美光科技 (MU) 2003 Q3 法說會逐字稿

Welcome ladies and gentlemen to the Micron Technology Inc. Third Quarter Earnings Release Conference Call. At this time, all participants have been placed on a listen-only mode, and the floor will be opened for your questions and comments following the presentation.

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

Thank you very much. I'd like to welcome everyone to Micron Technology's third quarter fiscal year 2003 conference call. On the call today are Steve Appleton, Chairman, CEO, and President; Bill Stover, Vice President of Finance and Chief Financial Officer; and Mike Sadler, Vice President of Worldwide Sales.

This conference call including audio and slides is also available on Micron's homepage on the internet at www.micron.com. It you have not had an opportunity to review the third quarter fiscal 2003 financial press release it is available on our website again at www.micron.com. Our call will be approximately 60 minutes in length. There will be a taped replay of this call available later this evening at 5:30 p.m. Mountain Daylight Time. You may reach that by dialing 973-341-3080 with a confirmation code of 3959620. This replay will run through Friday, June 20, 2003, at 5:30 p.m. Mountain Daylight Time. The webcast replay is available through Wednesday, June 25, 2003.

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

During the course of this call, we may make projections or other forward-looking statements regarding future events or the future financial performance of the company and the industry. We wish to caution you that such statements are predictions and that actual events or results may differ materially. We refer 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 current factors can be found on the Company's website. Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance, or achievements. We are under no duty to update any of the forward-looking statements after the date of the presentation to conform these statements to actual results.

With that, I'd like to turn the call over to Mr. Bill Stover.

Thanks, Kipp. I am pleased to report a quarter with notable improvements in operations. For those that may not yet have a copy of the release, I'll cover the highlights. Net sales for the quarter totaled $733m compared to $785m in the immediately preceding quarter. Net sales for the nine months year-to-date were up 20% over the preceding year. Selling, general, and administrative expense reduced $5m to $87m. Research and Development expense dropped by $12m to $162m, and the net loss was $215m or 36 cents per share. Note that all these numbers are devoid of any U.S. income tax benefit, given the Company's position of establishing a valuation allowance against U.S. operating loss carry forwards.

Average selling price per megabit was down approximately 15% quarter-over-quarter. That decrease, I realize, is a little bit difficult to analyze without looking at what occurred with individual devices. As a reference for you, the third quarter 256 Meg DDR average price was approximately $3.75, a decline of roughly 30% over the quarter. The 128 Meg Sync also averaged $3.75, that was an increase of approximately 30% quarter-over-quarter. As you appreciate, the ongoing price pressure stems from the continued supply of subsidized Korean parts in the market.

On this next slide, we reconcile the effects of NRV inventory adjustments and restructuring charges. The third and fourth columns show the timing effects of Generally Accepted Accounting Principles required lower up cost-to-market inventory adjustments. The third column notes $15m of new third quarter charge, and the fourth column shows an estimated $168m benefit from prior period write downs that relate to products sold in the third quarter. By working through these adjustments, you should be able to compare your models to actual operating results.

Of the cumulative inventory write downs, we estimate that approximately $130m remains in inventory at May 29, and we expect about half of that remaining amount to work through the current quarter, our fiscal Q4. In our second quarter conference call, we made reference to an estimated quarterly savings of $65m to be realized from our restructuring actions announced in late February. Those savings levels have been achieved and are principally recognized to lower cost-to-goods-sold.

The Megabit production increase for the quarter principally resulted from stronger progress on the Company's migration to 0.13 and 0.11 line width devices and manufacturing yield improvements. Selling, general, and administrative expenses came down slightly in the quarter due to cost containment measures, in part from the restructuring. Future quarters will likely run in the range of $85m - $90m.

Research and development expenses vary significantly with the number of laborers dedicated to new device development and qualification. The third quarter R&D totaled $162m benefiting from earlier qualification of the 0.11 512 DDR products than had been previously predicted.

As of the end of the quarter Micron had cash and investment balances totaling $1.1b. The overall cash and investment balance, including restricted cash declined approximately $150m over the quarter. Cash flow from operations was positive by approximately $50m, capital spending consumed approximately $160m, and the balance of the reduction was principally made up of debt payments. Capital additions for the third quarter totaled $190m and it looks like fiscal year '03 capital additions will approximate $1.1b. Fiscal year '04 capital spending is presently estimated at $1b.

I will turn over commentary to Mike.

Thank you Bill. Our view of demand in fiscal Q3 for products that drive memory consumption, principally PCs, consumer electronics and communications and networking gear was reflective of our typical seasonal patterns. In other words demand was relatively modest. In light of the environment, we are proud of our accomplishments with respect to production outflow and sales growth.

As was mentioned in the press release, sequential Megabit production output was up about 20% and Megabit sales were up appreciably as well. This enables us to access fiscal Q3 with less absolute finished goods inventory than we carried into the quarter, which is an even lower multiple of the weekly sales rate based on the higher sales in the quarter.

As Bill mentioned, we did experience DRAM market price pressure in fiscal Q3 with the majority of this occurring very early in the quarter. In the past several weeks, however, we have seen prices stabilize and most recently, subsequent to the start of the new fiscal quarter prices have been exhibiting an upward trend. Throughout fiscal Q3, we successfully completed platform qualifications of DDR 400 memory modules at all of the major PC OEMs. We are prepared to support volume shipments as the new platforms are getting traction in the market place.

