美光科技 (MU) 2009 Q4 法說會逐字稿

Good afternoon.

I will be your conference facilitator today.

At this time, I would like to welcome everyone to the Micron Technology's fourth quarter and fiscal year end 2009 financial release conference call.

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, Kipp Bedard.

Thank you, and welcome to Micron Technology's fourth quarter and fiscal year end 2009 financial release conference call.

On the call today is Steve Appleton, Chairman and CEO, Mark Durcan, President and Chief Operating Officer, Ron Foster, Chief Financial Officer and Vice President of Finance, and Mark Adams, 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 2009 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 an audio replay of this call accessed by dialing 706-645-9291 with a confirmation code of 30898947.

This replay will run through Tuesday, October 6, 2009, at 5:30 p.m.

mountain time.

A webcast replay will be available on the Company's website until September 29, 2010.

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 we will be attending.

Please note the following Safe Harbor Statement.

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

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

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

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

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

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

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

I will now turn the call over to Mr.

Ron Foster.

Ron?

Thanks, Jim.

Our press release is available on our website and it includes a reconciliation of the non-GAAP numbers discussed on this call.

Let me provide a brief summary of the financial results for the fourth quarter which ended September 3, 2009.

For the fourth quarter the Company reported a net loss of $88 million or $0.10 per diluted share.

Our net sales of $1.3 billion and cash generated from operating activities of $357 million.

These compare to third quarter results of a net loss of $290 million or $0.36 per diluted share on net sales of $1.1 billion, and cash generated from operating activities of $151 million.

The results for the fourth quarter reflect improving conditions in the memory markets, particularly for DRAM.

No lower cost or market write down is recorded in the fourth quarter, and gross margins reflect estimated benefits of $91 million and $242 million in the fourth quarter and third quarter, respectively, from sales of products previously written down.

Cost of goods sold also include charges of $37 million and $25 million in the fourth quarter and third quarter, respectively, for idle capacity from Inotera and IM Flash Singapore.

The charges from Inotera are expected to increase over the future quarters as Micron steps further into the capacity previously dedicated to Qimonda.

For the 2009 fiscal year, the Company reported a net loss of $1.8 billion or $2.29 per diluted share on net sales of $4.8 billion, and cash generated from operating activities of $1.2 billion.

Sales in fiscal 2009 decreased 18% compared to fiscal 2008, resulting primarily from declines in average selling prices for both DRAM and NAND Flash memory products in excess of 50%.

Across the same period, sales volume increased over 50% for DRAM and more than doubled for NAND.

Recall that one-third of the way through the fourth quarter we closed on the sale of a majority interest in our imaging business to private equity owners.

Micron will account for the retained 35% interest under the equity method with a two month lag from Micron's fiscal calendar.

Micron has continued to provide manufacturing services including wafer fabrication to Aptina, the results of which continue to be reported in our imaging business segment.

The fourth quarter financial results include one month of Aptina's results with sales to OEM customers, and two months of the Company's results as a wafer foundry for Aptina.

Wafers are sold to Aptina under a volume adjusted pricing model based on process technology.

Since we now have a few less than wholly owned operations, let me summarize how these appear in our financial statements.

Consolidated entities are fully represented in Micron's balance sheet and income statement with the interest attributed to other partners backed out of the balance sheet and income statement in the non-controlling interest line.

Entities in this category are IM Flash, both U.S.

and Singapore operations, Tech Semiconductor Singapore, and Singapore operations, semiconductor Singapore, and our MP Mask operation.

Entities which are not consolidated are accounted for under the equity method, and presented in a single line on the balance sheet and income statement.

Inotera and Aptina are our two equity method investments.

Total sales in the fourth quarter increased 18% compared to the third quarter, primarily as a result of higher sales of memory products.

Comparing the fourth quarter to the third quarter, sales of memory products increased 20%, while sales of image sensors decreased slightly.

Memory sales in the fourth quarter include royalties and technology fees of $34 million compared to $32 million in the third quarter.

DRAM sales increased 28% compared to the third quarter, to $708 million, primarily due to a 19% increase in bit shipments and 8% higher average selling prices.

Sales of NAND flash products increased 10% compared to the third quarter, to $469 million primarily due to a 23% increase in sales volume, partially offset by an 11% decrease in average selling prices.

Imaging revenue was $123 million in the third quarter and represented 10% of the Micron consolidated net sales.

The quarterly DRAM average selling price increased in the fourth quarter for the first time in 11 quarters.

Pricing for core DRAM products as well as specialty DRAM products both increased in the fourth quarter.

The overall 8% DRAM ASP increase in the fourth quarter occurred despite the effect of the continued shift in mix away from specialty DRAM products, which generally command a higher ASP than core DRAM products.

Substantially all the increase in DRAM bit shipments in the fourth quarter came in core DRAM products, primarily due to increased production from our 300-millimeter facilities.

While specialty DRAM primarily produced in our 200-millimeter facilities remained relatively flat compared to the third quarter.

DRAM production costs in the fourth quarter, excluding the effects of the lower cost or market write downs and idle facility charges from Inotera, decreased 21% compared to the third quarter, resulting in positive margin in the fourth quarter.

This substantial decrease in per bit cost was achieved primarily through increased bit production from continued transitions to advanced process technology and the effect of the mix shift towards lower cost core DRAM products.

