美光科技 (MU) 2008 Q3 法說會逐字稿

Good afternoon.

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

At this time, I would like to welcome everyone to the Micron Technology third quarter 2008 financial release conference call.

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

After the speakers remarks there will be a question and answer period.

(OPERATOR INSTRUCTIONS).

Thank you.

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

Sir, you may begin your conference.

Thank you.

I would like to welcome everyone to the Micron Technology's third quarter 2008 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 Mike Sadler, Vice President of Worldwide Sales.

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

If you have not had an opportunity to review the third quarter 2008 financial press release, it is also available on our website again at Micron.com.

Our call will be approximately 60 minutes in length.

There will be an audio replay of this call.

You may reach that by dialing 706-645-9291, with a confirmation code of 50696269.

This replay will run through Thursday, July 3, 2008, at 5:30 p.m.

Mountain Time.

A webcast replay will be available on the Company's website until June 26, 2009.

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

Please note the following Safe Harbor Statement.

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

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

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

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

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

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

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

(end audio recording)

I will now turn the call over to Mr.

Ron Foster.

Ron?

Thanks, Kipp.

For those listeners who haven't seen our Press Release which is available on our website, and includes a reconciliation of the non-GAAP numbers discussed on this call, I will review the summary of financial results for the third quarter which ended May 29.

For the third quarter revenue increased 10%, and gross margins moved into positive territory, as megabit sales and cost reductions outpaced ASP declines.

DRAM and NAND ASPs declined approximately 5% and 20% on a sequential quarter basis.

Megabit sales volume increased approximately 10% and 40% for DRAM and NAND Flash memory products respectively from second to third quarter.

The Company continued to make progress in driving down production costs, as cost of goods sold per megabit declined about 15% and 25% for DRAM and NAND respectively over the same period.

Wafer output, yield increases, and node transitions all contributed to the margin improvement.

The Company recorded a net loss of $236 million, or $0.30 per diluted share for the quarter, compared to a non-GAAP loss of $314 million, or $0.41 in the prior quarter.

The GAAP reported loss for the second quarter included a goodwill impairment charge of $463 million.

We are pleased with our progress in reducing the operating cost structure, as we continue to focus on driving down SG&A and R&D expenses.

R&D expenses improved in the third quarter, due to cost management efforts and progress moving key development products into production.

We expect to maintain the lower run rate of operating expenses, with SG&A and R&D expenses estimated to be flat to slightly down from the third quarter to the fourth quarter.

Operating cash flow was $217 million for the third quarter, and 775 million for fiscal year-to-date.

We ended the quarter with $1.6 billion in cash and short-term investments.

During the third quarter, TECH Semiconductor entered a $600 million credit facility, to refinance $240 million of outstanding debt, and to finance future capital expenditures.

Capital expenditures totaled $577 million in the third quarter.

Capital expenditures for fiscal 2008 will be at the upper end of our previously projected range of 2.5 to $3 billion, as we continue our investment in NAND and IMFS, and in DRAM at MTB and TECH.

Partners will provide net capital contributions of about $230 million to fund our 2008 expenditures.

In fiscal 2009, capital expenditures of approximately 1.5 to $2 billion are anticipated.

We expect about $300 million of this amount in net capital contributions from partners.

These capital expenditures do not reflect the previously communicated $550 million capital contribution for MeiYa, which is committed over the 2009 calendar year.

Depreciation and amortization expense was $513 million in fiscal Q3, and will total approximately $2 billion for the year.

In fiscal 2009, depreciation and amortization expense of about 2.25 billion is expected.

I will close there, and turn the commentary over to Mike.

Thanks a lot, Ron.

In 2007 and early 2008, we experienced a supply driven semiconductor memory pricing downturn, in the face of a relatively strong demand environment.

From our perspective, exiting fiscal Q3 of this year, we are seeing an improved supply/demand balance.

This has generally resulted in relative memory price stabilization, with an encouraging upward pricing trend across the commodity DRAM product portfolio in particular.

The computing industry continues to be the largest revenue driver for Micron's memory products, and we are seeing demand in-line with normal seasonal expectations.

Although we don't have perfect visibility on the specific demand drivers within the computing space, I would characterize memory content per system growth as being particularly strong, relative to absolute PC unit growth.

We are starting to see some traction for DDR3 memory in select platforms, and would expect increased penetration as we move through 2008, and into next year.

We are well positioned in this space with our industry leading DDR2 and DDR3 chip sizes on 78-nanometer and 300-millimeter wafers.

We have begun a transition to 68-nanometer in our Singapore facility, with initial production shipments and revenue generation occurring in the current quarter, ultimately contributing to aggressive cost of goods sold reductions.

All-in-all, we are pleased with our position in the computing space, and given the recent trends, encouraged about prospects for the second half of the calendar year.

Our specialty DRAM product portfolio continues to enable solid performance for the Company in a variety of non-computing segments, such as networking, mobile handsets, and automotive.

We are utilizing installed 200-millimeter capacity, to manufacture both image censors and the vast majority of our specialty DRAM products.

We are selectively moving high volume and high growth specialty DRAM products to 300-millimeter facilities, to achieve further cost reductions, and create headroom for supply growth.

