate that the real benefits of the operating system are observed with loading of 2 to 4 gigabytes.

And that's a huge jump from where we are today.

The industry today is at an average of about 800 Megabytes.

I'm not suggesting we're going to go from 800 to 2 gigabytes in the first quarter of 2007.

But my point is, that there are huge benefits to be derived from significant box loading from memory standpoints.

So I think from our view, the majority of the impact is going to be as a result of increased box loading.

Certainly there's some pent-up system demand in anticipation of the release of the OS as well.

That generated one last question for me and then I promise I'll go away, do you think there's a change in the rule of what DRAM can represent as the total cost, as a percent of the total cost?

Because that kind of box loading 2 to 4 gigs, would certainly kind of violate that rule given how much PCs are coming down.

But maybe that rule is going to change.

I believe so and I certainly hope so.

With the prices of other components and PCs, particularly CPUs and hard disc drives under a lot of pressure and coming down, one could argue that it certainly leads to more bill of material costs that should be in fact, should have been allocated towards memory for the past 20 years, in my opinion.

But there's room for that, it seems as though there's room for that to occur as we move forward.

Great.

Thanks so much.

Your next question is coming from David Wu from Global Crown Capital

Thank you for taking my call.

I have two quick questions.

First on the CMOS side, as I look at your revenue, it looks like it's been flat roughly quarter-on-quarter, and you mentioned that ASP has been rising slightly.

Does it mean that unit volume is kind of flat on a, on a unit basis?

And I assume when you talk about VGA on quarter, the business, you're talking about units, or are you talking about dollars?

And then I have a question on memory.

Which is NAND.

At this point, I guess everybody is gung ho on expanding capacity, and you have got a rich friend at Intel.

So you're not going to blink.

Do you think that we might have a bloodbath sometime in '07?

This is Mike.

I'll address the image sensor questions, and I'll let Steve address the NAND question.

If that's okay.

Thank you.

On the image sensor front, our revenues in Q4 grew slightly over Q3, and our unit volume grew slightly as well.

It is true that the growth rate certainly levels Q4 versus Q3, I would attribute that to a couple of things. #1, I mentioned we accumulated vga sensor inventory that we weren't sure whether to interpret that as a slowdown in phone sales, or perhaps just some indigestion of inventory, if you will.

In fact, things have picked up significantly since we accumulated that inventory.

So I think we're back on the growth track again.

Secondly, kind of from a macro view, we have been riding a growing penetration rate of cameras in mobile phones for the past three years.

And once we reach close to 100%, obviously that has to level off somewhere, and the growth that we are going to see is the result of two camera phones and replacement phones.

And I don't think we're there yet at 100% penetration, but certainly we are at a point where three-quarters of so of the world's mobile phones have cameras in them.

And naturally, that business is going to slow down somewhat.

We don't believe it's going to decrease, but we certainly believe the growth rate is going to slow down.

We also firmly believe that there are several other markets that are going to drive strong growth for us.

Video conferencing and notebook PCs will probably be the first one on the docket, in terms of driving big growth for us in fiscal 2007.

Do-it-yourself security systems is another.

Digital sale cameras is another, and Automotive would probably be the next.

So there are several other markets that we believe are going to be big demand drivers for us.

It just so happens that at this point in time, they haven't really become that significant in terms of driving big demand.

Now on the NAND front, Steve, do you want to take that?

Yes David, let me first throw out there that there are a number of reasons that we are doing NAND, so despite the fact that there is capacity coming on NAND, and we're a part of that, getting into that market because of the growth rate of that market is clearly one piece of it.

We have a couple of others, and I'll come back to those in a second.

In terms of whether we think there's too much NAND capacity, or not enough NAND capacity for '07, I guess would I add the comment, I don't know if anyone can predict it frankly, because we're in such a high price elasticity correlation on NAND right now, for growing density and lowering cost, that it would be tough for anyone to predict.

Clearly there's going to be more capacity that comes into the market.

But frankly, we've never said, and if you've listened to us for a while now, we have never said that we thought NAND wouldn't be volatile, or that it wouldn't have either great expansions, or periods of time where you have pretty heavy price declines because of oversupply.

We think in fact that that will occur.

Having said that, though, we don't think it's going to be any worse than DRAM.

And in fact we believe that the opportunity for NAND is better than DRAM, because we think that NAND really looks more like DRAM did in the up to 1995.

So pre'95 in which net/net, it was a pretty good business overall.

In other words, the up cycles were longer than the down cycles, and those of us that were good at it made money in the DRAM business, up through the period of time where the price elasticity was pretty good.

Although even then, it was in to a relatively narrow market with a narrow customer base.

NAND actually is a much broader customer base and a lot broader number of applications.

So the potential for margin pressure, if you will, I think is not eliminated, but reduced by virtue of this variety of channels it goes through, where you actually have opportunity to have more value added.

If you think about how long it took the DRAM market to get to a differentiated portfolio, effectively it took 25 years for that to occur.

Right now, there's a lot of discussion around NAND being the high density, et cetera.