The supply from Inotera constituted about 10% of DRAM bits produced in the fourth quarter.

The production costs in the fourth quarter decreased approximately 45% compared to the fourth quarter of fiscal year 2008.

Fiscal Q1 of 2010 DRAM cost reductions are expected to be in the low to mid teens, and bit production is forecast to increase in the mid-20s percent range, including the volume we purchased from Inotera.

Looking to NAND in more detail, overall NAND Flash bit costs in the fourth quarter ,excluding inventory write downs and idle costs, decreased approximately 13% compared to the prior quarter, and 62% compared to the fourth quarter of fiscal 2008.

NAND cost reductions in the first quarter of 2010 are expected to be in the low to mid teens again, and bit production is forecast to increase in the mid to high teens as our 34-nanometer NAND process technology achieves mature yields.

The decrease in NAND Flash average selling prices in the fourth quarter compared to the third quarter was largely attributable to the dilutive effect of NAND Flash sales to Intel, our NAND flash manufacturing partner.

At long-term negotiated prices approximating costs.

Selling prices for products sold to Intel from IM Flash declined significantly in line with our decline in costs, as the transition to 34-nanometer NAND process technology was substantially completed, allowing us to achieve industry leading costs per bit for these products.

Trade ASPs on sales to customers other than Intel were relatively stable in the fourth quarter compared to the third quarter.

Our cost reduction efforts in the operating expense areas are enabling us to near our long-term targets for SG&A and R&D as a percent of sales.

SG&A and R&D costs decreased 23% and 17%, respectively, when looking at a year-over-year comparison to the fourth quarter.

Cost reduction activities have included lower personnel costs as well as cancellation and deferral of other spending activities.

Staffing levels in SG&A and R&D areas decreased 9% and 16%, respectively, comparing their level to the end of Q4 to the end of Q3, principally resulting from the spin-off of Aptina.

We anticipate SG&A expense in the first quarter of fiscal 2010 to be around $80 million to $85 million, and R&D expense to be in the $130 million to $135 million range.

Other operating income in the fourth quarter includes a credit of $12 million which is a true up of our estimated loss on the spin-out of Aptina to the final amount of $41 million.

Now to the balance sheet.

Fourth quarter ended with $1.5 billion of cash which includes the cash flow from operating activities of $357 million in the fourth quarter or $1.2 billion for the whole fiscal year.

The growth of our cash balance during the fourth quarter reflects the operating results, including our cost reduction efforts, and our focus on working capital management.

Days sales outstanding and accounts receivable declined from 49 days at the end of the third quarter to 41 days at year-end.

Inventory decreased slightly in the fourth quarter reflecting lower costs per bit, partially offset by a higher level of bits in inventory as production outpaced sales in the fourth quarter.

The cost per bit reductions overcame the effects of selling products in the fourth quarter that were subject to previous lower cost or market write downs.

In addition, finished goods inventories were reduced by $56 million with the spin-off of Aptina, excluding the effects of inventory write downs in both periods.

Total inventory turns improved from 4.2 at the end of the third quarter to 4.6 at the end of the fourth.

We estimate the effects of previous inventory write downs is less than $40 million in the year-end inventory balances.

We paid down $429 million in debt in fiscal 2009.

The year-end balance of short-term debt is $424 million, which is scheduled to be paid down in the coming fiscal year.

Capital expenditures during the fourth quarter were $66 million, and were $632 million for the fiscal year.

Capital expenditures for fiscal 2010 are expected to be between $750 million and $850 million, loaded more toward the back end of the year.

With respect to our joint ventures, in the fourth quarter we made a capital contribution of $60 million to continue the 15-nanometer conversion at Tech Semiconductor.

In addition, net distributions were made to JV Partners of approximately $113 million in the fourth quarter, and $681 million for the fiscal year, which consists primarily of payments from IM Flash to Intel.

With tha, I will close here and turn the commentary over to Mark Adams.

Thanks, Ron.

For the second straight quarter we experienced a strong increase in bit shipments in both our DRAM and NAND businesses.

The bit increase was driven by better than anticipated demand across most of our key markets including computing, specialty DRAM and the consumer market.

In addition to the bit shipment growth in DRAM and NAND, we also saw ASP improve in our DRAM business and continued price stability in our trade NAND business, which when combined leads us to be optimistic about Micron's core memory business through calendar Q4.

Our overall DRAM bit shipments grew 19% quarter-over-quarter.

Revenue from our computing business which includes desktop computing and notebooks and netbooks was up 41% and bits were up 22%.

It's worth noting that we shipped our first product from Inotera Fab in our fiscal Q4.

Coming off a strong Q3 in our server business, we continue to see growth quarter-over-quarter with an 11% increase in bits shipped, translating into a revenue increase of 28%.

Our networking business achieved a 34% increase in bits, and remained a strong market for Micron specialty DRAM products.

All in all, demand for Micron's DRAM products has dramatically increased and has put us in a position where we'll be challenged to meet all of our customer demand throughout the remainder of calendar year 2009.

As I mentioned in our last call, we saw a market shift in demand toward DDR3 memory, for both server and PC applications.

Bit shipments of DDR3 product increased 40% quarter-over-quarter in Q4, and this was after Q3 where we experienced a 70% increase.