Our yield to the DDR 400 [spects]from the 0.13 micron 236 Megabit DDR chip are more than sufficient to meet the most optimistic DDR 400 market penetration projections. This will only get better as we continue the transition to 0.11 micron versions of both the 256 Meg and 512 Meg DRAM chips throughout the balance of this calendar year. The industry's next transition from 400 megahertz to DDRI to 500 megahertz DDRII is just around the corner in 2004. We are in great shape to support the front end of this transition as we are currently completing system level validations with our initial DDRII Silicon.

In fiscal Q3, we achieved revenue shipments with our 0.11 micron 1gigabit DRAM and with this accomplishment we are now generating revenue concurrently with six generations of DRAM products. Being a full line supplier of DRAMs enables Micron to support to just about any DRAM application requirement that our customers may encounter.

Micron's flash memory business experienced double digit revenue growth in the fiscal quarter. With the progress we are making in further penetrating the mobile handset, networking and digital set-top box markets, these strong growth rates should continue as we move forward over the next few quarters irrespective of the overall market demand environment. In the mobile handset area, in particular our CMOS sensor, Flash memory, cellular RAM, and low-power DRAM family put Micron in a unique strategic position to supply multiple technologies. In some cases these products have been integrated at the packaging level. As mobile communications application complexity and service offerings continue to advance, the ability to bundle these technologies makes a more compelling case for customers in the mobile space to engage with Micron.

A general over supply of DRAM attributed primarily to the Korean government subsidization program continues to supply the industry. The resulting economics present obvious challenges. We are facing up to those challenges by advancing the technology, browsing the entire product offering, and doing so with the focus on cost reduction. We are starting to see some signs of life in the PC demand and encouraging DRAM market price trends, which is at this point can best be termed as a seasonal uptick. This new demand is quite rich in both memory content and technological complexity, and this gives us reason for near-term optimism.

Thanks again for your interest, and with that I will turn back over to Kipp.

Thanks, Mike. What we would like to do now is to take questions from callers. And just a reminder, if you are using a speakerphone, would you mind please picking up the handset so that we are clear when you talk, we can hear you. So with that, we would like take questions.

Ladies and gentlemen, the floor is now open for questions and comments. If you do have a question or a comment, please press number "1" followed by "4" on your touchtone keypads at this time. If your question has already been answered and you would like to remove yourself from the queue, please press the "#" key. Please note that we do ask that you pick up your handset while posing a question to provide optimum sound quality. Please hold while we poll for questions.

Our first question of the evening comes from John Barton with Wachovia Securities.

Yes, good evening. Mike, you had commented that you have seen kind of a seasonal uptick over the last couple of weeks from relative strength in PCs. Based upon your conversations with major customers, any indication on typical Megabit usage per PC in the upcoming quarters and any other flavor you can give us with respect to demand growth in that same time period?

Sure. I think with respect to memory content that we're seeing today and will probably see over the next quarter, we are looking at 350-400 megabytes in a typical consumer desktop application, 250-300 megabytes in commercial desktop application, and probably about 350-400 megabytes in network applications as well. Some of the new platform technology comes to market, we would expect to see some pretty appreciable increases in the consumer desktop area in particular, as the dual-channel memory systems come to the market place and a lot of our customers are looking populating both channels with 256 megabytes each for driving wholesale of 512 megabytes per system. So we are expecting to see a pretty strong movement from the memory content, as the dual-channel systems come to the marketplace.

Any indications on customers on respective number of boxes to -- do you have to support next quarter?

It is really difficult for us to get that granular with respect to the number of the box demand. In general, our larger customers are projecting about 20% increase in Megabit consumption calendar Q3 versus counter Q2, if that gives you some indication.

Thank you.

Our next question of the evening comes from the Joe Osha with Merrill Lynch.

Hi gentlemen, just a few quickies. For starters, I am trying to do the math, you indicated that production was up about 20% in bit terms and that pricing fell about 15%. And wasn't doing something wrong, that should have led to an increase in revenues unless -- but you are saying that actual third quarter sales volume as measured in Megabits outpaced production, so I can't get those pieces to put together; can you help me?

Probably easiest Joe to take you offline and work you through the number --

Or perhaps let me ask the question in a simple way. Did Megabit shipments grow more quickly than Megabit production during the quarter?

No.

Okay.

You have keep in mind that Q2 revenues were strongly encouraged by a reduction in inventory. So the comparison-- when you are trying to compare percentage changes, you get a little bit messed up there

Megabit was higher than Megabit shipments during the quarter?

The Megabit sales roughly equaled Megabit production.

Okay we will take this offline. The second question I had was if you can give us some sense of your planned capital spending, how much you plan to spend on 300 mm versus 200 mm facilities?

Well round and about 16% of our CAPEX budget is going into 300 mm activities right now.

Okay. I am sorry the last question, I will go away, can you talk within DDR -- can you talk about what percentage of your mix was 266 versus 333 versus 400?

Mike would you like to take a shot at that one?

I can Joe, I don't have that data at my finger tips. I would say that the -- however the majority of it today is 333 transition throughout the quarter. Nearly all of its 333 now and we're starting to see some demand for 400 as you can imagine. In some cases, we have customers in the month of June that they are indicating approximately as much as 10%, are demanding 400. So we are seeing a pretty strong transition from 333 to 400.

