美光科技 (MU) 2006 Q3 法說會逐字稿

Good afternoon, ladies and gentlemen.

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

At this time, I would like to welcome everyone to the Micron Technology third quarter financial release conference call. [OPERATOR INSTRUCTIONS]

It is now with great pleasure to turn the floor over to your host, Mr. Kipp Bedard.

Sir, you may begin your conference.

Thank you very much and welcome to Micron Technology's third fiscal quarter 2006 financial release conference call.

On the call today is Steve Appleton, Chairman, CEO, and President, Mark Durcan, Chief Operating Officer, Phil Stover, Vice President-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 the opportunity to review the third 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 7520674.

This replay will run through Friday, July 7, 2006, at 5:30 p.m. mountain time.

A webcast replay will be available on micron's website until June 26, 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 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 in 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 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.

For Micron it was a strong operational quarter in period which seasonally isn't usually that strong.

Net sales totalled $1.31 billion, and the Company recorded net income of $89 million or $0.12 per fully diluted share.

These quarterly results for the period ended June 1st are the first period to consolidate TECH Semiconductor's operations and do not yet pick up any results of Lexar Media, for which our acquisition closed on June 21st.

Net sales grew 24% over a year ago and 7% over the immediately preceding quarter.

Our diversification effort continues to strengthen, as sales of our digital clarity image sensors now represent 16% of total sales.

Sales of NAND memory compromised approximately 5% of total sales in the third quarter, including the sale of NAND devices manufactured in IM flash technologies to Intel.

Gross margin for the third quarter came in at 25%, up from 19% in the immediately preceding quarter and 8% a year ago.

The gross margin improvement was principally the result of strategic shifts in product mix to higher margin devices and some strengthening of DRAM average selling prices.

Our specialty DRAM and CMOS image sensors continue to rank as our highest gross margin percentage product offering.

As this quarter is the transition quarter to consolidation of TECH Semiconductor, it seems appropriate to acknowledge the effect on consolidation.

Simply said, there was no significant effect.

The margin on products sold in our third quarter that were manufactured at TECH Semiconductor were essentially the same as that enjoyed by the Company under the previous pricing arrangement under the supply agreement.

On a go-forward basis, since TECH utilizes Micron's designs and processes, TECH 's manufacturing cost will inherently inherently be similar to Micron's other wholly-owned operation.

As of third quarter end, Micron had cash and investment balances of $2.8 billion, and subsequent to quarter end balances have ranged from $3 to $3.2 billion.

Total debt, now inclusive of TECH Semiconductor, is approximately $520 million, or debt-to-equity ratio of 7%.

Capital expenditures, that is in additions to property, plant and equipment, have totaled in excess of $1 billion to slightly in excess of $1 billion through the first nine months of the fiscal year.

Our estimate for the fiscal year ending August 31st is a total of approximately $2.2 billion.

The significant increase in the balance of the year is reflective of the activity we have mentioned for some time in our Virginia 300 millimeter fab and the Lehigh, Utah, flash technologies fab and our Boise R&D facility.

R&D expense for the third quarter was $168 million and SG&A expenditures reached $113 million.

The increase in SG&A is reflective of the stage of development of historical legal matters and the inclusion of TECH Semiconductor.

The current level of R&D is likely representative of near term expenses, but SG&A will increase to approximately $130 million for our fourth quarter, as a result of the inclusion of Lexar Media for ten weeks of the quarter.

For the last three years, we've had continued improvement in cash flow provided by operations.

The overall improvement reflects our successful diversification efforts, and the third quarter's operating cash flow reached $384 million.

Additionally in the third quarter, we collected $171 million from settlement of our call-spread option.

Quarter over quarter you'll note some increases across the balance sheet, notably: Inventories; property, plant and equipment; accounts payable; and the non-controlling interest in subsidiaries.

All of those are the result of the consolidation of TECH Semiconductor.

As a final discussion item, I'll make a couple comments on the expected accounting for our Lexar Media acquisition.

The purchase price, based on the approximately 50 million Micron shares exchanged and the values assumed, is about $900 million.

A large portion of the purchase price may be ascribed to goodwill and, therefore, be a non-depreciating asset.

Additionally, the Company does not expect our fourth quarter revenues to be measurably higher as a result of the Lexar acquisition, as the Company will not recognize revenue from Lexar products in the distribution channel at the June 21st, closing date.

With that I'll turn the commentary over to Mike.

Thanks, Bill.

We are pleased with demand strength in the various markets that we serve.

Our product diversification efforts in memory and imaging are resulting in a desirable mix of revenues in market segments such as desktop and notebook commuting, mobile phones, servers, consumer electronics, communications infrastructure, as well as a few other segments.

From a demand perspective, moving into what is traditionally a seasonally strong second half of the calendar year, we are optimistic about the Company's prospects.

We're not seeing many surprises in the PC market.

The industry is experiencing unit growth rates in the low double-digits.

When compounded by DRAM content expansion, this calendar year the PC markets are driving deman -- DRAM megabit demand growth of around 45-50%.

This growth rate has been sufficient to move our DRAM average selling prices up by approximately 20% in the PZ -- in the PC space over the past quarter.

We are happy with memory content per PC growth and look forward to what will essentially be a step function content increase occuring with the Vista operating system launch in 2007.

The DDR2 interface continues to gain penetration momentum in the PC space.

I now estimate the demand profile to be about 3-1 in favor of DDR2 versus DDR1.

Micron's output is more balanced between the two interface technologies.

We maintain the entire DDR1 product portfolio on advance processes and are growing market shares in the non-PC markets that still favor DDR1.

We continue to place priority on development and delivery of high density DRAM components and modules for the server space.

Recently we were first-to-market with the two gigabit DDR2 component on our 78 nanometer technology.

We are in process of moving this device to volume production.

This product differentiation strategy brings time-to-market value to our customers, enabling us to create a defendable market share position in high-value server applications.