Micron technology and cost leadership in DDR3 puts us in a unique position to capitalize on the growth in demand coming from both PC and server base related customers.

In addition to continued strength in DDR3, we also saw an increased demand for our DDR2 components, with an increase in bit shipments of 14% fiscal quarter-over-quarter.

The mobile business saw a strong rebound in bit shipments up 50% quarter-over-quarter.

Our MTP multichip package business experienced an 81% increase from previous quarter bit shipments.

We feel Micron's broad portfolio products aimed at the mobile segment, including MTPs, DS RAM products, NAND-based technology, and mobile memory cards puts us in a strong position to grow significant share in fiscal year 2010.

From a pricing perspective, ASP for DRAM was up 8%, which is especially strong performance given the mix shift driven by the commodity DDR 2 product, which has a lower ASP versus our portfolio average to go along with the lower cost per bit.

In the fourth quarter, our trade NAND bit shipments increased 23% from the previous quarter due to strong demand from the mobile, USB, MP3, photo and SSD markets.

During Q4 we shipped close to 80% of all NAND-based products on our industry leading 34-nanometer technology.

We feel this technology leadership puts us in an industry-leading price performance position vis-a-vis our competition.

Our retail business which includes the Lexar branded flash memory products, the crucial after market DRAM module business, and the crucial dot-com sites, continues to make positive contributions to Micron's overall operating performance.

We've had a very tough year in retail, but Lexar crucial business been able to achieve market share gains, both in the Americas and internationally.

Trade components for NAND memory remained stable in Q4, following significant increases in prices ending Q3.

As I mentioned on our Q3 call, the flash industry continues to migrate to higher density chips, and we feel bit consumption will continue to grow.

We remain optimistic that market segments like mobile, netbooks and SSDs will continue to drive strong bit consumption and that demand market will be in a fairly balanced supply and demand position throughout the remainder of the calendar year 2009.

We remain please with our performance in flash memory and work with constrained market conditions to best serve is our valued customers.

In Q4 we were encouraged by the relatively strong demand signals across most, if not all, of our channel segments.

One month into our first quarter of fiscal year 2010, we have seen stronger demand on DRAM and in price upticks than Q4, with an additional demand from our customer base.

In the meantime, it appears industry supply growth and capital spending remain at historically low levels, leaving many products on allocation.

We remain optimistic these trends will continue, and Micron will enjoy continued momentum in our overall performance.

With that, I will hand it back over to Kipp.

Thanks, Mark.

We will now take questions from callers.

(Operator Instructions).

We will pause for just a moment to compile the Q&A roster.

Our first caller is Tim Luke from Barclays.

Thank you so much.

Perhaps you could just remind us, based off where we are now in the quarter, where you perceive ASPs would be if they remain flat from here, first, for the NAND side and on the DRAM side.

And then separately, as we move forward, could you give us some color in terms of key variables here, say, with the gross margin dynamics given that you're describing a somewhat firmer pricing backdrop and some growth in the bit growth as well.

Thank you.

So the first part of that question, if it remained flat the rest of the quarter, both NAND and DRAM would be up 5% for us.

Can you repeat the second part of your question for me?

Just want to get some color on the key variables for the progression of the gross margin for the overall company.

Sure, Tim.

The way we address that, of course, is with the cost downs, we are looking at both DRAM and NAND, down sequentially in Q1, in the low to mid teens.

Separately, as you move forward you talked about the Tech JD, can you give us a sense of what your plans may be there in terms of capacity in NAND, and perhaps also some color on the progress on 34 and timelines associated with 22.

This is Mark.

Relative to the Tech JV, there are no plans for NAND at the DRAM only facilities, and that's our intent to keep it that way on a go forward basis.

Relative to NAND technology migrations, we are essentially complete with 34-nanometer at this point.

We do have our 2X mode.

I don't want to characterize exactly what that is today, but our 2X mode is in silicon running and manufacturing fabs today.

We would plan to sample that and ramp it in Q1 of 2010, calendar 2010.

Lastly, with the CapEx numbers, could you remind us what you had outlined previously for this coming fiscal year and just give us some sense -- it looks like it is edging up somewhat from the $750 million, $850 million level where that is likely to be principally directed.

.

I think you suggested it was going to be mostly second half

Yes, that's correct.

This is Ron.

We are guiding about $750 million, $850 million, partly related to carry-over from 2009 where we came in at $66 million in the fourth quarter, down from what we had originally estimated.

Some of that flowed into 2010 and it is heavily back-end loaded and it is always a little hard to call exactly how much of that will fall into the fiscal year versus fall out.

This is Mark.

Relative to uses, to the extent there is additional incremental CapEx, it is primarily around optimizing existing facilities, in particular, eking a few more wafers out of the existing operation in both Virginia and Singapore as well as optimizing the mix at IMFT.

Thanks so much.

Our next caller is Jim Covello from Goldman Sachs.

Just a couple of questions around CapEx and industry supply, what are your current expectations for sort of bit growth heading into next calendar year, and where do you guys think you might come out as you intend to pick up share in the industry?

Jim, we won't give you a year-over-year yet for Micron other than what Mark as referenced before for our sequential growth.

But in terms of industry, it seems like most of the third party looks at this or DRAM somewhere up between 35% and 50%, and NAND up somewhere between 60% and 80%.

For people who are -- as you guys commented, the CapEx is at historically low levels and it has been for the last couple of quarters.