From a market segment viewpoint, we did see particularly strong unit shipments and revenue growth, in both our image censors and low power DRAM products for the mobile handset market in the third quarter.

This speaks both to the health of the market, as well as to the strength of Micron's product portfolio for this particular market segment.

The NAND market is still dominated by consumer applications with a strong seasonal component of demand, continues to be challenged by excess supply.

Our technology and scale advancements now moving to an industry leading 34-nanometer process, are having a positive effect on cost of goods sold reductions, but the oversupplied market environment has continued to drive prices down further.

We are working diligently on strengthening the NAND product portfolio, much as we have already demonstrated in the DRAM space, to improve our average selling prices and relative profitability, independent of the market environment.

To this end, we are ramping up our NAND based multi-chip package efforts for the mobile phone arena.

We have a microSD compatible 1-gigabyte chip in the market today, and have recently begun shipments of microSD cards with Micron manufactured silicon.

We have initial single-level cell solid state drives in the evaluation stage with various customers in the enterprise space, and will introduce lower costs MLC-based solid state drives for the client markets later this year.

We are achieving higher average selling prices and margins via leveraging of the Lexar Brand in retail channel presence, and are continuing down the path of moving more of our NAND output through the Lexar operation.

We are proud of our operational performance and recent trends on key metrics, such as cost of goods sold reduction, and bit and market share growth, particularly relative to other players in our competitive space.

Looking to the second half of the calendar year, we are enthusiastic about improving market environment.

I will turn it over to Mark.

Thanks, Mike.

Micron had a strong quarter operationally.

Our bit cost reductions ran substantially ahead of projections, lead by solid yields and cost cutting in the operations.

The ongoing transition in Micron's NAND and core DRAM products leading it's capacity continue to pace, with 90% of the silicon now being built on 300-millimeter wafers.

Technology transitions also progressed well.

TECH Semiconductor continued it's ramp of the industry leading die size 1-gigabit product on our 68-nanometer technology, while making substantial progress on it's conversion to 300-millimeter.

During the quarter they completed the fabrication of their last 200-millimeter wafers, and expect to complete the transition during the second half of the year.

We are well-positioned with optimized products for the DDR3 transition coming.

Recently Micron and Intel announced sample availability of our industry-leading 34-nanometer 32-gigabit product.

That product and process node is now being deployed in our IMST joint venture, and will ramp throughout the second half of the calendar year.

We believe this positions us well for continued NAND cost reductions moving into calendar 2009.

Recently we closed on the creation of our DRAM joint venture, MeiYa.

The relationship with Nanya brings many advantages, including joint technology development and licensing of technology, as well as the capital efficient capacity associated with the joint venture.

MeiYa will start up it's initial manufacturing facility beginning early in calendar 2009, and ramp throughout the year, to an initial target of 10,000 wafers per week.

We anticipate this to be a cost effective way for Micron to add wafers in support of growing customer demand.

Licensing revenue and cost benefits grow throughout time as MeiYa grows capacity, and Nanya or transitions wholly-owned capacity, and as the scope of our R&D cooperation grows.

Looking to the current quarter we expect bit production growth of mid to high single digits for DRAM, and mid-teens for NAND.

Cost per bit reductions should mimic that, and be in the mid-single digits for DRAM, and mid-teens for NAND.

I will turn it back over to Kipp.

Thanks Mark.

We will now take questions from callers.

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

With that, we would like to begin the questions.

Thank you.

(OPERATOR INSTRUCTIONS).

Our first question is coming from Tim Luke of Lehman Brothers.

Please go ahead.

Thank you so much.

Kipp, could you just clarify how you would perceive bit growth in both DRAM and NAND, and if you were to see stable pricing from here, how you would expect the ASP direction to trend through the coming quarter?

Tim, this is Mike.

I will speak to the ASP topic, and I will let Kipp address the bit growth question.

Rather than answer the question directly, I will just tell you that quarter to date, we are 3.5 weeks into the current quarter.

Our DRAM ASPs on shipments are tracking right about flat, with respect to what we achieved in fiscal Q3, and our NAND Flash average selling price to date in the quarter, is about 20% below what we saw in fiscal Q3.

And on the production side, Tim, we are looking at NAND being in the mid-teens sequentially, Q4 fiscal over Q3 fiscal, and DRAM to be up mid to high single digits.

Could you also comment, Kipp, just on your expectations with respect to cost reduction, with respect to the new core businesses?

Yes, sure, as Mark mentioned they will be similar respectively in each one of those as the bit growth is, so mid-teens on NAND, and mid to high-single digits on DRAM.

Could you perhaps also a more strategic question, just update us on your perception of the Intel JV, and how you might view the degree of committment around that, and what other options there might be for you, should Intel decide to pursue other options with respect to the venture?

Well, this is Steve, Tim.

There is a lot you had in that question.

Yes.

First of all, our perspective is the joint venture is running well.

I haven't had any dialogue with Intel executives that they feel any differently, so a lot of I think what is out there in the market, whether it is speculation or somebody else is making comments other than what we are involved with, I can't really comment on, because we continue to move forward with the joint venture as we always have.

I will just point out that the agreement that as we announced originally when we did the joint venture with Intel, it is a 10 year agreement, so we are still in the first couple of years of that.

If I may, Steve, just as a follow-up, could you give a comment on industry consolidation, and opportunities for Micron therein?