As I've stated in the past, mobile phones are perhaps the most intriguing market segment for Micron.

This is a very large market, approaching one billion units per year.

The primary application shift from voice to data is creating a slew of new opportunities for Micron imagers,DRAM and NAND products.

As our Italy operation has reached critical mass in CMOS imager production, we were able to achieve in excess of 30% quarter-over-quarter revenue growth in the just completed quarter.

With Idaho and Italy fabrication areas running full steam ahead, we have both production redundancy and head room for growth.

We are currently taking additional share in a growing market and believe that we now are enjoying close to 40% share in the CMOS imager market.

In order to continue fulfilling growing customer demands, we have announced plans to initiate CMOS imager production in our fab in Nishiwaki, Japan.

Following this addition, we will have three fabs on three continents to support the expanding customer base.

Memory subsystems in the mobile phone market are shifting from a base NOR to NAND flash.

We have just introduced Micron's first version of a NAND multi-chip package, or MCP, combining our own low-power DRAM with an IMFT-produced 1.8 volt NAND flash device.

Micron is among a selec -- select few suppliers with this capability.

Many of our current mobile phone customers are designing our NAND plus DRAM MCPs into their future handset lineup.

We look forward to adding these products to our existing portfolio of discrete low-power DRAM and [pseudo-static] RAM devices for the mobile communications market.

Complementing the MCP effort, we recently announced the development of our managed NAND product family.

This offering combines Micron's NAND memory with a high-speed MMC controller, offering extremely high-speed data transfer rates.

We see attractive growth potential in the consumer electronics market and are, in fact, already experiencing significant memory and imaging business in this area.

One key difference in consumer markets is that we see even more fragmentation of memory solutions due to the wide variety of end products.

This is welcome, as it allows us to leverage the many product development and customer engagement resources that we have around the world.

The newest additions to Micron, of course, are the talented team, brand equity and channel expertise that comes with our acquisition of Lexar.

The Lexar capabilities are a great fit and perfect complement to what already exists within the Micron infrastructure.

We will leverage the Lexar brand and retail channel presence.

Combining this with IMFT-manufactured NAND will click -- quickly become a vertically-integrated provider of NAND solutions to retail and OEM customers, thus unlocking incremental value for Micron in the process.

We're quite pleased with today's demand and the longer-term trends in the various market segments that we serve.

We believe that our differentiated product strategy and execution to that strategy speaks for itself, with the continued delivery of positive results.

As the IMFT operation reaches critical mass and we take advantage of economies of scale and NAND flash production, the Company will realize even more efficiencies.

Thanks for your continued support and interest in the Company.

I'll turn it back over to Kipp.

Thanks, Mike.

And with that, we'd like to take questions from callers.

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

With that, we'd like to open up the line.

Thank you. [OPERATOR INSTRUCTIONS] Your first question is coming from Mr. Michael Masdea with Credit Suisse.

Please go ahead.

Yes, thanks a lot.

Maybe for Bill.

On Lexar, is there any more detail on how much you're expecting, and also given what you say about disti channel, is there any -- when do you think that sort of overhang in terms of getting that product sold will kind of be cleared up and we can kind of sort of assume more normal revenue growth for what hits your P&L?

Yes, Mike, it's somewhat coincidental, but the volume of product, which they had in the channel at the June 21st date essentially carries them through the August 31st coincident with the end of our fiscal -- our quarter and year.

So beginning with the new fiscal year looks like when we're going to pick that up.

That answering your question there, Mike?

Yes, and is there any more details you can give us on any of the financials, in terms of the impact of Lexar or is that kind of it at this point?

At this point we'll just be looking through the allocation of the purchase price.

And as I indicated in the opening comments, there's quite a significant amount that would end up in the goodwill, after the allocation to the assets, brand and channel intellectual property.

Okay, fair enough.

And I guess on the DRAM side, the big growth was actually a little bit of a decline this quarter.

Talk about that a little bit, especially in light of what you said about still seeing low double-digit growth for the PC market, so why did we see such little growth in the DRAM bits?

And kind of following on, it looks like your big growth this year is going to be pretty dang low for overall DRAM bits.

Kind of give us an update on that number, if you could maybe?

Sure, I can start on the production side.

We basically guided low single-digits last quarter because we're taking more wafers and moving them into the image sensor space.

And, so, how that relates to sales, relative to Mike's comments of a pretty decent quarter-over- quarter, we just didn't have additional bits to ship.

We had very little inventory at the end of last quarter, so basically what we -- what we produced during the quarter is what we shipped.

Mike, would you like to add anything on the marketing side?

Yes, Mike, I -- the reduced DRAM bit sales really are -- I think, probably to repeat what Kipp said, it's just a function of what we produced.

Obviously our -- our capacity is being shifted towards more image -- has been shifted towards more image sensor production and ramping up our NAND flash operation.

And at the end of the day, at least in the quarter we just completed, the DRAM megabit output did suffer somewhat from the increases in the imaging and the NAND business.

From a market standpoint, the market's clicking along about like we expected.

We probably saw a DRAM demand increase in our last quarter of somewhere between 10% and 15%, so we did lose DRAM market share.

And the quarter that we're entering, on a polling of our large PC customers, we're looking at, you know, DRAM bit growth, demand, anyway, of probably around 15-17%.

So, you know from a demand standpoint, markets looks pretty darn healthy.

And from a guidance perspective, Michael, just for your models, you'll be looking at about mid single-digit bit growth on the PC DRAM, higher than that on both NAND flash and image sensors for Q4.

Great.

And this last question on the image sensor, there's some concerns out there about handset strength.

Maybe Mike or someone can talk about what they're seeing from the order perspective.

And given the growth you saw, any concerns about inventory on that channel at all?

First of all on the inventory question, none whatsoever.

We are -- we're flush -- we're flush on inventory.