Okay. Thank you very much.

Thanks Joe.

Our next question of the evening comes from Mona Eraiba with Rosetta Group.

Could you tell me what is the average ASP during the quarter, and looking forward, where should we target the various expenses -- R&D, and selling and administrative?

The average on the 256 Meg DDR, we gave a reference device there of about 375 for the quarter. The 128 Meg sync was about the same -- 375ish, the 256 Meg synch was about 450.

But what is the average for the whole company that's used to give us this number average ASP?

Don't have a blended ASP, Mona. The SG&A total on the guidance or the suggestion that we're giving for Q4 is between $85m - $90m and on the R&D between $175m- $18m5.

What kind of pricing are you getting right now on contracts?

This is Mike speaking, Mona. The range on DDR in the high 3 to high 4 depending on the particular flavor.

High 3 or high 4?

High 3 to high 4 for a 256 Meg device; on the synchronous DRAM, it would be in the $4 range to slightly higher than $5.

Okay, thanks.

Thanks Mona.

Our next question comes from Robert Leats (ph.) with Bear Stearns

Unless I am deaf or something, did you guys give guidance for the current quarter that we're in or was there no guidance Kipp?

I am sorry. We can't hear you. You are cutting out a little bit.

Was there any guidance given for the current quarter that we're in?

Is there any guidance for the current quarter we're in; in what line items are you looking for?

I will take revenue and EPS as that.

We don't guide the revenues because we don't try and predict ASPs. We can talk to you a little bit about production bit growth, which we believe will be up single digits. And then will be up to you in your model to determine whether or not we'll build inventory or maintain it.

So you're saying production will be up single digit over the next -- sequentially not year-over-year, is that what you are saying?

Sequentially quarter-over-quarter. Yes for fiscal Q4, it will be up single digits from what we produced in our fiscal Q3 that we're reporting now.

And you're saying right now prices are holding up briefly firm and you've gotten some higher falls now in some other products; is that correct?

I think that's an appropriate interpretation, yes of Mike's comments.

Okay thanks.

Our next question comes from John Lau with RBC Capital Market.

Thank you, I was wondering if you can give us some more color on the demand post to SARS. When you said that you saw an uptick and that you mentioned it was also seasonal, re you also seeing that the demand in Asia coming back; and is that catch up mode or is it just seasonal right now with not much clarity on what's happening in the Asian Market? Thank you.

Hi John. I think with respect to demand in Asia, it has been pretty strong, strong in the last couple of weeks in -- I want to attribute it somewhat to speculation. You are probably aware that in particular DDR 400 prices have really skyrocketed in the last week, 10 days or two weeks or so. And my interpretation is that a lot of the customers in Asia that are buying DDR 500 were speculating on a tight supply situation in the strong demand environment driven by back-to-school demand. So, the Asian demand is probably folks trying to leg up on what they would anticipate to be pretty strong back-to-school demand.

That leads to my second question is that if you take a look at you chart, it looks like it's over 50% DDR400 as you exit the year. Are you seeing the corresponding support for that, maybe the Intel Springdale chipset. Are those inline for you to achieve that 50%?

Based on the data we have available to us, yes.

Okay, and--

One of the most significant variables on the DDR333 verses DDR400 demand, of course, is going to be price, our relative price premium for the 400 versus the 333. Obviously the higher the premium, the lower the penetration rate of 400.

Okay, and the Springdale actually is able to accommodate the 333 or the 400, so it's really a cost benefit choice for the consumer at this point.

It's a cost performance trade-off, that's correct.

Thank you Mike.

Our next question comes from Tom Thornhill with UBS.

Thank you. Kipp, several things. 0.11 micron program target was 50% of starts by -- in the calendar year, has that been pulled in some?

This is Steve, Tom. The chart that we showed is still the chart. We are right on schedule. The 0.11 continues to grow and we still have the target by the end of the year. Nothing's changed and it looks pretty good.

Great. The bit gross projection for fiscal years had been in the sort of 45-50% range for '03 fiscal year and '04 fiscal year. Can you update those numbers a bit please?

Nothing's changed the notes, still looks like mid 40s for us for fiscal '03 and that's about the same place we will start people for fiscal '04.

How long can you run with inventory given that we were under two weeks at the beginning of the quarter. It looks like you used some. So we're quite low at this point. How long can we run at these levels of low inventory?

Mike, maybe you'd like to address that?

We can run on the inventory level we are at, we just can't run very well. So we are having anything less than two weeks. In every way, we are going to have stock out situations at certain customers and in fact that it what's occurring today . So, I would say that at the low inventory levels that we are at, it's very difficulty for us to maintain a level of service that we would like to. And it also obviously low inventory levels facilitate price increases.

And how do you reconcile your production bit growth for Q for the quarter in low single digits, customer demand that you were seeing at about 20% and the inventory already at low levels.

Yes, you can draw your own conclusion for that I think.

Thank you very much.

Thanks, Tom.

Thank you, Mr. Thornhill. Our next question comes from Clark Fuse (ph.) with Fulcrum Global.

Yes, thanks. Do you have the wafer stocks that you're currently running at 0.11 micron?

We tend to speak in terms of outs. And we haven't given that as a specific really, but like Steve said, we are on target with the chart we showed at the analyst meeting and still on target for at least 50% by the end of the year.