With respect to which --

Just with respect to Qimonda, Inotera, and other possible opportunities?

Yes, I see.

Well I have been commenting on this for 20 years now or so, and always what I thought would happen doesn't happen, and what I don't think would happen sometimes happens, so we have said many times, that as opportunities present themselves, we will consider them, and today is no different than it was a year ago, or five years ago or 10 years ago.

But it obviously requires more than just Micron, and to the extent that there are opportunities where there are other willing partners, whether it be joint ventures like we just did with Nanya, or something else in a different form, we are certainly willing to consider it.

Thank you very much.

Good luck.

Thank you.

Our next question is coming from Shawn Webster of JPMorgan.

Please go ahead.

Yes, thank you for taking my question.

Can you talk about how you expect your operating expenses to evolve for the rest of this year?

This is Ron.

As I commented, the SG&A and R&D costs have come down in Q3.

We are on a good trend line on an absolute basis, and we expect as we go into the fourth quarter, that they will be flat to slightly down both in SG&A and R&D in the fourth quarter as well.

Okay, thank you.

And then on your wafer production, what did it grow sequentially in Q3, and what do you expect it to grow in Q4?

Shawn, we were up a couple percent in Q3, and we are expecting another couple percent in Q4.

Okay, and then finally, in terms of channel inventories, can you talk to us a little bit about what you are seeing out there in any quantitative way you can, for both, well all of your DRAM and NAND and sensor markets?

Sure, I would be happy to.

On the DRAM side, my view of things is very little inventory on our shelves, as a matter of fact we are bumping up against the minimum that we think we can carry, and still do an adequate job of servicing our customer, so the demand environment is quite strong on the DRAM side.

With respect to the inventory levels that our customers have, as I have said many times there is very little incentive for our customers to carry inventory, unless they want to do some price hedging, and those questions would probably be better addressed to our customers, but just based on my view of things, I wouldn't imagine there is much inventory in the hands of our customers, in terms of the channel so the minimal value adders and the traders, and so forth, my view is that the market is being fairly semi traded right now, and there is probably little inventory in place, at least in the DRAM arena, with respect to the speculators.

The NAND side is a little bit different.

I do believe that the spot market traders, if you will, have probably accumulated a fair amount of inventory.

My view of things is there is probably a fair amount of inventory in the spot market with respect to NAND Flash.

We are pretty flush with respect to our own inventories.

We have got 1 to 2 weeks worth of finished goods in the NAND area, and I don't believe our customers are carrying much inventory either.

In terms of image censors, we are a little heavier on finished goods inventory.

We have got about, if I remember correctly, somewhere between 6 and 7 weeks of finished goods inventory on image censors, that is relatively flat with where we exited the prior quarter.

We are expecting a pretty strong seasonal demand uptick as we move through the second half of the calendar year.

I don't believe our customers are sitting on any inventory, with respect to image censors, so that is kind of my color on the inventory situation.

Finally and I am sorry if you said this already, I had to leave for a fire alarm, what do you expect pricing to do if it stays flat for NAND and DRAM in your fiscal Q4?

I answered the question already, just real briefly --

Okay, sorry.

Yes, you can probably pick it up on the transcript.

Will do.

Thank you.

Our next question is coming from Daniel Berenbaum of Cowen and Company.

Please go ahead.

Hi, thanks for taking my call.

Can you talk about what percentage of your mix was DRAM and NAND in the quarter, and also, it seems from the numbers that DRAM was profitable.

Can you comment on that, and then also, can you talk a little bit about your NAND bit growth guidance?

It sounds like you guided for kind of mid-teen percent bit growth.

I thought the big comment on your Q1 call was to think about 30 to 40% sequential bit growth over the next few quarters, so I am wondering if something has changed there, other than the Singapore push out?

Sure, let me start with a revenue break down for you, Daniel.

We had core DRAM in the high 20% range, specialty DRAM in the low 20% range, imaging as we noted on the press release was around 11%, and we had NAND in the mid to high 30% range.

And on the profitability, we did note the last quarter that we felt we were probably 1 of 2, if not the only DRAM manufacturer to keep positive gross margin, and of course with the data we provided about our cost reductions, relative to our ASPs, we certainly improved on that, and I might let Mark talk more to maybe the NAND bit?

Yes.

Dan, on the NAND bit growth, it is going to be in the mid to high teens on a quarter-over-quarter basis going forward, but it will be a little lumpy.

We may exceed that in certain quarters.

As you will recall, a couple, a quarter ago I think we outran our estimates substantially, pulling some bits forward, and that actually happened a little bit this quarter as well, so just I think sort of the running mid to high-teens number is a good one to think about, but think about it exceeding that on a quarter-over-quarter basis, depending on exactly where we are at relative to technology transition.

So mid to high teens, not 30%, did I just misunderstand that from previous quarters?

As a good run rate going forward?

No, well, yes, you have got both references correct.

In the past we have used the term "averaging" around 30% over the next several quarters, and as Mark is pointing out sometimes we will get that pulled in as the manufacturing folks execute better, and that makes the next quarter's comp a little bit more difficult, but we will still have a pretty tall year-over-year bit growth as we go into '09.

Okay, and then can you just help me understand how much of your NAND production goes to Lexar products now?