And that's where you see a lot of the commoditization, but the fact of the matter is, there is already a lot of opportunities for differentiation in the NAND space when you start combining it with controllers and interfacing software, et cetera, that we think it will have a lessening effect on the volatility.

Having said all of that, NAND for us will clearly help drive a lower R&D cost per wafer, and help drive, if you will, a scale gain for us, where a lot of the other costs get shared as well.

Because for two reasons.

One, not only do we have more wafers to spread the R&D over, but we also now have a partner sharing that cost.

And then obviously as the total business grows, we get the benefit there.

Because if you think about what was happening to us in the DRAM business, we had less and less product that needed an advanced process.

And that's just by virtue of the market bifurcating and splitting between Specialty and what you would think of as PC DRAM.

But it didn't cost us any less to develop that advanced process, it cost us the same, so we had a rising R&D charge per wafer internally.

It didn't look like it externally, because our silicon production was the same or growing.

We are actually reversing that trend now.

In fact, we internally in all of our numbers will show a decreasing R&D cost per wafer, by virtue of the scaling of our total silicon business, and by virtue of the crossover of development, which is essentially about 70% the same between NAND and DRAM.

And so we aren't really focused on just the supply side in '07.

I can't predict whether what you said is a bloodbath will occur or not.

But that doesn't change our course, not because we have a rich partner, but because of the other benefits that it brings to our business.

Thank you very much for the answer.

Your next question is coming from Douglas Freedman from Amtech Research.

You may go ahead.

Thank you for taking my question.

Steve, if you could, we just missed on basically execution of spending the CapEx, and moving 500 million from '06 into '07, if you heard you correctly.

When are you expecting that catchup to occur?

And we should be sort of entering the peak, if I remember your slides you presented recently.

Yes.

We, on the CapEx side, I think we need to be careful to differentiate really three things.

One is, when does the equipment get installed?

The second is, when do you actually pay for it?

And the third is, when does it show up on the books?

And oddly enough, all three of those are slightly different.

So the greatest, it was interesting, because when we came out and our CapEx was lower by this number that you just mentioned, half a billion or so, I think a lot of people thought that we were, in fact there were articles that we were deferring or pushing CapEx.

We have done nothing of the such, nothing even remotely close.

All of that is being driven by when this stuff gets on the books, or when it gets cash flowed.

In terms of schedules of capital implementation, I suppose it's always possible they are off a few weeks here and there.

But the fact of the matter is, the big capital products are on schedule, within a pretty tight band, so in terms of producing products based on the capital that we've been forecasting to spend, all of that is happening.

The rest of its frankly is just an accounting nuance, as to when the timing hits in cash flow and assets.

And by the way, to answer your question, more specifically, it probably gets almost all - caught up by the end of '07 or the first part of '08.

So we're still in sort of a peak spending period as we exit '07?

That's right.

It's fair to say, for us I think fiscal '07 will be a peak for us, in terms of what we report as capital expenditures.

Okay.

And then if I could just move on to, you have given us some targets on wafer loading as a percent of wafer loading by product.

Any change to those numbers for the end of '07?

Steve, I can take it.

Go ahead, Kipp.

You bet.

We had previously talked about being core DRAM being around 40% towards the end of '06, Flash being 10 to 15% or mid-teens.

CMOS image sensors, 20 to 25, and Specialty DRAM 20 to 25, those are still appropriate.

For the end of '07 timeframe, expect core DRAM to be 20 to 30%, Flash 30ish, plus or minus.

CMOS image sensors 20 to 25, and Specialty DRAM 20 to 25.

And lastly, my last question for you, Lexar bits, how many, what percentage of Lexar sales are being supported by IMFT fabs presently, and what do you think that will do in Q1?

Well, I don't think it's going to change much by the way,for the next couple of quarters, we have customer commitments for the current output.

So Lexar right now, as you may know, they had good supply relationships in place prior to the acquisition.

Those supply relationships have remained in place.

And we actually don't have much if anything to allocate to Lexar in the next couple of quarters.

And even if we did, by the way, I think our strategy is to have downstream, external and internal supply to meet their needs.

And in particular, when you look at their product portfolio and our product portfolio, they don't necessarily match up exactly right now, for part of their products, because they had historical practices of buying things from these other suppliers.

And there are some differences, and I think it will take quite a bit of time for those all to sort their way out.

Any clue on what we should think about for gross margins for their business next quarter, given the fact that spot NAND prices seem to be declining?

Yes.

Well, we're still sorting through that ourselves.

I think that suffice it to say, that if you looked at their historical results, they had pretty poor margins.

And we are weekly if not daily working on fixing all of that.

So the guy that runs that business for us, Mark Adams, who was a new COO at Lexar, he's only been there a couple months when we engaged to acquire them, is really focused on that, and I think doing a pretty good job.

So I think the answer is, they will improve from where they were.

How much progress is made in the next quarter, I don't know, I can't really respond to that yet, because they are still working on it.

All right.

Thank you.

Thanks, Doug.

With that, actually we'd like to wrap up the call now.

We would like to thank everyone for participating today.

If you would 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 for joining us.

This concludes today's conference call.

You may now disconnect.

Have a great evening.