So the tight supply is very likely to continue for the next couple of quarters.

How quickly or when is the soonest that meaningful incremental supply can get brought online, if not for Micron, from the rest of the industry?

There's a couple of different parts to that, Jim.

One is how much supply can come back online from the existing capacity that was or still is in place.

Obviously, there is not going to be anything coming out of (inaudible) been liquidated.

With respect to Micron itself, you know what our plans are.

We don't have any silicon come out, or we're mostly optimizing, And then when you look at what's going on with [Alpeta] and probably [Ionics] are in the same boat.

I don't think there's a lot of news, There's a lot of discussion around trying to bring them up to the generation of technology we are already on.

But that's pretty limited.

And then you talk about the Taiwan Inc.

in general.

Mark Durcan just got back from there, he knows probably more than any of us at the moment.

But essentially, we think about promos in power chip, there is just no capital available to do either an upgrade of the current technology or any kind of additional entry into the marketplace.

So it really just leaves what Samsung is doing, which is not a whole lot.

But kind of in line with what they've have always done.

So in terms of new facilities being built for capacity, they're just -- there aren't any that exist, and in terms of optimizing some of the facilities that are in place, there's not a lot of capital for that to happen outside of what we're probably doing and obviously [Alpeta] is out trying to raise some money, but if you look at the debt structure, it is pretty ominous in what they have to deal with in the next twelve months.

So we are not real sure how they do both.

We talked about CapEx being at historical lows or very low.

There has been clearly CapEx at the moment for the DRAM world, no one has ever seen it this low before.

But what is happening is that there's not a lot of capital to try to change that metric at the moment.

So we look at good supply coming from industry in general.

We don't see that much activity that could happen actually any time in the -- let's call it probably one to two years.

So if you're dealing with part of the equation, you have to look at is the similar equipment guys.

And what is interesting there is this downturn has been hard on us, but it was particularly hard on them as well.

They went to a model where they in large part outsourced a lot of their manufacturing and really focused on the technology development and design.

If you look at how that got outsourced, it got outsourced to a lot of these very small, I don't want to characterize them as mom-and-pop shops, but a lot of people in small businesses doing a lot of specialty stuff around powder coating materials and so forth.

And this downturn has -- I mean they're gone.

These people they're just gone.

They don't exist anymore.

That's when you read about all of these 50-year- old, 100-year-old businesses going out.

These people are just gone.

So I actually don't think there's a whole lot of quick recovery in the equipment supply chain.

In the event that there's some additional need for capital, and so I think it -- of course we don't predict the economy, and we can't as to what is going to happen there.

It doesn't take a lot of demand in this environment to consume whatever supply exists for us to come online probably next year.

Mark, did you want to add a couple of comments?

I just want to add more Micron specific things.

First of all, relative to Micron, I've commented on previous calls, I just want to reiterate, we can get a long way on relatively little capital here at Micron, primarily because a lot of work is relatively new and we feel like we have invested that capital pretty wisely.

So that on a go-forward basis, the leverage we can get in terms of technology migration is a little bit better than others.

And in particular, I would point you to the scanner situation which I think is probably a key bottleneck for the industry generally.

And where a lot of the capital dollars are going today .

We feel like we played that about right in terms of positioning our capacity so we have a minimal spend relative to others.

But having said that, it is probably limit the amount that the industry in general can

Thank you so much.

Good luck.

Our next caller is Tristan Gerra from Robert Baird.

Hi.

Could you provide us with an update on the manufacturing (technical difficulty) in terms of when we can expect for the next two quarters?

We are having a difficult time understanding.

Can you repeat the question one time, please?

Sure.

I'm not sure if the question was already asked.

Could you provide us with an update on the manufacturing transition at Inotera and (technical difficulty) the production ramp up in the next few quarters?

Sure.

This is Mark.

Let me handle that and make sure I heard it properly.

If the question is how is the transition at Inotera going, what sort of a schedule for that generally.

We are in pilot production now on our 50-nanometer process.

That is progressing smoothly.

We are happy with the early results in terms of the unit process modules and how those are all coming together.

It is a mixed tool set.

So it is not a direct copy from Micron, but so far we are not seeing any (inaudible) and we're very encouraged with the way that's going.

Our plans are to have our first real significant output in calendar Q1 and be approximately 50% transitioned by the summer.

As we move through the rest of the year it is likely we would complete that towards the end of calendar 2010.

Okay.

And also, can you remind us what percentage of your D1 production is going to be at (technical difficulty) this calendar year?

So by the end of this calendar year, on a -- on a bit basis --

This is Ron.

I can tell you that there was a little bit under 10%, this latest quarter production, and it is going to approximately double in the next quarter.

And then it will ramp up some from there, but it will be a significant jump this next quarter.

Let me add that, one more thought on the technology transition.

I spoke to 50 nanometer a minute ago, it is highly likely that as we get into the summer timeframe we will also be introducing a 4X mode.

Okay, so maybe about the quarter (technical difficulty) kind of year end?

We think we heard you right.

Ten percent in the current quarter we just reported doubling as a percent of total bits outs in Q1.

Thank you.

You bet.

Our next caller is Glen Yeung from Citi.

Just in response to one of your responses earlier, is it safe to say that Micron does not view this period of low capital spending by the industry as an opportunity to maybe get more aggressive here and try to take share?