Yes, it is about 20% for the quarter.

20%, in that assumes that 50% goes to Intel, and another 20% to Lexar, and 30% elsewhere?

That is correct.

Okay, great.

Thank you.

You bet.

Thank you.

Our next question is coming from John Pitzer of Credit Suisse, please go ahead.

A couple of questions here.

First when you look at your FY09 CapEx budget of 1.5 to 2 billion, what kind of market assumptions are you you making around that number, and what kind of growth can you see in both the DRAM and NAND output relative to that CapEx number?

Well, first of all the CapEx number, I mean, the reason we have these ranges is because we constantly look at this quarterly, and it could go up or it could go down.

I think what we looked at on a long term basis on our CapEx number frankly isn't so much based on what we think the market is going to do next year, because obviously we will adjust to that if we need to.

It is really around either planned expansion or planned technology conversions that we have, that obviously needed to be executed upon in terms of planning much before we ever get there, so that is why we have the ranges that we do.

So is there a larger portion going to DRAM or to NAND, or how does that break out?

Well, again, it is a little hard, right off the top of my head to break that out, because remember we essentially have two projects in '09, in addition to what we would be doing on a normal basis among the wafer fabs.

There are the IMFS and there is MeiYa, and those will both be in '09, at least calendar '09, so the timing of those are obviously independent of each other, and it is just a little hard for us to know exactly when one hits versus the other.

And then guys on the NAND side, can you talk a little bit about the solid state drive market especially going into enterprise, and I guess help me understand at what point will solid state drives be a big enough percentage of your NAND business to start breaking it out, and at least telling us 10 to 15, or 15 to 20% range?

Well, it is critical to us from a strategic standpoint certainly, and we have got products both in the enterprise space, as you noted as well as in the client space, primarily for the notebook computer market.

There is no revenue generation whatsoever today for us from either of those markets.

In terms of it being meaningful, it is a little bit subjective, but we are probably looking at into well into next year, before we feel like it's meaningful, in terms of a revenue standpoint.

I don't know that we are going to break it out publicly, but we are looking at it being meaningful for us sometime in the 2009 timeframe.

And then guys my last question, DDR3, what percent of the calendar second half capacity on DRAM do you think will be dedicated towards DDR3, and if you could help me understand the die size penalty that might exist, as you migrate to DDR3, that would be helpful.

Sure.

Our in terms of our output in the second half of the calendar year, we have got by the way plenty of flexibility to dial it up or down, because it utilizes essentially the same tool set in the fab, but we are currently targeting high-single digits of our DRAM output as being DDR3 in the second half of the year, and the penalty that we pay on the same density, DDR3 chip versus DDR2 at 78-nanometers, is about 3 or 4%, something like that.

Great.

Thanks, guys.

Thank you.

Our next question is coming from Glen Yeung of Citi.

Please go ahead.

[Peter Karazeris] for Glen Yeung.

Earlier you were asked about potential consolidation, specifically around Qimonda and Inotera.

I realize that it you get this question quite often, and you have looked at it before.

I guess if you could layout for us some of the pros and cons, as far as taking on Qimonda, and then maybe separately just taking on, or Qimondas taking Inotera.

It is almost impossible for us to do that, because we frankly don't have any kind of inside knowledge on what exists at Qimonda, Inotera, et cetera.

So we have to, we will just have to look at that opportunity if it were to surface, but I couldn't lay anything out for you, because I don't know what it would look like at the moment.

Okay, and then you talked about specialty memory, it sounded like image sensor, demand for image censors, and demand for low power DRAM was pretty healthy.

Could you give us any sense of what kind of participation you have in embedded NAND, I guess there have been some, obviously there have been new smart phones launching, fairly high density, fairly large amount of NAND in them.

If you could just talk about how those are impacting overall market demand?

Yes.

They are very important to us strategically particularly because, we are one of just a couple of semiconductor makers that have the ability to package it's own silicon in a NAND and low power DRAM multi-chip package.

We have got quite a bit of activity with virtually every handset maker under way today, in terms of designing it and qualifying our products.

In terms of revenue contribution today, the NAND-based [MCP], which I believe you referred to as embedded NAND, are insignificant in terms of revenue generation.

And then lastly, can you give me a sense of, as you look at second half capacity for DRAM, what kind of assumptions are you taking in for the market, as far as what Samsung's capacity build would be, going in Q3 and Q4, calendar Q3 and Q4?

Probably the best we can do is quote those industry experts that are out there, and largely what we have read is that Q2 marks the high water mark for sequential bit growth in the DRAM industry for calendar '08, so you would be looking at calendar Q3 and Q4 to be lower than what we just saw in calendar Q2.

And in general, most of the "experts" are ranging some where between 55 to 65% total bit growth production year-over-year for the industry.

You don't have any thoughts one way or another on what Samsung or any kind of your equipment, discussions with equipment manufacturers, what you think Samsung might be doing in the back half?

No, and the obvious answer it is probably best to just ask them.

Sounds good.

Thank you, I appreciate it.

You bet.

Thank you.

Our next question is coming from Daniel Amir.

Please go ahead.

Yes, thanks for taking my call.

First of all on the 34-nanometer production of NAND, what percentage of your production in the second half of the year will be at 34-nanometer, and kind of what is the outlook there for the first half of next year?