Do you have any problem making a capability of the demand issues for DDR400 by name and account recorded say at the 50% level?

I am not quite sure what you mean by the question, sorry could you--?

Well, if demand is 50% of production, we have colander yearend, do you see any issues in getting there?

We could meet that today.

Great. And can you give us the bit statement split by density and technology for the last quarter?

I can vaguely, we're running about height 60% 0.13 and a little bit of Trailing Edge technology for some of the lower density parts and the rest is all 0.11. And that will go in Q4 to mid to high 70s on 0.13 and everything else will basically be 0.11.

Great. And by the density splits, 256, 512? Any 128?

Yes we have run pretty stable on 256 megs for quite a while, and run 65% - 70% on 256 megs, and the 512 is picking up steep. It's been single percentage points up till now and may be Mike would like to address how he sees customer pulls on 512 maybe go on forward.

It will be interesting to see. I think we are likely going to see strong demand for 256 megabyte module, so those customers that choose to populate 512 megabytes in a typical desktop application are probably going to go through two 256 megabyte modules to take advantage of the performance enhancement from the dual channel system.

Okay, great. That will be enough. Thanks.

Thank you.

Our next question comes from Manish Goyal with Neuberger Berman.

Yes I have a few clarifications. Actually, I want to go back to the very first question. What was the sequential growth in bit shipment?

We are not going to give that out specifically, Manish.

I see. Can you explain if your total inventories declined in bit, why the value of the inventory is so much higher? Is it because of all the increase in raw material or what is it?

No, it's the reduction, Manish, in NRV component. So you appreciate to the extent that the earlier beginning of quarter -- end of prior quarter inventory had a significantly larger write-down, of the $197m if you remember write-down at the end of the second quarter and only $15m at the current quarter.

So that's what is increasing the value of the inventory there?

It increases the cost of the inventory.

Oh sorry, yeah, okay. All right, thank you.

Thank you Mr. Goyal. Our next question comes from Gus Richard with First Albany.

Thanks guys. Can you talk a little bit about the relative cost differential between 0.13 0.11? Are you at sort of cost parity there in terms of [inaudible]?

Well -- obviously in the early stages of the transition, there is not a cost advantage the day you start gaining up some volume. I think in general you would expect somewhere at least probably the 50% range, as we mature the 0.11 over the 0.13, and it depends on where we are at in the stage of conversion.

Right and at this point with 20% of your mix roughly is the wafer outs, I am assuming that the cost on 0.11 is a little bit higher still?

Yes, it all depends on where you got on your yields. I would say it's probably a little bit but that's partially reflective of the volume. So you know, it's not a lot though. Obviously we wouldn't ramp the 0.11 unless we thought it was ready and that we were going to achieve a cross over. So -- but we are pretty close, we are right there. We don't see any obstacles that's not going have it be a very cost forbid for us as we move forward.

Okay. So in the coming quarter, you should start to see cost reductions based on the ramp of the 0.11?

Well, we still have a couple of things going on. You have the finishing out if you will of some of legacy staff moving to 0.13. Obviously that will flow through. We are continuing to mature the yield on the 0.13 even though we have obviously made a tremendous amount of progress. And then of course that 0.11 will continue to be a better cost structure in its own right as we move it forward. So there's really three effects going on here in the cost structures you know to the next quarter and quarter beyond.

Got it. All right, thanks a lot.

Thank you Mr. Richard. Our next question comes from Dan Scovel with Needham & Co.

Thanks, a couple of questions. What was your stock market percentage entertained in the last quarter?

It was in the 30% range, 30% of bit shipped.

Yes, okay.

And let's make sure we qualify that, we -- that includes SpecTek brand and normal stock market activity.

Can you comment on your current wafer out, are they still around the 50,000 I guess a week?

Correct.

Okay. In terms of -- how was the ramp on the 300 mil in Virginia going there?

Actually it's going pretty well. We have obviously indicated we were starting to ramp it. You know we have a [dye] coming off that line now while it was going to continue to ramp to a stage to [proof] up the technology. And probably the more relevant question is where do we think the endpoint is for this phase that we are in right now on wafer [starch] but we are still ramping towards the 500 wafers a week, and when we get to that stage then we will make a determination as to how much will we go beyond that.

And you gave some bit production guidance for this quarter in terms of costs that we look, can you give us an idea what you think your cost is going to do this quarter?

We won't give you a specific target but it will not accelerate like we did here in our fiscal Q3. That was a pretty big quarter for us. The past guidance, we have given is average of about 10% a quarter, 50% for 12-month period. And obviously we are pretty ahead quarterly in Q3, don't expect that same kind of cost reduction as a percent in Q4, but we will get cost reduction.

Okay, finally one more, can you comment on, I guess the difference between the Boise Fabs and some of the international Fabs in terms of where they are in technology?

Well, obviously the Boise Fabs were the leading fabs as we started the conversion, essentially all over, it is on phase now, either running 7.11 or trying to run at 7.11. So we felt pretty good, and if you look at the maturity of the [inaudible] and all the fabs around the world, obviously there is a little bit more maturity here in Boise but rest are more and far behind.

Right. Thank you.

Our next question comes from Sumit Dhanda with Banc of America Securities.

Hi, couple of quick ones, first what do you estimate channel inventories were at the end of the quarter here?

Mike, would you like to take that one?