I mean, we get stuff out of our production line and it's gone the next day, so we've got no inventory concerns.

And based on a polling out of our customers, you know, the handset customers as well as the camera integrators, not concerned at all about inventory in the channel.

In the handset market, we're benefiting both from growing number of handset units, as well as continued penetration of cameras and handsets.

So we're -- we're still looking at pretty significant growth.

Obviously, the last quarter was huge growth for us.

And you know, as I look out over the -- of course, the next couple of quarters, we're looking at pretty significant growth in both units and revenues on the image sensor side for mobile home cameras here, at least over the next couple of quarters.

Great, thanks a lot.

Thank you.

Your next question is coming from Shawn Webster from JPMorgan.

Please go ahead.

Hi, yes.

You talked a little bit about the bit growth in PC DRAM.

Can you talk about how you see your global -- how you see global DRAM shipments in calendar '06 and how you see Micron's bit supply growing in calendar '06?

Could you give us an update on that?

And then I have a follow-up question.

Yes, I think -- your first question, I presume, was a reference to the overall market in terms of DRAM bit growth and I'm going to give you a -- probably around 50%.

And I say that 50% is going to be accurate within plus or minus 5% in terms of demand growth and supply is going to be in line with that.

So, roughly balanced, supply and,demand, environment in calendar 2006 of 50% plus or minus 5%.

Our DRAM bit growth in calendar '06.

I don't have this specific figure at my hand, but it's going to be less than the market.

Kipp, can you comment more specifically on that?

Yes, we're looking, Shawn, at about 20-30% PC DRAM bit growth in fiscal '06 over '05?

Okay. nd then just to clarify, did you say that you expected end bit demand in calendar Q3 for PCs to be up 15-17?

That's correct.

Okay.

And then on capital expenditures, it looks like you lowered it from prior of 2.6 to 2.2.

Can you talk to us what some of the moving parts there were?

Yes, every year it always seems to be difficult to predict just how the final months are going to play out.

It's just timing of acceptance of tools and we have a large volume that are hitting, but honestly doesn't come down to accepting and getting those capped to the rate that we had thought six months ago.

So the 2.6 has been brought down to 2.2.

Okay, and was there any change to your fiscal '07 guidance at this point?

Fiscal '07 we're looking at $3.5 billion.

As we previously indicated, there's about $1 billion of that, which you can think of as historical Micron.

There's $500 millionish that's in the TEXT Semiconductor and $2 billionish is in the IM Flash.

Okay, and then turning to NAND, can you tell us what your NAND bit production and NAND bit shipments were sequentially in May?

We're not going to get into bit guidance on NAND flash.

We will give you some wafer references.

We ran about -- out of our total productive capacity, about 11% was aimed at NAND flash, and that'll go up a percent or two this quarter, meaning the fiscal Q4.

And then you'll see a pretty substantial ramp in wafer outs moving into the calendar Q4 area.

Thank you very much.

Thank you.

Your next question is coming from Mr. Glen Yeung with CitiGroup.

Please go ahead.

Mr. Yeung, your line is live.

Please go ahead.

Thank you.

Your next question is coming from Jim Covello with Goldman Sachs.

Please go ahead.

Hey, guys, thanks so much.

We got the bit shipment or, you know, bit guidance for the next quarter.

Any thoughts on ASPs in the different segments for us?

Well, you know, it's very -- we can't predict what ASPs are going to do going forward.

I can tell you, Jim, that, in the DRAM environment, if we were to see a stable DRAM price from now through the end of our quarter, the overall for the Company -- the DRAM average selling prices will be roughly flat quarter-over-quarter.

So, in other words, they've been relatively stable for the first three weeks of our fiscal Q4.

In the NAND flash area, prices in early Q3 for us, they went down, they stabilized somewhat toward the middle of the quarter, and they actually trended up a little bit in the latter part of our Q3.

If prices were to be flat going forward for the balance of this quarter in the flash world, they'd be flat to down 5% quarter-over-quarter.

We're expecting the price -- flash prices are going to continue to tick up a little bit through our fiscal Q3 and, in fact, through our fiscal Q4 -- I'm sorry through our fiscal Q4 and through our fiscal Q1.

Okay, that's helpful.

If I can ask you a quick follow-up.

On the image sensor business, I know on the 10-Q you guys break out the margins by product.

Can you tell us now what the image sensor margins were this quarter and, then, what you think the long-term margin target in that business is?

We ran about 40% gross margin, Jim, on image sensors this quarter.

I'm not going to make a stab at trying to predict what they'll be long-term.

Okay.

And -- and that, if I'm not mistaken, I have to go back and check numbers, but that was up quarter-over-quarter?

Down slightly. 44% last quarter.

Oh, because you were talking - I'm sorry, you were talking about gross margins there.

Okay.

And I guess without predicting a long-term model, do we think that amount -- that level is sustainable, given the competition or --

Yes, I'm really not -- you're asking me to predict a couple of different things.

I can, obviously, argue a pretty good case for cost reductions we have coming, but trying to predict pricing in this business, like anything else, is pretty difficult, so I'll leave that to you experts.

[LAUGHTER] Yes, not.

Thanks so much, I appreciate it.

You bet.

Thank you.

Your next question is coming from Mr. John Lau with Jefferies & Company.

Please go ahead.

Great, thanks.

You gave us an indication with regards to how strong the image sensor market is and the inventory levels for the wireless handsets.

I was wondering if you can also give us your market indications for what's happening in the PC market, especially given that there's a lot of concerns out there of a stall coming?

We noticed that the pricing has been stable, but we wanted to get your input on that.

Thank you.

Yes, John, this is Mike speaking.

My visibility on PC inventories, either in the channel or at our customers, would not be of much value, because I just don't have visibility into that.

I do have pretty good visibility into, of course, our DRAM component and module inventory, whether it be in the hands of the customer or our own hands, and I tell you that the shelves are bare.