Well, Dan, it is Mark.

We will have to dial that as the ramp progress and see how we are doing, but on a wafer basis, we would expect to be 50% transitioned at the end of the year, something on that order.

50%?

Okay, and then the second question, related to the Nanya opportunity, and what they have discussed so far, when should we start seeing kind of a kick-in on the royalty or payment front, that would really start impacting your model?

I mean, is it second half of calendar year '09?

Is it before that?

I think that in two different terms you would see some of the up front payments upon close, I believe it is, and then royalties obviously would be associated with the ramp of either the Nanya converted capacity, and/or the joint venture.

And in terms of the timeline?

I believe Nanya had quoted first half '09 for their conversion, their captive capacity and second half '09 calendar for the joint venture.

Okay, and then third question related to the image sensor, can you give us an update where we potentially stand, and I guess the separation of the business, and also follow-up of the question to that, your margin has obviously increased quite nicely there this quarter.

Can you comment I guess a bit about the mix there, and what were the factors?

Yes.

First of all, for that business as you noted, it did improve, and that is I think largely reflective of the image business migrating more towards the higher end pixel density, and obviously, it was 2 meg, and I think as we move throughout this Fall, second half will probably be dominated by 3 meg, and we were very well-positioned in that space, and clearly, that business bottomed really a couple quarters ago, and if you noticed our results compared to some others, we have continued on a pretty good path of recovering, I think where we had lost for a couple of other reasons out of the marketplace historically, so that is actually going pretty well.

We know what our targets are for that business, and as we have noted before, we are exploring with potential partners how we can achieve those targets, and I would just say that we hope to accomplish, or I suppose another way to frame it was complete, something some time in this quarter, in terms of separation.

Okay, thanks a lot.

Thank you.

Our next question is coming from Vijay Rakesh, please go ahead.

Back to the licensing question with Nanya.

Apart from the JV RAM, when would you start seeing up-front licensing from Nanya?

I think we answered that before, the first licensing piece, comes in conjunction with the closing or technology transfer, and then the royalty stream begins as they ramp production.

And so, when do you expect that closing?

We haven't put any estimated date.

Got it, okay.

And on the DRAM side, this is last question, on the DRAM side you said pricing so far looks flattish.

Any [thought] on various things for the quarter that ramps out at everything with demand and the channel?

Well, a comment on the quarter today pricing being flat.

There are several components of that of course, and the commodity DDR2 pricing for the PC industry is actually up quite nicely.

It has actually been raising, rising every two week negotiation period since February.

As a matter of fact, the commodity DDR2 contract pricing today is up about 35%, relative to where it bottomed out in February, so that is moving up quite nicely.

The other part of our DRAM product portfolio is our specialty DRAM products, and actually that is quite significant for us.

So the logical conclusion is that that pricing has come down somewhat to affect an overall basically flat pricing quarter to date.

And just one last question on R&D and SG&A, you said they should be coming down slightly.

Any quantification that it would be down 5% as you look for Q4, and kind of quarter to quarter?

We couldn't quite hear your question.

You said R&D and SG&A should be coming down slightly as you look at Q4.

Just quarter-over-quarter, any more quantification kind of, will it be down 5% quarter-over-quarter on the R&D and SG&A line?

Any more color on that, was that your question?

Yes.

We are continuing to do cost management efforts across all of our SG&A and R&D areas, and also have got the benefits of some of our prior actions that are continuing to flow through, so in general, we see that it will be as I mentioned flat to slightly down in the fourth quarter, and we will keep managing that in a downward trajectory as a percent of revenue also, as we do our planning process going forward.

And Vijay, probably some other color around that is we think that we have probably run out not all but a lot of what has happened on the SG&A side, and obviously we continue to go through these technology transitions, and make them more efficient, and we think that that is probably where we have some left, to continue to wring out, and I think it is probably worth noting that it is in that at least quarter-over-quarter, probably somewhere in that 5% range that we would expect to see.

And this mid-quarter separation on the imaging sector should also happen, right?

Well I think that is a good point that you make.

Right now, as you might imagine, because it is a slightly different business that the imaging business actually has slightly higher R&D and SG&A as a part of that business, than the memory business does, and so that also happens to be lifting the index if you will, for our current operating expenses, and as we expect to get that separated this quarter, then that obviously would then come out of the calculation as well.

Okay, great.

Thanks a lot, guys, good job.

Thank you.

Our next question is coming from Jim Covello.

Please go ahead.

Thanks, just a couple quick questions.

First on content per box, I heard what you said has been sort of going on up until now with good increase in content per box.

Question is kind of short-term is, are you concerned at all that because of some of the price increases that the OEMs are feeling from other components, or some of the commodity costs getting passed through to them, that the content per box could slow a little bit, and then longer term, any concerns around the issues with 32-bit operating systems, and only recognizing 3 gigs or so of DRAM, and what workarounds we might see for that?

Sure.

On the pricing issue with respect to its impact on content, absolutely there is whatever we see increasing PC component prices, and in this case, an increasing price on the component that we provide, certainly we get concerned about that applying margin pressure to the PC makers and potentially resulting in them slowing down content.