We have been taking share over the last ,I don't know how many quarters you want to try to measure it over.

The low CapEx numbers, I think, are reflective of the difficulty the industry faced.

And even though we're able to increase our CapEx to some degree, I suppose we could think about doing more of it, but as Mark already noted, as you know from our CapEx for the prior two years, we spent quite a bit of money to position ourselves to have a pretty good leverage model.

With respect to the rest of the industry, most of them, they don't have any money.

They just don't have any access to capital.

And by the way, let me point one more thing out and then you can jump back in.

The equipment financing market, the equipment leasing market was a huge part of what a lot of these companies did when they went out and spent CapEx.

They basically got the equipment companies in cooperation with the GE Capital guys, and those guys to finance it.

That market just closed.

There is no equipment financing available to anybody and if you did -- or if you were able to find some, it's incredibly expensive.

So it's gone, it's just not available.

If you can't generate the cash internally, then you have a very difficult time trying to raise the money, obviously, because of the shape the Company is in and it is just not available.

I think that's why you are seeing in general the low CapEx.

If you are a company like Micron, we actually view it as pretty advantageous to be in the position that we're in, not having to spend a lot of CapEx and already being on the advanced process mode, which is the exact opposite of what a lot of our competitors are facing right now.

Maybe a question on the demand side of the house, thinking near-term about any effect you may or may not be seeing from the pre-Windows 7 build.

And then as you think about next year, and the potential for Windows 7 to be more of a 64-bit operating system.

What are your thoughts on bit per box gross next year, and is it accelerating because of Win 7 or not?

This is Mark, I think we've seen a later back to school drive a lot of that anticipation, and we feel pretty good about it going forward, the type of growth we see in total PC bit per box is in the mid-20s, a lot of that, obviously, is desktop, notebook business.

The netbook business will be confined, obviously, given the architecture of that product.

But we feel good about it, given the current results we are seeing today and what we are hearing from our customers.

Great.

Thanks.

The last question here is thinking about the -- in between the supply and demand, how inventories look to you.

You made the point that you grew a little bit of inventory on a unit basis in the quarter.

What is your sense of overall industry levels of inventory, what you see in the channel and your level of comfort with your visibility on it?

Well, my earlier comments in my message earlier was that it is pretty tight.

We are -- when you see inventory build that you're referring to earlier, it's around us making strategic decisions on serving key markets and mix.

But the overall inventory in the channel is pretty tight, and it's causing us some problems around allocation and how we serve our customer base.

Thank you.

I just might add one other point and that is our inventory on a volume basis was only mildly up and actually finished goods was down.

As I mentioned, our inventory turns are up significantly, so I would characterize the inventory movement as being more of the production volume related to the business versus any buildup.

Got it.

Thanks.

Our next caller is Shawn Webster from JPMorgan.

Thank you.

Can you share with us what your bit production for DRAM and NAND was for your August quarter?

Sure.

We had DRAM bit growth about 20% in Q4, and NAND was up around 10%.

You were talking about just now the tightness in the channel inventory, do you guys have an expectation on when that tightness will be alleviated.

Right now we can't see that far out We think our current supply output and demand profile looks pretty tight at least through the rest of the calendar year 2009 as best we can see .

And so I hesitate to try to guess when that

Okay.

What are your OEM customers telling you to expect in terms of PC DRAM bit demand in calendar Q4 and maybe even calendar Q1 to the extent they have given you some visibility here.

On the demand side?

Yes.

The PC OEMs are right in line with my other commentary.

We've had customers come back and almost ask us for 2X of what their forecasts were just 60 days ago.

That's probably some of of the tightness, but it doesn't stop the PCs, the server business -- our server business, the demand side, has been dramatically up from forecasting in the July-August time frame, and the networking business is very similar.

So it has been tight all over, but the PC OEMs are kind of impressing on us that they need more in terms of output.

And then on the pricing side, you said that if things stay flat with where they are today on pricings for -- I think DRAM demand, you said it would be up 5% or so.

That, when you look at the spot in the contract pricing we observe on DRAM Exchange and sources, it would suggest something significantly higher than that.

Can you walk us through why you think it will be 5% versus 25% plus?

Sure.

Same as last quarter, what we have done is quarter to date we are not predicting we'll only be up 5% for quarter to date.

It's up 5% and it really reflects the increased units coming out of Inotera core memory.

Which obviously have a lower ASP than, say, specialty memory do.

Okay.

Maybe one last one for me.

On the wafer side, what is your wafer production sequentially in Q4?

It was up low teens including Inotera.

Thank you very much.

You bet.

Our next caller is Daniel Berenbaum from Auriga USA.

Hi.

Thank, guys.

It is from Auriga.

Quick question on cost reduction for the year.

A couple of questions may be danced around this.

Can you talk about how much you can reduce costs on a per bit basis for year-end over the course of the next 12 months.

Not at this time.

We will keep you on a quarter to quarter basis for now.

Okay.

Would you expect to be better or worse than industry average cost reduction, and is there any reason we should think about things a little differently for Micron versus your competitors?

It's tough to know exactly what our competitors are going to do, but if I was going to venture a guess on that, I would say we are probably going to be better on both counts, NAND and DRAM.

Is there a reason why you will be better, can you be a little bit more specific?