We've got virtually nothing in stock here at Micron and when I look out into the Micron-owned inventories at the customer sites, they're basically empty, as well.

So, from our viewpoint, there is no inventory.

Now, obviously, I couldn't speak for our competitors nor could I really speak for the channel players, but we're not putting much, if any, Micron product in the channel, either.

So, from an inventory standpoint, we're in great shape.

But, Mike, if you take that thought one step further, the current inventory levels are pretty lean as you say.

Looking into the second half of this year, with the inventories lean at this level, do you think the industry will -- what do you think the supply-demand balance will be for the rest of the year, given that you're actually slowing down your wafer starts?

Well, you know, our -- certainly the amount of product that we're going to be putting into the market is going to be less than what the market appetite is, basically, so we're not going to be putting pressure from a supply standpoint in the market.

My view is that we're going to be -- supply and demand are probably going to be roughly in balance, if anything.

I think we're in a situation where there are going to be pockets of market segments where demand is going to exceed supply.

In general, we feel quite bullish about the prospects in the DRAM market for the second half of the year, if nothing else based on the seasonal demand pull for consumer items, mobile phones, PCs, as well as the continued content growth that we're seeing.

So we feel pretty good about it.

Great.

Thank you very much.

Thank you.

Your next question is coming from Mr. Glen Yeung with Citigroup.

Please go ahead.

Thanks.

Sorry about that last time, guys.

We hear you saying we want to make image sensors, more NAND flash, and having to reduce wafers in DRAM to do that.

Is there a [ster] target that you're willing to live with in the DRAM side, just so we get a sense as to when you may have to stop moving wafers around?

Well, I think that the objectives on our market share aren't really specific in terms of a particular percentage.

We, clearly, are willing to accept less and less market share until we've reached an exposure into the PC space that we're comfortable with, and I can tell you that it's still significantly less than where that is today.

But the transition of our wafer portfolio is rapidly occuring away from the product that goes into the PC space.

You just saw from our release that, for the first time in our history, sales of product from Micron into the PC space just dropped below the majority. would expect that, over the next 12 months, maybe 18 months, to probably at least go in half as these other products grow for us.

Now, it doesn't mean that we're necessarily producing that much less DRAM, because the DRAM we're producing, the low power DRAM, the pseudo-static RAM and the stuff that goes into wireless and packets up MCP's, that is actually pretty good.

And, in fact, that'll continue to grow, in particular when you look at the consumption in the server space.

A lot of the product and a lot of the effort that we have right now is in that area, so I'm not projecting that our overall DRAM market share will cut down in half.

I'm just saying that the stuff that goes into the PC space will drop dramatically.

But, yes, we'll lose some market share in the DRAM space, but we're willing to accept that, down to the level where we have to have some presence there, not down to the level where we actually would be considering adding more wafers back into the DRAM space for the good markets as they grow.

Okay.

Makes sense.

When you look out into the competitive environment, if we're just speaking on the DRAM side, is there anyone over which you will lose sleep at night?

Are there any players out there where you think, boy, these guys are looking to be very aggressive here in terms of adding capacity or trying to gain share or both?

For which space are you talking about?

For the DRAM space.

Oh, I see.

Well, I don't know anything more than what you know from the public reports on capacity being added.

One of the reasons that we're pursuing the strategy that we have been is because of irrational capacity that comes along.

In particular, the easiest space to get into is the PC space because, you know, they have the least demand in terms of all of those things that companies like ourselves are capable of doing on the service side, whether it be quality or supply chain hubs, et cetera.

And so, I don't see anything particular out there that we're concerned with, other than just the historical behavior that, when the market's pretty good, they tend to plow too much capacity into it.

Okay, all right.

Well, thanks.

Your next question is coming from Eric Gomberg with Thomas Weisel Partners.

Please go ahead.

Yes, I was hoping you could talk a little bit about what's going on in IM Flash in terms of the road map lf -- maybe talk about the densities of the production at this point and where you'd expect it to be over the next couple of quarters?

Sure, this is Mark.

We -- as we've said before, we're running primarily 90 nanometer NAND today and well into shipping 72 nanometer SLC products.

We did qualify and are sampling our 90 nanometer MLC, but primarily, we're focused on the 72 nanometer node and and qualifying and shipping 72 nanometer MLC in 2007.

With respect to further out, I think we'll wait and talk about nodes beyond that with our partner, Intel, probably sometime later this year.

And in terms of chip capacity, this is primarily two gigabit at this point?

Two gigs and four gigs now moving into the marketplace, also.

Okay, and would you expect later this year to begin integrating that into your Lexar acquisition or is that something that's more kind of '07 and beyond?

Well, we have -- that's a good question, actually.

We have use for all of our current NAND forecasted output, at least in the near term.

And as a result, we would expect Lexar to continue on their business model with their supply agreement,s as far as we can see in the future.

So, we think that actually would be incremental to what we're doing here.

Okay, that's helpful.

And just one last question.

Can you just -- maybe I missed it before, but talk a little bit about the composition of the inventory?

You talked earlier that it's up a little bit, so where is that sitting in terms of product?

Yes, this is Mike speaking.

Actually, the finished goods inventory is down quarter-over-quarter, and what we have is -- it is not much, but what we have, we have -- it's probably an equal distribution of NAND flash, DRAM and virtually no image sensors.

We have some inventory in place at the customer, our vendor-owned inventory, very little in Micron warehouses.

Basically there's not much inventory at all.

Certainly less than two weeks, probably closer to one week worth of finished goods inventory.

If the question in part was looking at the balance sheet and the growth in inventory dollars, the finished goods down, as Mike indicated, raw materials flat.

The only growth is in the work-in-process, and that's just as a result of the TECH Semiconductor consolidation.

Overall WIP throughout the historic Micron operations are actually down.

Okay, very helpful.