I can only say to date, Jim, that we haven't seen that be the case, and right up through the last couple of weeks, as I mentioned we have seen roughly 35% price increases in the DDR2 area.

We haven't seen a slowdown in content yet, but it is something we need to watch very carefully, because there is no question that rising prices may have some impact, in terms of content per system.

The second question on the memory limitation of a 32-bit operating system, is it a longer term concern, certainly it would be a longer term concern, if we weren't going to see the industry move to 64-bit operating systems.

That obviously lifts the head room or the cap tremendously, above the 4-gigabyte limitation of 32-bit operating systems, but it is probably worth noting that today the average content per box is 2-gigabytes roughly, and we have still got plenty of room to go to get to that 4-gigabyte per system limitation on 32-bit systems.

It seems like there is a little bit of confusion among the industry folks, whether it is 3, or a little over 3-gigs, or 4-gigs is the limitation.

Your clear view is that 4 gigs is sort of you, you can go up to 4 gigs?

Well I think it is a mathematical calculation as opposed to speculation, so 4 gigabytes would be the limitation, and we have got pretty clear sailing to 4 gigabytes I believe.

Maybe if I could ask a final question of Ron.

How are you feeling about the balance sheet, and what potentially might be put there that or what would be the pressure points to cause you to want to go out and raise some additional capital to get you through the current period?

Yes, Jim.

As I mentioned I gave the perspective on '09 CapEx, and our plans and commitments going forward.

We are continuing to monitor a couple of things.

One is the memory market trends and pricing, and secondly, obviously taking a look at business growth opportunities that we might foresee in the future, and evaluating our cash position and cash requirements.

We are going through our planning process now, and we are obviously mapping out our cash requirements and plans, as I mentioned them, in terms of the CapEx structure.

Great.

Thanks so much.

Yes.

Thank you.

Our next question is coming from Hans Mosesmann.

Please go ahead.

Question on CMOS censors.

What is the market environment?

Is it something that you would characterize as seasonal going into this quarter?

Sure.

I think we are starting to head into what should be a strong seasonal period.

Actually, the results that we delivered in fiscal Q3, were in what would typically be regarded as a slack seasonal period in terms of demand, and I think the growth in Q3 can be attributed to our strong position in the high pixel density area, and that would be in the 3 and 5 megapixel density area.

We are probably about a month away from seeing what should be a good seasonal uptick, in terms of mobile phone and PC demand, and so forth, so the real strong tailwind that we expect to get from the market seasonally, is still just a little bit in front of us.

Okay, and then one follow-up on that subject, but on DRAM.

Have you seen any indication of Back-to-School, or are we just a little too early for that?

I believe that we are a little bit too early for it.

My hopes are that pricing trends that we have carried forward since February in the commodity area will get us through, what I would call the slack seasonal period, which extends now about another month, and if that is the case, I feel really good about the strong seasonal demand, kind of carrying it through the rest of the year.

Okay, thank you.

Okay.

Thank you.

Our next question is coming from Krishna Shankar.

Please go ahead.

Yes, the 55 to 65% bit number that you mentioned for DRAM, is that industry DRAM production this year, and is that despite the CapEx cuts of 40 to 50% in memory?

Yes, that is correct.

And what is your best assessment of NAND Flash production this year, bit production this year, and demand, given factors, such as SSD drives, and things like the new iPhone and 3G phones?

Yes, I will stay my comments more to the supply side, and again those same industry groups that track that are looking at 130 to about 145% on bit growth, and I don't know if Mike would like to put any additional color around the demand side?

Demand in the NAND Flash area, the card demand is really a function of market price as well particularly the density of cards, so the demand is going to be sufficient to absorb that 130 to 150% supply growth that Kipp talked about, but the only question is at what price is it going to sell, so I hope that provides some clarity.

Just a note by the way I will answer a question before that was asked with respect to DRAM demand, our best view is in the computing space because we have got some pretty tight connections with the various computer manufacturers, and what we are looking at in terms of demand, DRAM bit demand growth in calendar 2008 is about 65 to 70% at least from a sampling of our customers, and in calendar Q3 relative to calendar Q2 so on a sequential basis about a 25% growth.

Okay, and in NAND Flash, I know that you have pushed out your Singapore fab.

Are you seeing other evidence of NAND Flash capacity cuts?

What do you see out there, in terms of capacity additions or cutbacks to NAND Flash?

Well, we have seen a couple of competitor announcements but those are all public, and I think you can pull those off the public websites, probably better than we can report them to you here.

Okay, thank you.

You bet.

Thank you.

Our next question is [Kan Wilkel] calling for John Lau.

Please go ahead.

calling in for John Lau.

My first clarification on your 2009 CapEx budget that 1.5 to $2 billion range you gave, does that include your joint venture in IMFT and Nanya, your contribution to those ventures?

This is Ron.

The 1.5 to $2 billion is our gross CapEx expenditures we anticipate for our facilities in 2009.

It does not include capital contribution that we have committed to 550 million for the MeiYa joint venture that is for the 2009 calendar year.

Okay.

And how about IMFT in Flash?

Yes, that is included.

Yes.

Okay, that is included.

And my second question was regarding your image censors business.

Could you give us a percentage of your sales coming from higher end, or higher megapixel image censors, like 3 megapixel and above?

Could you please repeat that question?