I'd love to give you credit, but I have to pull that string a little bit.

I think as we move, as already alluded to earlier on, as we move into calendar Q1 we'll be ramping our 2X mode on NAND.

We think that's a significant sale size reduction that could drive out performance on NAND cost per bit reductions relative to our competitors, who we heard anecdotally are struggling a little bit on the 3X mode.

On the DRAM side, we have are a significant increase in low cost product coming out of Inotera, and we think that will drive out performance on the DRAM side.

Just circling back on Aptina, so how should I think about revenue, and maybe you can give guidance even just one quarter out for the Aptina segment, either in pricing or units or both.

And then how is the revenue that's going to be consolidated now, the revenue that's still Micron revenue, the sell-in of the units to the spin off, is that a break even business, what's kind of the structure there?

And do I model that revenue trailing off at some point or how should I think about that moving forward?

Well, let's see, I'm going to let Ron answer.

Essentially think of that converting from a component sales business to a wafer selling business.

So we are still manufacturing the wafers.

And we sell to the partnership on a wafer basis, and that pricing, like you would expect in other foundry arrangements, that pricing will vary in volume.

But that's how we're going to account for it.

In terms of forecasting revenue, that's hard for us to do because no one has been able to really forecast much revenue with much accuracy because of volatility.

Remember, it is seasonal for them right now.

So we would expect this quarter we are in right now to be seasonally strong.

And then what happens after that will be -- depend upon what happens in, primarily, the consumer markets.

So we don't know -- we're uncertain what happens there, but we are manufacturing the wafers and selling them on a wafer basis.

So I think our revenue won't fluctuate as much as you think or expected them to had we been selling the end product to the customer, because we are selling on a wafer basis.

They may fluctuate based on volume changes, but not based on pricing or margin changes.

Ron, do you have anything to add to that?

Yes.

I think Daniel, this is Ron.

In terms of projection, obviously, we are a supplier to Aptina now.

So they're looking to volumes and we are delivering supply into them.

Steve mentioned if you look at the structure, two of the last three months of this last quarter we were on a foundry basis.

So what happened this last quarter in terms of the structure is two-thirds of the way to what you would expect to see, which is somewhat lower gross margin, if you will, and also correspondingly lower OpEx both in R&D and SG&A, which has been sold to the Aptina operation.

So we are two-thirds of the way there in the fourth quarter, and going into first quarter it would be a whole three months worth, but a similar kind of construct would play out.

Okay.

Great.

One just longer term question, and when you talk about guiding for CapEx, the $750 million to $850 million, and obviously fiscal 2009 came in a little bit lower than expected, it is a lot lower particularly in terms of CapEx to sales than it was, say, in fiscal 2006, fiscal 2007.

How do you think about what a sustainable capital intensity level is for Micron, and then maybe more broadly for the memory industry.

Well,I think fundamentally the answer to that question is primarily associated with how long do you think the equipment will be -- what's the useful life of the equipment.

We have done studies historically in the industry, and that number was in the neighborhood of 3.6 to 3.8 units.

So that means you in general will have a capital replacement every 3.6, 3.8 years, and so you can do the math and that says you need to spend 20% of your revenue or somewhere in that neighborhood to continue to stay up with the innovation of the equipment.

But that curve, I think most people think that curve is probably slowing.

And that a lower CapEx number will sustain a more useful life, if you will, of the equipment.

So it is hard for us to know what the number exactly is.

If you look at the data, on average, that number -- we're in an historical low this year, probably in the neighborhood of maybe 11%, 12%, at least for the memory industry, and I am merely speaking about the memory industry, not the DRAM industry, and you've seen it as high as 22%, 23% and everything.

I think you would expect to kind of settle out somewhere between -- in our particular case, you can run the math and obviously we were down, too, like everybody else.

But I think in general, the precise answer is going to be dependent upon how long the useful life of the equipment is and how long it takes us to make these generation switches.

Our next caller is Vijay Rakesh from Thinkequity.

This quarter for a couple of questions, I'm wondering what the bit build should be for NAND and DRAM as you're looking at the next quarter here.

And also, what were the licensing revenues for the August quarter and where do you see that as you look at fiscal 2010?

So, the first part of that bit growth, I think we are looking at up mid 20% for DRAM, and mid to high teens on NAND bit growth.

I don't think we caught the second part of your question.

Just the licensing revs, what it was for the August quarter and how do you see licensing kind of going as (inaudible) also picks up, how do you see that for fiscal 2010?

This is Ron.

We had $34 million in licensing revenue in the fourth quarter, $32 million in the third quarter.

That was related to technology fees.

Going forward we expect to see ongoing technology fee licensing as we have seen here in recent quarters in 2009.

In addition, as we migrate Inotera to stack technology, there will be licensing fees coming along on a volume basis or per unit basis related to the ramp of that production used by others.

So that would come on as the stack technology transition occurs as Mark articulated earlier.

Got it.

And as you ramp up Inotera as a shared output, what do you see as a stable OpEx level for Micron as you look forward a couple of quarters here?

Well, we said that historically we said that we think in a normalized market, not one that could rapidly increase or not one that's rapidly declining, we think an OpEx model that would be benchmark is somewhere in the 50% range total.

And obviously that will be split between SG&A and R&D depending upon what's going on at the time.