All right. I have no idea, I mean, obviously all we know is what our inventory levels are and we put all our material out in the market at the end of the quarter, some of it was obviously is the speculators that they may be sitting on but I mean, my feeling would be, it is probably relatively lower but I couldn't really quantify it.

Okay and then the next question. You talked about feeling to ramp up your flash production here. Could you take a stab at what it might be at the end of the calendar year as a percent of revenues?

Yes, our flash productions are running around 2% - 3% and I don't see big change anytime in the near future on that but that's okay for us.

Okay thanks.

We will move on to our next question. Our next question comes from Clark Westmont with Smith Barney.

Hi guys, thanks, all of my questions have been answered, but I am wondering if you can comment second half of June contract prices, have they been settled and did you get any increases or just give us any color if you can?

They have been settled and we did get increases across the Board, 5% - 10% range. Anything else, Clark?

No. Thanks.

Our next question comes from Aaron Marsh (ph.) with ROM (ph.).

With regards to the CMOS sensor market, I was wondering if you can talk a little bit about your market share right now, and how you should that change in throughout the course of the year?

Yes, the market share really depends on which segment to be cleared, we have three primary segment focuses. Obviously we are in the digital still camera market. Then we had what we called mobile which would be things like wireless and the PDA stuff and then we have emerging markets which would include sayings like the automotive and, of course, the given imaging pill that is being becoming more and more popular across the U.S. for the replacement of endoscope procedures, of which of course that is our sensor that's in there. As far as been the other one is FDA approved in combination with the given pill.

So it depends on the segment. Obviously in that one trigger segment, I just mentioned, we probably have 100% of market. But with respect to some of the -- what we consider to be a larger market, its pretty formulary. It is tough for us to tell exactly where we are at because the markets are just developing in terms of -- we started thinking about the cell phones most of them have been used in some type of Sipra [delusion] but they are quickly moving to VGA as we speak and then almost all of the new designs, so at the end of the year are going to be probably at 1.3 megapixel.

As you know, when we acquired Photobit about a year and a half ago, we were behind, although since during the design work under the umbrella of Micron if you will and manufacturing of all of those products, we have briefly caught up. We think in the mobile space, we are equal or ahead watch out there in CMOS images; in terms of the digital still camera market, we have a 1.3 megapixel out and we have our 2 megapixel out, the 3 is going to be shortly hereafter and the 4.

So we think by the time we get -- some time in the next few months or at the end of year, we will be completely caught up and may be ahead in that market. I don't really want to speculate on market share, I can tell you at this an incredible number of designing is going on right now and so we think we are pretty well positioned.

Are you actually shipping 1.3 megapixel right now?

Oh yes.

In what kind of volumes?

We're not going to give that way for competitive reasons.

Okay. Thank you very much.

Our next question comes from Ben Lynch with Deutsche Bank.

Yes. Steve, a couple of questions for you. One is when could we see another accelerated cost reduction quarter like the past one due to 0.11 micron really getting into its sweet spot? And secondly could you just give us a - as qualitative as possible the feel for what your DDR 400 deals are at 0.13 and 0.11 micron please? And then I had one question for Bill.

Okay. Well, the cost reductions, obviously we got the [inaudible] out. We're trying to continue to have great quarters. We're just not going to commit to the type of quarter we just had until we achieve it just like we didn't commit to it last quarter until we achieved it. So we have got a lot of things in the works.

We had said we were targeting some pretty significant cost reductions by the end of the year. We're certainly on schedule to do that. Obviously, we're going to try to delivery beyond that. So I don't want to speculate, but we haven't stopped anything in terms of effort and we feel pretty good about what we are at.

In terms of the DDR 400 percentage total production, we just think we are in a great position. Kipp already said that if they wanted -- the market was at 50% today, it wouldn't be a problem for us. I think as we move, the conversion at 1.1. Frankly I think the focus for probably start being on the DDRII and more along the lines of the 533 and who can do 667, we think we're in great shape for that. We'll probably be one of the first if not the only that can do the 667 at DDRII level. So we think and by the way on DDR 400, we think we're probably one of the few if not the only one today who can delivery in volumes.

You know the DDR 400 just barely serviced its head in terms of some reasonable demand and we've immediately heard, kind of, across the world that there are shortages and when our customers are telling us is that we are [inaudible] delivery right now. So again we feel pretty good about what we are at. A lot of people can talk about being validated and having been able to sample and have some parts, but when it comes to delivery in volume production, we're right there.

Great. And then a quick question I had for Bill was you guys even as recently as a couple of weeks ago were publicly saying that you expected your cash balance to be I think about $900m - $950m definitely at the end of the last quarter. What happened that you did a lot better than that?

Well, we had some strengthening as Mike indicated and -ASP's probably the easiest thing for me to say was that there was nothing unusual. There was no activity in our part to manage that balance. It came in as it came in, a little better than expected.

Okay. Great, thank you.

Our next question of the evening comes from Matt Gabel (ph.) with Calypso Capital.

Yes. Could you get into your fiscal 2004 CAPEX and what percent will be 300 mm. I don't -- I think you said 16% but I don't know if that was for fiscal '03 or '04?