Yes, how much of your image sensor business is 3 megapixel and above?

Oh, we just reviewed that data before the call.

It is more than 50%.

And to which end markets are you you selling it, handsets versus digital cameras?

The majority of the revenue is going to handsets, that is correct.

Okay, so are you seeing trend wherein digital cameras, your image censors are replacing CCDs, or is that domain pretty much left to CCD?

No.

We have selectively won a couple of camera designs away from the CTD makers at the 8 and 9 megapixel area, so I certainly would not suggest that today, CMOS censors are taking over CCDs in the DSC space, digital still camera space, but we have between 5 and 10 active projects under way in the digital still camera area with our CMOS censors.

Okay, that helps a lot.

Thanks a lot.

Thank you.

Thank you.

Our next question is coming from Bob Gujavarty, please go ahead.

Thanks for taking my question.

One question, do you see any impact, I know it is hard to figure out but judge decisions, but Rambus' IPC action, or asking for an injunction when Hynix, would that impact your business at all, or just kind of curious, if they happen to get that injunction?

Yes, we haven't even considered that.

We don't really have any comment on it, because that is a legal proceeding between somebody else, but that hasn't gone into our thinking.

Okay, great.

And also if you look at some of the data from DRAM exchange on the NAND contract prices, it seems like the 8 gigabit MLC is taking a particular pounding.

Can you talk about maybe why that may be?

Is that the mainstream density now?

Is that where everybody's production volumes are, just kind of curious if you have some insight there?

Our mainstream volume is the 2 gigabyte or 16 gigabit chip.

I have not seen a significant difference in the market prices for 1 gigabyte versus 2 gigabyte chips.

Certainly the sweet spot in the market today in terms of volume, is MLP, either 1 gigabyte or 2 gigabyte chips, and yes as I mentioned in my comments at the beginning of the call, the market today is clearly oversupplied, and there is price pressure certainly across the sweet spot products, and that would include the 8 gigabit chip you referenced.

Okay, great and the final question, can you give us an update on Lehi, and where you are in that ramp, whether the wafer starts are as a percentage of the peak capacity, whatever way you would like to put it?

Sure.

We hit our initial peak capacity about six months ago, and we are in the process of adding roughly 15% additional, and that will play out through the second calendar half of the year.

From a technology perspective, the vast majority is on 50-nanometer today, but we are into the 34-nanometer transition now.

Okay, and would the 15% additional capacity, what is the capacity at Lehi, on a wafer starts basis?

It will be roughly 60,000 a month.

Okay, great.

Thanks.

Thanks.

Thank you.

Our next question is coming from Kevin Cassidy.

Please go ahead.

It looks like on the specialty DRAM, you had pricing coming down tremendously.

Can you talk more about that?

I wouldn't say pricing came down tremendously by any means.

We had virtually all of our bit growth in the quarter came from the commodity DRAM area, so think of our specialty DRAM product portfolio as being, I don't have the specifics at any finger tips, but relatively flat quarter-over-quarter, virtually all of the bit growth came from the commodity DRAM area, and just as a reference point, the average selling price per bit in the commodity DRAM area, think of it as being in the 3 to $4 range, and the average selling price per bit in the specialty DRAM product area think of it being in the 10 to $12 range, so the net effect of the growth in the commodity DRAM area even though we are in a very strengthening pricing environment, was that 5% quarter-over-quarter DRAM ASP decline.

Okay.

I hope I explained that clearly.

No, that it is very good.

Very good.

How about going forward?

Do you see the pricing still flat going into next quarter?

Well, as I mentioned if you weren't on at the beginning of the call, I mentioned that our average DRAM pricing through the first 3.5 weeks of Fiscal Q4 is relatively flat.

It is flat relative to what we actually achieved in fiscal Q3 , so as the mix may change going through the balance of the quarter, who knows what impact that will have, but with no speculation whatsoever, just actual results, the first 3.5 weeks of the quarter DRAM pricing is flat.

I think you did ask where we may be seeing pricing pressure in the specialty DRAM area, and in particular, the low density synchronous DRAM area for a variety of consumer applications, hard disk drives, and so forth, we are seeing I will say more severe price pressure in that area, than we are in any other portion of the specialty DRAM product

Okay, yes.

That was my question, thank you.

Thank you.

Our next question is coming from Atif Malik.

Please go ahead.

Thanks for taking my question.

On NAND, if you look at the last two years, June is a typical month that you start seeing some seasonal strength, and here June has been quite weak.

Is there a scenario where NAND could be in oversupply for the rest of the year?

Is that a scenario that you guys are modeling and factoring in business conditions?

Of course, we are modeling all different kinds of scenarios in our business conditions, but we can't speculate on what the relative balance is going to be through the rest of the year.

I think we are confident that demand is going to be stronger in the second half of the calendar year, than it was in the first half of the year, due to the consumer nature of the products that consume NAND Flash, whether it is going to be strong enough to absorb the supply, that is your job as well as our job, to try to sense that, but we aren't going to make a prediction.

Okay and Fiscal Year '09 CapEx any comments on the loading of that CapEx, or is it just too early?

This is Ron.

We haven't specifically broken it out by quarter, just given a view, and we will work through it during the year.

Okay, and one last question.