But essentially think of it as I think targets in the neighborhood of, oh, call it 10% or 11% on R&D and maybe between 5% and 6% on SG&A, and the combination of trying to get to 50% as being a benchmark.

It was tough to achieve that with obviously falling revenues, but we think that again is probably a right level of benchmark for the industry.

Okay.

And moving over to the NAND side, you said the 30-nanometer transition going on, you're ahead of the industry.

Where do you see 30-nanometer output in, let's say, March quarter, June quarter of next year.

What should that be as opposed to the rest?

I think we are going to hold off on any output numbers relative to that, until we will see how well -- how we're (inaudible).

Okay.

Good.

Thanks a lot.

Our next caller is Uche Orji from UBS.

Thank you very much.

I hope you can hear me.

Let me just start by asking you, I am still not sure I am convinced as to why bit per box will not at least in the near term come down.

Now I understand there's supply shortage, but at the same time, it would seem that the ASPs for PCs coming down quite significantly in a market also where DRAM is rising, (inaudible) have been very expensive.

So how vulnerable should we think about your bit per box comments in light of the fact that your OEM margins are struggling.

And if you don't think DRAM is vulnerable, what components do you think are vulnerable and can provide some for what the OEMs are that may be (inaudible) right now because of the declining ASPs in light of (inaudible).

And that's my first question.

So let me try and summarize what you are asking and turn it over to Mark for the answer.

It sounds like you are interested in the bump to the cost of memory today versus historical, and is there something pending that might trigger that.

And B, kind of the general overall question is the convincing you of the growth of memory per box.

Is that --

That is correct.

Yes, maybe on the [bellum], there's, I think, a couple of different ways to look at it.

One is that there's no question that the memory content today is lower than it has been in the prior few years.

It was right in the 8%, 9% range.

Today it's more around the 4% or 5% range.

If you're looking at netbook, it's actually still running about 3% or 4%.

And in desktops, characterize it as something in the 5%.

But if you go back to the 2005, 2006, 2007 time frame, it was up around 8%, maybe 9%.

So it could be when the economy struggling and as a result margins are struggling.

So what's the appetite to keep it where it's at today or have it slightly go up?

I don't know.

I think it is segment dependent and customer dependent.

And it is, I think, more predicated on whether it's enterprise.

Which the enterprise guys seem to be doing pretty well, as opposed to the guys on the consumer side.

And I guess I'll turn it over to Mark and let him comment on that.

The only other comment I would make is, while we have seen some positive movement from a pricing standpoint, we need to go back a year ago, last August and we are still not at the prices of that level.

And the density per box was slightly below where it is today, so before we over-react to increasing pricing on DRAM per box, I think it is fair to say we are right where we were a year ago and certainly from the -- around the silicon, but I don't think we are seeing a dramatic shift away from the densities, I think it's a little bit ahead of ourselves here, and certainly from the demand side that all the signals we are getting are consistent with that.

All right.

That's helpful.

Let me just ask a different question.

If I look at the inventory comments you've made.

It looks like the OEMs, ODMs may not have a lot of inventory.

But can you just tell me what is happening, what your view is as to inventory and sales (inaudible), I'm talking about module guys, distributors, all of your companies and OEMs.

So can you just give a comment as to your baseline?.

My sense is that given the pressure on the OEMs, what you're referring to is the aftermarket channel.

That's even tighter because the OEMs are probably getting preferential treatment in most cases.

So my suspicion is that the aftermarket module markers, card markers, are in a tighter position, and thus one of the reasons why you see a spot marking pricing higher than where the OEM contract prices are.

Right.

Right.

Okay.

And then just finally, are -- in context in terms of the availability of equipment, the ability to spend CapEx, especially (inaudible).

Do you have any comment about the longer times to deliver equipment, is it a concern to Micron?

It is not a concern for us currently as we have got a pretty good forecast out there and a good relationship with our suppliers.

So we think we have got the slots we need to execute to our plan.

l would say for people coming in now with new plans and trying to get new slots, it is not going to be easy because that is a constraint in the marketplace.

Perfect.

Thank you very much.

Our next caller is Hans Mosesmann from Raymond James

Most of my questions have been answered, but just the one regarding -- the situation regarding shortages.

I think Mark may have mentioned there were some shortages.

Can you explain, give us more granularity, in terms of what was the shortage, was it DDR 3, NAND and what flavors, and so on.

Thanks.

Sure.

So I think where we have seen shortages early on was in some of the specialty and server market opportunities, and what we have seen lately has been consistent with that staying -- of course, we were a little tight there, but then a pretty heavy request for DDR 2 parts from one of our commodity businesses.

So on the DRAM, we started out with DDR 3 and it looks like DDR 2 has caught up a bit.

On the NAND side, it is been pretty tight as well.

You have seen a little more up tick both on the mobile side of the business, and I would suggest that the SmartPhone business has been the healthier of all the segments , and that's where the density on NAND has driven some appetite in a couple of those supply situations.

But I would say across the board it's been pretty tight.

DDR 3 first, DDR 2 caught up with it, and there's still a lot of unmet demand there, and the NAND side has been more consistent, throughout the back half of the

Okay.

Thanks.

And then one last one, DDR 3 in terms of Micron's overall mix and how does that compare to industry mix?

That's it.

Thanks.