Just for record Matt, I think -- yes, Kipp said it was fiscal '04. We're running about 15% - 20% on 300 mm. The only caveat I would throw out is that the better it looks, the more we are going to spend on it. We haven't really decided the end point in terms of the ramp of the wafers and for fiscal '04 decisions that we make obviously in the next several months will impact the CAPEX. We still have lots of time to impact the CAPEX on fiscal '04. So I would just say, the 1-1.1b CAPEX for '04. and how that breaks out to 300 mm then the current thought process is a long line to what Kipp said, but I am going to say don't hold this to it, because we made a statement we will try to change that in sometimes in the next 3-6 months.

All right. And secondly, what were your comments on your current levels of inventory right now. How would you describe them?

Essentially, the current levels in inventory in terms of when Mike Sadler was talking, he said a couple of weeks and essentially it's down a little bit. I don't want to over exaggerate how much it's down from the last quarter. Obviously, as Mike has already told you that once we get down to this level, it is very difficult for us to operate service to customers as they like.

Right. Okay. Thanks a lot.

Our next question comes from Andrew Root with Goldman Sachs.

This is [inaudible] on behalf of Andrew Root. I just wanted to follow up on the question on the percent of your output that goes into the stock market. What is the reason that in the last couple of quarter it's been a little bit higher than historically?

Well, I think just the demand has been -- the demand from OEM customers has been as I mentioned modest and we have been in the mode of moving all our effort into the market and sometimes that requires us to be a little more aggressive in the stock market. In fact that was the case last quarter. Just to add what Kipp mentioned when I addressed the question earlier. You know, a good portion of that well over half of the material that we put in the stock market was under the SpecTek brand.

Okay. And I would just say that a lot of that is from the .13 line?

It is representative of our total output of the majority in the last quarter would have been from the .13.

Okay. And just a second question if I may following up from the GDR3 discussion. There is a new chipset that has been talked about from Intel which would be a variance of the Springdale, but be single channel GDR400, are you guys going to see any kind of impact one way or the other from this on either demand or the premium that you are seeing now for 400?

I am not familiar with that chipset, so I couldn't really comment on that.

Okay, and then just lastly given the kind of current environment in the market, what are your updated thoughts on accessing the capital markets right now?

We could barely hear you, but I believe the question was our thought process in terms of accessing the capital market, is that correct?

Right, right.

Okay. You saw the cash burn as we reported it for Q3 roughly down a 150m. It is surely difficult to anticipate what the ASP environment is going to be for the second half of the year. You have heard some description here of what we see of the strengthening markets. So, you will have to make your estimates as to ASP. From our perspective, we certainly key current with capital markets, but at this point in time, we are not anticipating anything.

Steve, thank you very much.

Our next question comes from David Wu with Wedbush Morgan Securities.

Hi. Good evening gentlemen. I am going to talk 2 things; first are there any signs, Mike, about this financing called corporate upgrades -- the desktop cycle or the corporate upgrading cycle becomes a desktop swapped out through notebook kind of cycle this time around? The second thing is can anybody comment about as you know the government is coming into putting these I guess, little things for the Hynix suit and in actuality if everything is passed as expected, what impact does it really have because I guess Hynix is making products out of Eugene and I don't know whether anybody checks, whether US government actually checks stuff from third countries that have Hynix in it when enter the US. How does this thing work?

Yes. Let me take the last question first and then Mike can jump in afterwards on the corporate upgrade cycle. Obviously, what you are referencing is the counter[vailing] duty case -- the CVD case which deals with subsidization , and as many people know the commerce department released yesterday a final duty determination of about 44%. We are pleased with that.

We still have, I think it is important to point out we still have the ITC final ruling. There is a hearing on that next week, and then there will be a final determination sometime in July. You know, we feel like the case is going well, although of course there are still a lot of things to be determined.

The way that it works is that as you mentioned it is true that the output from the Oregon fab would not be subject to the CBT and that it would be shipped just as normal if you will to the customer base here. Although, we have to consider the market shares that exist, I don't think that output would service the entire US market. Obviously, there is a European market as well that is a case in progress and I think there preliminary was 33%. Should that be finalized, then obviously Hynix would have to try to service that market as well out of the Oregon facility. And there is just really not enough to go around.

I think the more relevant question you brought up was how do you deal with product that comes in some other form beside the either a component or module. It is very clear, the customs level that if shows up in a component or module that there is a duty applied and it has to go through that process. If it shows up in some other end-product, then it is a lot more difficult obviously to try to deal with.

I don't have a good answer for on that other than to say that, our experience is that our customer base isn't going to do anything unusual in order to try to buy products. So they are not going to ship where they are sourcing, they are not going to built product in some other country, etc, just to try to accommodate one producer who has the particular challenge or difficultly.

So, we, you know, to the extent that that would be the case, I think it's just that the -- I think it's a matter of what the customer is willing to do and they haven't really been willing to change much in the past. So it will be an issue for him to deal with., Again, I think I just want to say and conclude on the subject before I turn it back over to Mike, Micron has never tried to, if you will, only have that as a task that we are trying to improve ourselves, we have always been worried about things that we can control.

We believe that duties are appropriate, there has been subsidization, we have laws in U.S. for a reason, we expect them to be enforced and that's really the issue at hand. Trying to predict Hynix's behavior or how they might try to get around it really isn't what we are trying to focus on, and so we are doing really what we think is the right thing to do and whatever effect that that will have, it will have.

Mike, you want to jump it on the second question.

