美光科技 (MU) 2003 Q4 法說會逐字稿

Welcome to the Micron Technology fourth quarter and fiscal year-end 2003 conference call. (OPERATOR INSTRUCTIONS)

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'm like to welcome everyone to Micron Technology's fourth quarter and fiscal year end 2003 conference call.

On the call today are Steve Appleton, Chairman, CEO and President; Bill Stover, Vice President, Finance and Chief Financial Officer; and of course, 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 Micron.com. If you have not had an opportunity to review the fourth quarter and fiscal year-end 2003 financial press release, it is available also on our website at Micron.com.

Our call will be approximately 60 minutes in length. There will be a taped audio replay of this call available later this evening beginning at 5.30 PM Mountain daylight time. You may reach that by dialing 973-341-3080 with the confirmation code of 4173983. This replay will run through Monday, September 29, 2003 at 5.30 PM Mountain daylight time. A webcast replay will be available on the Company's website until September 24, 2004.

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

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 certain factors can be found on the Company's website.

Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance or achievements. We are under no duty to update any of the forward-looking statements until after the date of the presentations conform these statements to actual results.

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

Thanks, Kipp. It's a pleasure to report continued improvements in our core operations and acknowledge solid gross margins for our primary products. For those that may not yet have a copy of the release, I will cover some highlights.

Net sales for the quarter totaled $889m compared to $733m in the immediately preceding quarter. The 21 percent increase in net sales was the result of both higher average pricing and increased megabit sales volume. Net loss for the fourth quarter was 123 million, or 20 cents per share.

Average selling price per megabit was up approximately 15 percent quarter-over-quarter. As one reference for you, the fourth quarter 256b Meg DDR average price was a little over $4.50 cents.

For the 2003 fiscal year, net sales totaled $3.1b, an approximate 20 percent increase over the prior year, and a loss per share after the second quarter restructuring charge was $2.11 cents.

Note that all these numbers are devoid of any US income tax benefit given the Company's position of establishing a valuation allowance against US operating loss carry forward. Similarly, when the Company returns to profitability, there will not be any US taxes provided until the operating loss carry forwards are utilized.

On this next slide we reconcile the effects of net realizable value, inventory adjustments and restructuring charges for the fourth quarter. The third and fourth columns in the tale show the timing effects of GAAP required lower-cost to market inventory adjustments. The third column notes that there was only a $4m net realizable value charge taken, and that the related -- and that related to Flash products. The fourth column shows an estimated $76m benefit from prior period write-downs that relate to products sold in the quarter. As of cumulative inventory write-downs, we estimate that approximately $65m remains in inventory at August 28th, and about half of that should come through the current quarter.

Selling, general and administrative expenses came down another $5m in the quarter to $82m as a result of continued cost containment measures. Our first quarter of 2004 is fourteen week quarter, as compared to our typical quarter which has 13 weeks. SG&A for the first quarter is forecasted to be between $90 and $95m.

Research and development expenses varied significantly with the number of wafers dedicated to new device development and qualification. The fourth quarter totaled $166m, with the significant elements being qualification costs of .11 Micron devices and development of our 300 mm line in Manassas. First quarter is expected to run between 190 and $200m, again recognizing this represents fourteen weeks activity.

As of quarter end, Micron had cash and investment balance of just over $1b. The balance a couple hundred million dollars better than many anticipated as we entered this quarter.

Our press release reports today's investment by Intel capital of $400m -- excuse me, $450m -- for rights to 33.9m shares of Micron common stock. This additional capital is not represented in the year-end cash numbers. While further bolstering our balance sheet, the investment is evidence of the strong cooperative relationship existing between the companies.

Capital spending was just over $1b for the year, and we anticipate that fiscal '04 capital spending will be up slightly from this level.

With that I will turn it over to Mike.

Thanks, Bill. We observed a strengthening DRAM demand environment throughout fiscal Q4 and it was driven primarily by growing system demand and memory content in the computing markets. Based on input from our customers, we attribute the strength to both back-to-school consumer PC seasonal demand and the early stages of a re-energized commercial computing environment. The end result for us were sequentially increases in megabit sales exceeding production output, further declines in finished goods inventory levels and a steadily rising DRAM pricing environment. In the three week period since the close of fiscal Q4, demand has continued to remain strong for all DRAM products in the Micron portfolio.

In fiscal Q4 we established ourselves as the industry leader and primary market enabler for DDR400 memory technology. As the rest of the DRAM industry catches up, price premiums for DDR400 have dissipated somewhat, as expected. We do, however, continue to enjoy premier market position and price premiums as they are available in the high-performance applications.