Thank you.

Thank you.

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

Please go ahead.

Thanks.

I just wondering with respect to Lexar if could give us any framework for how we should think about that in terms of revenue in '07 -- in the fiscal '07 period?

Well, I don't -- there's not a lot of guidance we can give.

You have to keep in mind, we only closed a week ago.

And you obviously have all of their numbers up until that point in time.

Our intent is, of course, to take where they're at and try to grow that business.

So, to the degree that we're able to do that, that's something we'll have to look at in the future.

But just keep in mind that the -- as Bill said, there won't be any revenues for this quarter by virtue of the way the accounting works, and it'll just be incremental after that.

But in terms of SG&A, we're going to model 130 going forward, and the R&D you think is going to be fairly flattish going forward?

Yes, both of those are correct.

The 130 reference was specifically for a Q4 and ten weeks of the quarter incremental Lexar activity.

And just -- can you just clarify, or Bill, on the 17% [inaudible] or wha -- and how that breaks down [inaudible] with respect to 15-17% growth?

Oh, sure, that was Mike's reference to the demand profile we see in, basically, our fiscal Q3 over fiscal Q2, driven by PC units and content.

Okay.

And with respect -- and going forward in the coming quarter in terms bit growth, then, and how you see that breaking down between the segments?

Oh, bit growth on a production side, from our standpoint?

Yes.

Mid single-digits on PC DRAM and then higher on flash and image sensors.

Higher than that.

Lastly, just in terms of accounting.

In the May quarter, it looks like there was this non-controlling interest and net loss of 17 and then there was -- seemed to be like a pretty high interest income number of 27.

How should we think about the interest income number going forward, and is this non-controlling interest in net loss a one-time thing?

How should we think about that?

Thank you.

Yes, on the interest income it was indicated we had some very significant growth in cash and short-term investment balances.

It was also indicated that we've got a pretty good capital spend profile ahead of us, including next year's being a bit front loaded in the year.

If you look at the next couple of quarters and thought about $20 millionish in interest income, you're probably in the right ballpark, with that tailing off a bit later next year.

On the non-controlling interest, the $17 million was virtually all a result of the TECH Semiconductor consolidation and the purchase accounting as -- if you think about TECH Semiconductor on a go-forward basis over a time horizon, it's more likely to be close to break even.

So, not anticipating a significant non-controlling interest charge in any quarter is the right way to look at it.

It'll be somewhere lower than 17 going forward, between that and zero?

Likely.

Thank you very much.

And Tim, just let me correct one thing I said before.

I guided you to PC DRAM bit growth of mid single-digits.

That'll actually be flat this next quarter, but costs will actually come down in the mid single-digit range for --

Say that again?

Your bit growth in terms of production for PC DRAM will be flat?

Will be flat Q4 over Q3, yes.

And the other areas would be higher than that?

You're going to see growth in the other areas, correct.

Why would you not see bit growth in PC?

Similar to what we've talked about before.

Wafers are still moving into other products, specialty DRAM, image sensors, NAND.

But you would think it's a growth quarter in terms of revenue for you?

Well, that would be your own -- your own discussion about ASPs.

I think Mike's probably described an environment where we're going to ship what we make, and so then you'll have to determine what the ASP effect will have on revenues.

NAND's clearly down and the other one is -- DRAM's flat.

Okay, thank you.

Alright, thank you so much.

Well, that's assuming your ASPs.

That's not what Mike has said, but if that's your take away, then we respect that.

I thought you said that if it remained as it is for the rest of the quarter, it would be NAND flat to down five and DRAM, if it remained as it is, flat?

Yes, that's giving you a reference point that, if those two things happen, then that would be the effect.

Mike's not making a prediction on what the market will do for the next two months.

Sure.

Sure.

Thank you so much for that clarification, Kipp.

You bet.

Thank you.

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

Please go ahead.

Hi, Kipp, I'm sorry, color me confused following on the previous question.

It sounds like you are saying that the, you know, approximately, you know, half or somewhat less than half of your DRAM business that doesn't go into the PC end market is going to see production growth and presumably shipment growth.

Is that correct?

Correct.

Okay, so -- and you talked about the PC business overall being less than the majority of revenue.

That presumably includes the other non-DRAM business.

Are you able to give us some color in terms of what the end market composition of your DRAM business is?

Yes, we can do that, Mike.

We can share that was you promptly.

I can -- from a market standpoint, Joe, I can give you some ballpark figures here.

I apologize, by the way.

This is not necessarily broken out by -- it doesn't single out DRAM, but in terms of our overall revenue mix, jow about if I rank the market segments for you.

Desktop and notebook computing would be the largest.

Okay.

Mobile terminals would be number two -- handsets, basically, would be number two, servers would be number three, consumer electronics would be number four, and then there are a mishmash of segments beyond that, but that's the ranking.

And those four that I named are -- are very significant, in terms of revenue dollars.

Okay.

Well, then it sounds to me like, if your PC business is -- you know, and obviously, you know, hypothetically assuming that pricing's flat, nobody knows -- but if it is then, you know, your PC DRAM business doesn't generate any revenue growth, but, you know, if you're able to see shipment growth in your non-PC DRAM businesses, that would generate revenue growth.

Is that correct?

That is a -- that is a fair assumption.

Okay.

And then the last question would be then, is the margin on your non-PC DRAM stuff better than your DRAM stuff or PC stuff or the same or worse?

I can rank them for you like we have in past quarters, Joe.

Image sensors are still the highest gross margin, followed by specialty DRAM.

In this quarter, NAND and DRAM flip-flopped -- PC DRAM flip-flopped, so PC DRAM's third and NAND fourth.

Okay, understood.

And then last question, I guess for Bill.

Seems like there would have to be some operational and gross margin synergies for Lexar, so would it be fair to say, as I think that year to 18 months out, that the objective is to meet all of Lexar's NAND requirements with internally sourced wafers, that's the first thing?