So, of our core commodity DRAM parts, DDR 3 comes in about 35%, and that's good to note though that is still a pretty -- you have to factor in the Inotera piece of that, also would have been higher pre-Inotera.

Okay.

If you going back a quarter, you have to factor that in this quarter.

Okay.

How does that compare to industry mix per quarter?

My sense is based on our mark-to-market channel, our mix is significantly higher than our competition.

Okay.

Fair enough.

Thanks.

Sure.

Our next caller is Bob Guiavarty from Deutsche Bank.

Thanks, guys, for taking my question.

Just curious under utilization in Singapore -- you mentioned Inotera was -- under-utilization of Inotera was declining, how about the Singapore side, will that be relatively constant or how do you see that coming out?

We are fully utilized in Singapore, and have been.

We are continuing to eke a few more wafers out of the fab as we optimize process times, equipment utilization, et cetera, but that fab has been running fully utilized.

Relative to IMSS, that fab is in an idle state.

Okay.

So for IMSS we can think of some modest under-utilization continuing for at least the next quarter or two?

It's -- yes, it is empty and it will stay that way for sure for a quarter.

Okay.

It would take us a lot longer than if we already had things in place.

A quick question about your view is about fungible capacity between DRAM and NAND.

It seems to me there's a less of that given the technology nodes each is on.

But can you talk about if that's an issue that affects supply and demand across DRAM and NAND now, and for Micron and the industry?

It is still possible to flex capacity back and forth with enough lead time to plan.

That's a relatively small amount of capacity.

It is not sort of wholesale that people would be able to switch back and forth.

Because there is a significant divergence, I think, as we move to more advanced nodes in terms of the tool sets that are required.

So while, at the margin you may be able to flex small amounts of capacity to take advantage of market opportunities, if you're running both in the same fab as we do in Virginia, it is a relatively limited opportunity.

Great.

Thanks a lot.

You bet.

With that, we have time for one more caller.

Our final question is from Bill [Develom] from Titan Capital Management.

I'm going to try to make it two question, please.

The first one, you may have partially addressed the floor, but I don't have great -- feel like I have great clarity still yet.

This quarter you had mid $60 million CapEx run rate, and if we take the midpoint of the fiscal 2010 CapEx of $800 million.

That's roughly $200 million on quarterly basis, and you said it would ramp as the -- as the year progressed.

What is it you are going to be doing with that significant increase in capital as the year progresses.

And secondarily, I think, to Mark to talk about Taiwan since you were just there, and like your insights as to where the various companies are at in the process of thinking about consolidation or not, and how the government is thinking about it also.

I will take the first part and I'll let Steve take the second part.

Relative to CapEx, I think first of all I think that the $60 million in the current quarter is probably a slight aberration and abnormal depressant relative to the ongoing spend we have been experiencing.

But having said that, where is the money going.

It is probably roughly one-third to NAND and two- thirds to DRAM.

And it is the vast majority is around moving technology forward to either 2X node on NAND or 5X or 4X node on DRAM.

And in terms of where that spend is going, as I have alluded to the advanced step or capability is expensive and adds up quickly.

So that's a significant piece of it.

If you want me to take the second part, too, relative to what's going on in the competitive arena, clearly in Taiwan, everyone is pretty much back on their heels with the exception of our partners, who I think have renewed energy as they look at how the technology deployment is going.

So they feel pretty good about that, and you can look for a resurgence from both (inaudible).

[Hinex], they clearly have spent some capital.

Whether they can get efficient use of that I'm not 100% convinced yet.

You guys probably have a better read on that than me.

And LP is a challenge relative to their technology position.

Samsung, of course, is always a very strong competitor.

Bill, it is probably worth adding that we were talking to a lot of the investment community in our most recent financing.

I would comment at the time, I think it is even more so now, that it doesn't appear that there's some grand large activity that's going to happen in Taiwan with the government participating in terms of supporting all these companies and trying to figure that out.

(Inaudible) but for any dollars.

So as Mark said we feel great, actually, about the relationship we have and what we have got going there.

And we think that's clearly going to emerge as the leading entity in Taiwan for for manufacturing, and that the others are terrifically challenged in this environment and are likely to continue to be for quite some time.

With that in mind, are you inclined to be thinking that the corporate activity within Taiwan has already taken place, meaning that your tie-up with Inotera and NAND and really nothing else is going to take place for now?

Well, I think that is -- I will let you make those comments, Bill.

I think it is hard for -- we don't want to speak on behalf of Taiwan or Taiwanese government, , but I think that the recent investment in LPN from the Japanese government and with the pretty solid relationship that we have with NAND in Taiwan, I think we are viewed as -- and that entity is viewed as a good solution for Taiwan in terms of having access in development technology for that.

And there'scertainly less, if any other need, to do something else.

There might be something on a lower scale.

So, I think that is probably a pretty good basis for how to think about

That's helpful.

I would like to follow up, Mark, in terms of the comments that you made relative to your CapEx and also tying that in with comments that have been made sporadically throughout this call.

Would it be fair to say you are believing that over the course of the next year, that you will be improving your competitive position relative to where you are at today versus the industry, both on the DRAM and the NAND side of the house?

That would be my view.

Thank you, both.

Thanks, Bill.

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

If you will 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 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 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-K and 10-Q.

Thank you for joining us.

Thank you.

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

You may now disconnect.