I mean, some of your peers are struggling with their yields, we keep on hearing about defective dies on both the NAND and DRAM side.

Is your sense, what is the probability of some of your peers hitting defect challenges, and migration issues in second half '08?

Is the probability higher, as we move to the 5X nanometer of DRAM, and 4X and below nanometer NAND?

Well, yes, it is hard to say exactly what our competitors are experiencing.

I would say generally, that yes, the technology is not getting any easier.

For Micron, however, I have got to tell you we are pretty comfortable with our technology from a defect and yieldability standpoint.

We are particularly pleased with where our 34-nanometer is coming out, and we think that we have done some things in a different way there, that are going to give us potentially significant advantages as we do that ramp.

Okay, thanks.

Thank you.

Our next question is coming from Bill Waite.

Please go ahead.

Hi, guys, thanks.

A quick macro question for you.

Any commentary in general would be appreciated.

You talk about the geographic breakout of revenue, and whether you see the trends from the last couple of quarters, and the last year or two, continuing through the end of this year and then anything on out --?

We are having a tough time hearing your question, but I think I have got it.

Basically, we have split our geographic revenues out a couple of different ways.

75% is outside of the U.S., and 50% is in the outside of the U.S.

in to Europe.

Is that about right, Mike?

Yes.

Okay, can you hear me better now?

A lot better.

Thanks.

Can you talk about going forward whether you see the trend sort of continuing, and how that, in talking to your customers, what they are seeing as well and the degree to which they have communicated that to you?

Where will you recognize revenue, it can be somewhat misleading.

Certainly the trend is that more and more shipments are going into what we would, we call Greater China, which is Taiwan, Hong Kong, and China itself.

That has been the trend in not only at Micron, but it has been the trend in the industry for at least the last 10 years, and I would certainly expect that trend is going to continue.

Can you put any quantify that and put any numbers on it, with regard to you all specifically?

No, I can not.

Many thanks.

Thank you.

Our next question is coming from Kevin Vassily, please go ahead.

Please go ahead.

Thanks for taking my call.

I want to go back to the demand dynamics in specialty versus commodity DRAM.

Would flat bit growth or no bit growth in what is essentially a calendar Q2 and your fiscal Q3, be a normal trend for you guys, and as you look out into your fiscal Q4, do you see a resumption of growth, in what is obviously a more profitable part of your business?

A flat quarter for us, especially in the DRAM area would not be normal, although it wouldn't be, we are not particularly concerned.

It is not alarming, let's put it that way.

We expect our specialty DRAM business to continue to grow.

We do expect the components of that are going to be slightly different.

We believe that our business in the mobile handset area, our DRAM business in the mobile handset area is going to grow rapidly.

We believe our market share in that area is relatively low, so there is plenty of room for us to take additional market share, and exhibit some pretty strong growth in that area, and that business in particular does have a seasonal, the demand in that business does have a seasonal component to it.

Okay.

Were you at all constrained in terms of some of your trailing edge lines, in terms of being able to kind of grow that business?

I heard you mention that you were transitioning some of that product over to 300 millimeters.

Yes, we were.

I am glad you asked the question.

We as I mentioned in the comments I made earlier, most of the specialty DRAM portfolio is manufactured on 200 millimeter wafers, and as you know, we are not growing 200-millimeter capacity.

In Q3, I can't quantify it, but certainly that 200-millimeter capacity envelope, limited our unit and revenue shipments in the specialty DRAM area in Q3.

We have selectively identified a few of what we believe are going to be high growth specialty DRAM products, and reporting them over to an advanced process technology, as well as from 200-millimeter to 300-millimeter diameter wafers, so we can grow the business faster.

And one last question on this.

You don't think that part of what might have happened here in Q3, had anything to do with a slowdown in the end markets for your specialty DRAM, particularly in kind of the handset business, which it sounds like you are targeting as a growth opportunity?

No.

In the handset piece in particular, we achieved growth.

Okay.

I will leave it at that.

Okay, thank you very much.

You bet.

Before we take our last question, I would like to just clarify one small point.

Earlier we had a question come in about when we expected to close the Nanya deal, and as I recall now Mark actually mentioned it in his opening statements.

We actually closed it during the last fiscal quarter, and recognized a small licensing fee already.

So I just wanted to clarify that, and with that we would like to take the last question.

Thank you.

Our last question is coming from Dan Morris.

Please go ahead.

Thanks for taking my question.

Just looking at your cost declines, you did another good quarter there.

You mentioned three factors, wafer output, node transition, and yield improvements.

Is there any way you could quantify how much cost declines are coming from those segments?

We don't normally break that out, so if you would like to try a second last question, I would be happy to entertain that.

Okay.

If we look at 68 nanometer, you mentioned that you are ramping there on DRAM.

Could you talk about how much that might be as a percent by the end of the year?

On 68 nanometer?

Yes.

Well, it is really running in the TECH facility only today, so by the end of the year on a wafer start basis, we would certainly expect that to be north of 50%, probably closer to 75% of total output there.

Thank you.

You bet.

With that we would 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 this call that are not historical facts, are subject to a number of risks and uncertainties and actual results may differ materially.

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

Thank you again.

Thank you.

This does conclude today's Micron Technology third quarter 2008 financial release conference call.

You may now disconnect.