And then the second thing is at the operational level, do you have some thoughts about the synergies you might ring out of that, as we think about how the integrated model look?

Yes, Joe, this is Steve.

I -- first of all, given the growth profile of, I think, the NAND demand and given some of the other obligations that we already have in place or that we're working on, I'm not confident in 18 months from now all supply comes internal.

However, having said that, as Lexar's noted themselves publicly, they have some pretty good supply relationships now, that I think would be advantageous for them, that remained in place after the acquisition.

So, if the Lexar team does their job, that business will grow, in addition to the business that we have growing.

So we will get operational synergies.

There's no question about that.

And we'll -- I think we'll be able to bring some efficiencies to that model.

But also remember that part of the business is in controllers and a few other things. o, you know, which we're actually looking to invest more R&D in what they've been doing, as opposed to less R&D, but get more efficiencies on the operational and the SG&A side.

Okay, understood.

Thanks, guys.

Thanks, Joe.

Thank you.

Your next question is coming from Mr. Dave Wong from A.G. Edwards.

Please go ahead.

Thank you very much.

Can you give us any idea of what gross margins the Lexar product is currently selling for?

It's too early for us to do that, Dave.

We just got our hands on it.

And the thing that I think is important to highlight -- and I think this question's come up a number of times -- and that is, boy, have you really done enough planning, you know, given we had enough time to know that we were going to close on Lexar or not close on Lexar, but the fact of the matter is there was some uncertainty, you know, until we were able to complete the transaction.

And as a result, I think this is the first time in my history where both boards approved a transaction and yet there was still uncertainty in the market as to the transaction closing.

And as a result, we really didn't know whether we were going to be competitors or partners, so to speak with Lexar right up until -- right up until the closing.

So we couldn't do as much planning -- and, frankly, we didn't get as much visibility into some of the things that they were working on and developing as we would have normally liked to.

Obviously, we had enough insight to know we wanted to do the deal, but it's just going to take us a little bit of time to get our hands around that.

And, so, it's just too new to us right now.

Okay, fine.

If I could just follow-up quickly on that, though.

For all your non-Lexar stuff, you had really good margins this quarter, about 25% gross margin, I guess.

Are there any things that improve -- pricing aside, are there any other things that affect gross margin in coming quarter?

Are there any special costs?

The [B Tech] -- there's no effect from B Tech now, right?

Are there any other factors there?

Sure, we have costs coming down on a multiple of our products.

We've got, potentially, some -- as Mike described earlier, some upward trending, NAND flash ASPs, potentially, on the quarter, as well.

And, of course, the more allocation comes from the higher margin products, which is image sensors and specialty DRAM, then that's a positive impact on gross margins, as well.

Great, thanks.

Your next question is coming from Mr. David Wu with Global Crown Capital.

Please go ahead.

Yes, Mike, can you talk a little bit about CMOS sensor.

I was just wondering two things.

Number one, your gross margin dropped sequentially.

Is that a function of VGA graphics being a bigger percentage of the total?

Secondly, roughly how much are these 1.3 and two meg megapixel sensors versus these VGA graphics for you guys?

Yes, David, from a mix standpoint in terms of pixels, that's the -- roughly half of the units we shipped were VGA sensor and half one, two, three, and five megapixel sensors, and so -- that's ballpark, ballpark half.

And I think it's safe to say that our margins on VGA sensors are lower than they are on one megapixel -- actually, the greater pixel density per chip, the greater our margins are.

That's a pretty safe general statement.

And, actually, I may have answered all your questions there in on once sentence.

Is it --

Yes, pretty good.

Are these going to be shortages, though, because I think in the last time you talked publicly, you said you were 15% under shipping the market?

Yes, I think we left some business on the table from a supply standpoint in our fiscal Q3.

In fiscal Q4, which is -- you know, we're three weeks into fiscal Q4, I think we're about right even with demand, basically.

I think we're able to meet the demands of the market in fiscal Q4.

With Italy really running full steam now and some capacity additions that we've got planned in Idaho, as well as, you know, looking at starting sensors in Japan, certainly by fiscal Q1, I think we're going to be in great shape from our -- you know, a standpoint of being able to meet the needs of the market from a supply standpoint.

At some point would CMOS sensors possibly as much as half of your profits at a gross profit level?

Because that number keeps going up and your market share keeps going up, so you look at that number and you say, well that's approximately a quarter of the earnings.

And if you keep that up, it won't be too long before it approaches half.

Was there a question in that?

Well, I just wonder what your plans are, because you've got another fab coming up, right, Japan?

Yes.

I think our approach is we have, obviously, significant capacity in a variety of products worldwide, and we want to optimize that capacity, and we'll continue to do that to the improvement of the margin.

The only caveat related to that is we do have a partnership with Intel on NAND and we have obligations to continue to invest and ramp that, as we would want to anyhow to substantiate our position in that market.

So, you know, again I think as its somewhat already been capsulated, we expect the imaging business to grow.

We'll keep alloqua -- allocating resources to that business as the demand grows with it.

We expect the NAND profile to grow rapidly by virtue of the investments we're making in IMFT.

And when it comes to the DRAM, we're going to allocate those resources among the best product for return to the Company.

And it's just going to go wherever it goes, and we'll just have to, you know, keep our eye on it.

Okay, well, thank you.

Have a good quarter.

Thank you.

Your next question is coming from Mr. Doug Freedman from Am Tech Research.

Please go ahead.

Hi, guys.

Getting to the bottom of the barrel here, but could you talk about your overall production increase in wafer starts from Q3 to Q4?

You bet.

It'll be up just a couple percent in terms of wafer outs,Doug.

And any plans or any update on the scheduling and node at which you're going to convert the Singapore Tech JV to 300 mil?

Can you repeat that again?

Can you offer us some details on the timing and node point which you'll convert the 300 mil at Tech JV?