Yeah, sure. On the corporate upgrade cycle David; our customers are generally operating on an assumption of 5% or 6% unit growth for the calendar year 2003 versus calendar 2002 in terms of numbers of systems. I think, embedded in that is not a real strong assumption of big corporate upgrade taking place in calendar 2003. Within that demand growth, if we break it down by platform we are definitely seeing notebook demands. We know that because the notebook require memory in different form factor, small outlined in. We are seeing notebook demands much, much stronger than in the desktop.

Thank you.

Thank you, next question.

Our next question comes from Quinn Bolton with Oppenheimer & Co.

Hi, I just wanted to ask a clarification, then a couple of following question. Did you gave a actual number for the percentage cost reduction achieved during the May quarter, and if not, can you try and quantify that on a percentage decline basis? I know you talked about 10% per quarter this quarter and the August quarter, it sounds like you are ahead of that?

We did not give a specific reference. It would not be inappropriate for you to assume that the -- something of a cost reduction similar to the debt increase that we have for the quarter.

Okay. And then the second question just has to do with the -- you started to talk a little bit more now about the 512-megabit density. I am wondering with these cost reductions, if you could talk about when do you think you might achieve your better cost per bit at 512 megabit densities than the 256?

Well, we think that the 512 crossover will happen pretty quickly. It doesn't take a lot of volume that to be the case, but I will just remind you that in some cases it doesn't drive the whole market. We are going to deliver the granularity, the configuration, the customer wants. I think that the 512 is obviously going to be adopted earlier probably in the server space before it gets to the desktop space, just because of the price premium, because we are in the price premium currently as Mike Sadler mentioned on the 512.

It will really be driven by price premium as opposed to the big crossover for those applications that care. For those who don't care, obviously, and its not a granularity issue and they just simply want it in whatever form we deliver it in. I think you could expect us to accelerate the ramp on the 512 in order to slight that market because there will be cost effective for bit force.

Could you care to take a shot or David can you take a shot as to when you think it might see that crossover, is that sort of six month out, 12 months out?

No, it's a shorter time than that.

Okay, and then the last question is with the bit shipping growth you achieved in the May quarter. It looks like you had gained some market share based on the WSES data, but just wondering if you agree with that or do you have any numbers you could look beyond market share gains?

Well, we don't really have numbers on it. We get all the same data that you do.

Okay, great. Thank you.

You bet.

Our next question comes from Richard Whittington with American Technology Research.

Steve, if you would, the subject of increased market share in advanced products over the next period of time. Could you address the issue of the expense that those might be able to command premium prices and secondly whether there is enough capital spending taking place industry wide to narrow that premium any times since.

Yeah, well with respect to the market share there and -- obviously we look at the same data that you do, but with respect to where we are at in advanced products there still a market check, with respect to where we are in the advanced markets, in advanced products sides. As I mentioned I think we are in great shape and probably ahead of everybody others. Total market share of different issue, I would just point out.

In terms of pricing, I think Mike probably said it best. Absolutely, if there is a premium to pricing we are going to participate in it because of our worldwide position. You know it's worth pointing out there we are the first to deliver the one gig on the .11 technology. We think that we're the first to deliver on the 4 gig games. We are talking about the modules etc and we ran the DDR 400. So I think we stayed in the premium pricing to the extent that exists for the market penetration that we're able to achieve as a result of that premium being in place.

On CAPEX, it's an interesting question. Last year, I think that the DRAM industry spend somewhere between 5b and 6b on DRAM specifically. And if you go back and if you look at the number for 2003 which nobody is really talking too much about yet although a couple of the equipment seen on the right top view expected to be maybe similar or little bit less. I think it's very, very deceiving. Because we've had a fundamental shift I think in the demand profile. As you know it historically ran at 75% and the whole world is now talking about 50% at least in the last year and this year.

But the revenues have decreased so dramatically in the industry, as you know, Rick, we peaked at about 42b in prior years and we got up into the 30's again in 2000. And here we are running along in a 15 billion-ish range. So as a percentage basis, we're still increasing the output. It is consumed more 50% per year. So on a percentage basis, it looks like it is a high percentage of revenues but when you look at historically where the industry has been, the shift down with CAPEX being spared compared to historical levels revenues.

I think that its pretty moderate by comparison and I think that what are you seeing right now is we have a substantive a relatively lackluster demand that everybody's talked about the PC consumption shifted and the wireless markets is been tough and the networking markets being tough. To keep in mind more and more of these bits have been consumed outside the computing world. Even in all that lackluster demand in the pricing relatively flat, Mike mentioned its kind of treated up and -- the result I think a lower supply that's been coming online and in fact it's the direct results of this last Capex.

It sounds that you are making as if you are making a case for a normalization of pricing back to some sort of historical earnings group?

Well, you are not going to catch me in that discussion. You know we're not going to give the guidance going forward, but you can drive your conclusions. We've been our industry has been very much supply side driven and obviously and that's why we continue to point out the kind of subsidization that Hynix received and that is not rational and we've being driven by some supply side economics and we are hoping that that's going to a become more rational and in fact I think if that does turn out to be the case, the market will be healthier.

And lastly is there SDRAM or Flash capacity that can be brought back into the DRAM.