Oh, I got you.

Mark, you want to answer that?

Yes, we are -- we're in a -- building out some space right now.

We'll be in sort of pilot line mode late this year on 78 nanometer, and that would be a 78 nanometer, 300 millimeter ramp of that.

And -- alright, terrific.

Thank you.

Thank you.

Your next question is coming from Mr. Hans Mosesmann with Moors & Cabot.

Please go ahead.

Thank you.

Most of my questions have been answered.

I do have a question, though, regarding the multi-chip module solution in the wireless space.

Are you saying that there's certain OEMs in the wireless space that are abandoning NOR and going with the NAND/DRAM solution?

No, I would not say it's accurate to say that certain OEMs are abandoning NOR.

I think what's happening, as new applications are demanding more memory content, all of the OEMs are making decisions to implement NAND-based MCPs in certain phone models, as opposed to NOR-based MCPs.

So, what you have today is, basically, every handset manufacturer that I'm aware of with a lineup of NOR-based MCPs as well as a new lineup of NAND-based MCPs.

And what's typically occuring is that the high-density MCPs are deploying NAND plus low power DRAM and the lower density MCPs are deploying plus pseudo-static RAM.

Okay, and then one follow-up.

These higher-end phones with NAND and DRAM, do they have any kind of NOR in the handset?

In some cases they do.

In some cases they have a discreet NOR.

In other cases they may be implementing the NOR via an MCP with a DRAM or a pseudo-static RAM and a stand-alone embedded NAND flash chip.

So there are a whole variety of different configurations.

Okay.

I --

n general, the trend towards more NAND content and less relative NOR content in cell phones, that's a guarantee that's going to occur.

It'll -- you know, it'll unfold over the next several years.

Perfect.

Thank you very much.

Thank you, your next question is coming from Mr. Michael Lucas with Appaloosa.

Please go ahead.

Yes, I just wanted to confirm what you said.

Where did you think PC shipments are going? 15-17% up, did you say?

No, did not say that.

DRAM demand from our PC customers in the calendar Q3 trime -- time frame, which roughly corresponds with our fiscal Q4, will be up quarter-over-quarter 15-17%.

Now there are two components to that.

One is the content per box and the other, of course, is the number of boxes.

I can't give you the granularity on that, but overall demand from the PC customer is up 15-17% quarter-over-quarter.

Sure.

And also, I guess, what is your guys' assumption on -- do you make any assumptions on what PC shipments will be or what handsets will be in a given year?

All we do is take the published third-party data, and I think in the PC area that's 9, 10, 11, 12% for the year, and in the handset, maybe slightly higher than that.

Sure.

Do you see a demand from some of these lower-end markets that they want phone -- they want cameras in those phones?

Yes.

Okay.

And lastly, I know that the market kind of throws you off a little bit here on revenues, God only knows why.

But if everything stayed flat, we would see a margin expansion relative to all the information you've given out in the call because your costs are going down.

Correct?

That's the right assumption.

That's correct.

How much will the cost go down in each one of those areas in terms of image, DRAM and NAND?

Yes, we're not going to give you specifics on each one of those.

You know, in general we've guided that PC DRAM, which is the largest portion, will be down mid single-digits.

Okay.

All right, great.

Thanks.

You bet.

Your next question is coming from Krishna Shankar with JMP Securities.

Please go ahead.

Yes, can you give us the gross margins for each of the four segments for the last quarter and this quarter?

You mentioned that CMOS image sensors was 40% versus 44% last quarter.

What are gross margins in the other segments?

We don't break those out.

We give you the image sensor and then the corporate.

Okay.

And you said that this year DRAM bit production would be fairly well-balanced with demand at about 50 plus or minus five.

Can you have hazard a guess as to what it might be next year, given your knowledge of capital spending in DRAMs?

It looks like, from what's shaping up for spending today, probably a calendar '07 over '06 is on production bit growth probably in the 50% range, maybe a little bit lower.

And if Vista has a reasonable ramp, what could demand growth be, you think?

Well, there are some pretty positive trends, which I'll let Mike speak to here.

I think -- you know, again keeping in mind that PCs now are -- they're not driving the whole DRAM market, but it's certainly the majority of the market.

In a non-new operating system year, what we're seeing from the PC market is growth rates -- demand growth rates of 45-50%.

It stands to reason with a new operating system being layered in 2000, the incremental affect of that is going to be positive, obviously from a demand standpoint.

And I don't know enough to make a specific prediction, but I think it's safe to say that, if a normal year is 45-50%, next year, with the Vista introduction, it certainly stands to reason it'll be significantly greater than that.

But I couldn't give you a specific figure.

Okay, and my final question is on NAND demand and supply in the second half of '06.

Last year we saw the nano iPod phenomenon, which really shot -- demand shot up and pricing really escalated.

What's your viewpoint on capacity additions NAND flash going into the second half of this year verses incremental new market drivers?

You know, on -- I don't know on the capacity side.

Of course, I know what we're doing from a capacity standpoint.

There's no way I could predict with any precision what's happening from a competitive standpoint.

I can say that we have pretty good visibility into demand and, you know, the big drivers of NAND demand -- the MP3 audio players, a variety of cards for digital cameras and so forth -- hugely dependent upon the seasonal point in the year, and we're already seeing significant strength in demand, relative to what we were seeing for the past six months.

So, again, we're quite bullish on demand for the second half of the year on NAND.

And you guys are probably in a lot better position to estimate what the supply growth is going to be and how well that's going to be lined up to support the growing demand.

Then final question, you talked about the growth of MCPs with NAND flash and NOR DRAM, would that take away from the growth of adding cards in cell phones where you have these SD cards, would that take away from the growth of adding cards to Lexar cell phones or will it be [inaudible] incremental to add in cards in cell phones?

I don't believe that's going to be the case.