I think that it's becoming more and more difficult. And if you look at some of the public data on Samsung they are converting DRAM to flash, not the other way around. And if you look at the most the world, they are talking about maybe more flash capacity and converting more to flash, not the other way around. So I don't really think that's a likely trend as we move forward. I think you will more likely -- by the way in even for our case and the flash grows in strength, we'll be converting bits to that. I think that the higher probability is that the DRAM producers and on existence will be the DRAM producers. Just maybe some foundry arrangements that effect that some degree but looking back over to the SDRAM capacity I mean what SDRAM capacity. It is not just very much SDRAM capacity out there anymore. That market has been relatively flat obviously we exited it and those people are staying in the SDRAM market other than Samsung which claims that their converting capacity and other things such as SDRAM there is nobody else in the entire business that are going to make DRAM anyways.

Thank you.

Thanks Rick.

Our final question of the evening comes from Dan Niles with Lehman Brothers.

Thanks Steve maybe tied into the CAPEX question, how do you think of the need for CAPEX that you move to 300mm? In other words, some of the larger guys like Intel talked about not needing to spend as much money and being a lot more capital efficient to move to 300mm. I mean you kind of see that as well in your business when you do make some move? And is that sort of the reason why your CAPEX is more flat next year versus this year, despite the fact all these inventory trends would seem to be pretty good and hopefully there is some improvement in the global IT spending environment or is there something else going on there?

I again, just feel sort of a personal perspective Dan, the -- I think you have to bifurcate it a little bit. It's somewhat a philosophical discussion, but we continue to have even if it is only 50% big growth in DRAM arena. That's pretty good growth and we know that we cant simply get a 50% increase off of process newer transition alone. In fact it doesn't matter when they are on 200mm or 300mm. We know that by going from 200mm-300mm you can get obviously an incremental growth, but you have to make assumption and you have the same wafer equipment in house which we know isn't the case either because their equipments are more complicated, takes more processing, difficulties in advanced notes]and sort of wafer outs generally tend to come down. But if you made the effort to increase the silicon, then you would have quite a bit of growth.

I heard all this when we went from 4 inch to 5 inch, 5 inch to 6 inch, 6 inch to 8 inch, now we are here 8 inch to 12 inch. And I think that yes, you are right there is a transition that will have to occur. The rate of that transition will determine how many bits come on, how efficient they are but in terms of even if you have a 50% growth profile, processed newer transition to only given way for diameter will not solve the total problem. It's true that your silicon growth would be less but it will not solve the problem.

Now, take that and separate that discussion from the rest of the semiconductor world and I think you are dead on. There is a real issue there because when I have interaction other CEOs of other companies, they are very concerned how anyone is actually going to consume the capacity of 300mm. So I think that's why you are seeing quite a few of them join together in these fabs to try to have a combined effort to consume as much as an entire fabric produced, obviously the cost reason, but also I think just fundamentally for being able to use the capacity of the silicon. That's the challenge that they have and that's the challenge that the semiconductor equipment industry is going to have. I mean you have better days than I do. I think the efficiency of that is a lot of concern to the equipment industry and that's probably why you will continue to see the consolidation there as well.

And I guess maybe moving back to Hynix for a second. In terms of the duties and etc. being imposed on them, do you see that fundamentally restricting supply to some degree. Or do you just see kind of shifting in other words trying to take more of the Asian markets so there is more white box related stuff and you stand in front of the guys who don't have these issues take more of the Dell, IBM, HP market share and so its more of a shift of the quality of the OEM that you have as opposed to necessarily any change in supply or do you see it fundamentally hopefully changing the supply situation that Hynix can put onto the market?

Well, it's pretty tough for us to predict what the reaction of Hynix will be and the Korean government. But frankly, I think there will be some impact, but frankly there is not our objective. Our objective is to have trade laws that hear to and that's why we filed saves. We work market principles to government's industry and not government's observation and that's why we filed.

Whatever the outcome is, the outcome will be, whatever the impact, the impact will be and it's obviously not going to be determined by us. It's going to be determined by the behavior of the Korean government of Hynix and we are hoping that this will move the industry towards more market rationalization of its capacity and that we will get kind of subsidization that we have in the past.

That's great. And maybe a final one for you or Kipp. In terms of the bit growth you have given for this quarter, you were talking about the single digits. You moved some of that equipment from Virginia to your other facilities, so that should be giving you some incremental benefits I would assume this quarter and plus some of that 0.11 micron start should probably start to show up now. It's given 60 days cycle time I would assume in this quarter. Is there something else going on that potentially restricts it or are you just trying to be fairly conservative as you go into and waiting for the wafer outs, I guess?

Well, it's probably a little bit of everything. We obviously have the process of moving the equipment and you made a good point it's very true that we are going to get some benefit from moving that equipment. And in fact we will have the CAPEX savings that we otherwise thought we were going to spend. It's tough for us to dial in the exact bid growth, and as a result we would rather give a number that we feel comfortable with.

If we we're able to execute better, obviously the bit growth will be higher; if we don't execute as well as we should have, then the bit growth will be lower. So we are trying not to give so much granularity to it that, it gets picked apart. We are just trying to, you know, give kind of an indication what we are comfortable with, and at the end of the day frankly, it doesn't matter what I say in this room or what Kipp says in this room; it matters what the operations do and that's why we approach it like we do.

All right, perfect. Thanks a lot, Steve.

Thanks, Dan. 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.

Ladies and gentlemen, we thank you for your participation in today's Micron Technology conference call. You may disconnect your lines at this time and have a pleasant evening.