We're seeing so much growth going forward, by the way, both volatile and non-volatile memory content in mobiles phones that it would be -- I'd be hard-pressed to say that cards were going to de -- take away from embedded growth or vice versa.

In the non-volatile area, just next year alone we're looking at a tripling of con -- of flash memory content per phone, and on the DRAM side I think we're easily looking at a doubling of memory content per phone.

So, tremendous growth from a memory standpoint in mobile phones as we move forward, both from embedded as well as detachable cards.

Great, thank you.

Your next question is coming from Mr. Thomas Smith with Standard & Poor's Eq.

Please go ahead.

Yes, following up on phones, I wondered if you had some sort of dollar figure arranged for the addressable markets that you now have in phones, now that you have sensors and various kinds of memory?

You know, you're -- well, I'm sure we could scratch one up here pretty quickly, and -- but I don't have it off the top of my head.

In terms of the addressable market, considering our image sensor offering and our memory offering, I really don't have one off the top of my head.

Okay.

Safe to say that it's considerably more than it was a year or two ago as far as your Company history goes?

Certainly.

Let me share a little bit of data with you here.

If you take the number of phones that are manufactured this year -- I don't know if it's 900 mill -- closer to 900 million or a billion -- from a DRAM content standpoint, probably about 25 megabytes of content per phone, and from a NAND flash standpoint about 45 megabytes of content per phone.

From an image sensor, probably about 70% penetration and the average selling price of a image sensor in a phone this year is probably about $2, between $2 and $3.

So that -- you know, you can do your own back of the envelope calculations based on those figures, I think.

Okay, great.

Thank you.

Thank you.

Your next question is coming from Lilian Li with TIAA-CREF.

Please go ahead.

Hey, it's Manish Goyal.

I have a couple of questions.

It seems like you sound pretty positive about the DRAM -- PC [inaudible] DRAM supply demand equation for the second half of the year, yet you are not investing putting capital.

Can you reconcile that for me, why you're not putting capital for that?

Yes, I think, Manish, the -- it's not a matter of not putting capital in.

We're putting capital in, we're just not allocating it to produce that particular product, and the answer's really pretty simple.

Just look at the last seven years.

There's only been one or two, really, that've been okay.

So, we're allocating the resources to products that we think are more differentiated and have a higher gross margin over a sustainable longer period of time.

You know, we cannot -- one thing that I think is worth noting is, you can't go out and develop and [inaudible] some customer or MCP customer, et cetera, where they have to design this product in months in advance or maybe a year in advance, finally get it to market, and call them up and say sorry, we decided to make some PC DRAM instead.

It just doesn't work that way.

So we have commitments to make on these products as we build those businesses, and as a result, we're going to allocate the resources to do that, rather than just short-term change it.

Having said that, it's not that we won't benefit from an improving PC mar -- pricing environment for a product that goes into PCs, because Mike already stated, that's still a big part of our business and we will.

But our focus, as we've said now for several years, is to develop a more differentiated product.

It's not to exit the PC space; it's to --it's to lessen our dependence on it, and that's what we're trying to do.

So, I guess the follow-up here is if, as you said, you know, of the last seven years, there was one good year for PC [inaudible] DRAM, then why even be in this business or have such a large portion of your total business in PC [inaudible] DRAM?

Why not shrink it to something more substantial?

How do you think about that?

Well, we are shrinking it.

As I said, it's continued to shrink for several years now., and we're continuing on that path.

So, the other thing is, you say why make it at all?

Well, we do have a lot of capacity and pace.

We do need scale.

We need to be -- we need to continue to be cost competitive in our memory business in totality, and that requires that you produce, I think, in all of these spaces.

So, it's not just as simple of cutting that all off and thinking that it solves all the problems.

It doesn't.

You have to go through a transition , and the intent and what we've been trying to achieve is go through that transition without damaging the Company in any particular quarter or 12-month period.

We're trying to keep a Company that is growing in the areas we want to grow, continues to make the use of efficiencies that we've learned and developed from the commodity DRAM business, and keep that, by the way, knowledge base moving forward by participating in those markets, but leverage that in the more differentiated product.

I have one more question, this is on capacity.

Could you talk about how much of your total wafer starts today are on 65 nanometers and if you -- when you look at 200 mm installed capacity, how much of that you will be converting to 65 nanometers to produce PC [inaudible] DRAM?

Well, first of all, we're not running much of all on 65 nanometer in production.

Sure.

I don't think [inaudible] hardly anybody else is either. obviously, we're working on the technology, but we haven't deployed it.

Sure.

So, obviously, we're working on the technology, but we haven't deployed it.

And, really, when you think about NAND for us, it's likely to be a 50 nanometer node as opposed to a 65 nanometer node, and we're doing that -- obviously working on that now as we speak, as well.

In terms of converting from 200 to 300, interesting enough, our model now greatly leverages the 200 millimeter capacity that we've had in place, even though it's a relatively advanced process node.

So, you know, will a 200 millimeter capacity that we have today, that we're utilizing for more different data products, eventually become 65 nanometer?

I don't know.

You know, probably if you can still -- if you continue to make that product based on the 200 millimeter.

The higher probability is that most of your 200 millimeter will eventually convert to 300 millimeter, as it becomes less expensive and more applicable to a generation minus one or two or three product portfolio base.

So, eventually I think all 200 millimeter will convert for products that we probably make.

It just takes quite some time for that to happen, and to leverage the 200 millimeter that we're actually taking out of the facilities that have been running advance processes for NAND DRAM.

So I think the probabilities of 200 millimeter becoming 300 millimeter are actually much higher than 200 millimeter moving all the way to 65 nanometer for most of our product base.

Thank you so much.

You bet, and I'd like to thank everyone for participating on the 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 the 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 very much.

Thank you.

This concludes today's Micron Technology conference call.

You may now disconnect your lines at this time and have a wonderful evening.