Today's yield distribution on our .13 Micron version of the 256b Meg DDR chip exceeds the current and projected 400 MHz market penetration rates. The industry will begin a transition to DDR2 technology in 2004. And as was the case with DDR400, Micron is in a lead position to drive this technology into the marketplace. We have DDR2 product in commercial production today. Our 256 megabit chip is used to configure modules ranging in density from 128 MB to 512 MB, which makes it an ideal cost density enabler for a wide variety of systems.

Right on the heels of the 256 megabit chip are our 512 megabit and 1 gigabit chips, which enable both further cost reduction and module density expansion to 4 Gigabytes, meaning availability for Micron of essentially every module density and configuration possible. We're working closely with Intel in the development and commercialization plans for this technology, and today's announcement is the culmination of a much closer collaborative working relationship between Intel and Micron on technology for the computing environment.

Outside of computing, we're seeing strong and relatively stable demand for lower-density DRAM products being driven by consumer electronics and network equipment. Micron has made a significant effort to develop cost reduced versions of these products on advanced process technology, and those efforts are paying off nicely with strong customer relationships, increased exposure in non-PC markets and average selling price per bit premiums relative to the PC main memory products.

Execution of this legacy DRAM strategy is also paving the way for introduction of newer Micron products, such as low-power DRAMs, pseudo-static rams, CMOS image sensors and multi-chip package products, or MCP’s, which incorporate a combination of a NOR-Flash chip and a DRAM, or pseudo-static RAM.

The MCP product area in particular is intriguing for Micron, as we are one of only two or three large semiconductor manufacturers with the ability to integrate our own volatile and non-volatile silicon into an MCP. The mobile phone market is clearly headed in this direction with respect to deployment of MCPs as the primary memory subsystem solution.

We are quite encouraged about the market's acceptance and validation of our CMOS imaging technology. In fiscal Q4 we made first volume shipments of Micron produced one megapixel sensors for the digital still camera market. We're following that effort up this quarter with increased production output of the one megapixel chip, initial production volumes of the two megapixel chip and sampling of our VGA sensor products for the mobile phone camera market. Although not a significant revenue generator for Micron today, this technology positions us ideally to be a significant player in what is expected to be a high-growth marketplace in future periods.

We are obviously pleased with the recent trends in the external marketplace and the strength of the current business environment in addition to our progress in reducing costs, introducing advanced technology and broadening our product portfolio. Our customers are quite receptive of this type of execution, and we're all optimistic about the prospects for continued demand strength.

Thanks very much your contended interest, and I will turn it back to Kipp.

Thank you, Mike. We would now like to take questions from callers. (OPERATOR INSTRUCTIONS)

With that, would like to open it up for questions.

(OPERATOR INSTRUCTIONS) John Lau, Banc of America.

I wanted to see if you can give us some more color with regards to some of the product mix that occurred throughout Q4, and if you can help us come to some of the elements that we had there. You mentioned that the bit growth rate was about 8 percent, ASP's were up about 15 percent, but the revenue was -- the revenue growth rate up about 21 percent was a little shy there and the inventory was lower, so I am hoping you can help us reconcile that.

I'll talk a little bit about the product mix. I'm going to be a little presumptuous here on the specifics on the question. But about 75 percent of our megabit sales in fiscal Q4 were of DDR technology, meaning 25 percent synchronous DRAM. In terms of the density, I don't have the figures at my fingertips, but the vast majority of our sales were in the 256 megabit density. I think about roughly 20 percent of our megabit sales were in 128 Mb or lower, which would include 128, 64 megabit and 16 megabit products. And it's pretty safe to assume that all of that lower-density sales were outside of the PC environment.

Otherwise I think we can work with you off-line to specifically work with that model. Because the percentages are off of different basis, you can't literally compare one percentage and aggregate them to the sum. But we will be happy to work with your model off-line.

One final question is that in terms of the ratio, do you see the continued migration over to DDR400? And what do you anticipate the market will be as a percentage for DDR400 as you exit the year?

Just a snapshot today, our perception is it probably represents about 30 percent of megabit demand in PC applications today, is DDR400. We do see it increasing. I think the slope has probably flattened out somewhat. We may hit 40, 50 percent towards the latter part of this year or the earlier part of next year. Our yield percentage to DDR400 is significantly higher than that. It's basically a non-issue for us at this point.

Thank you.

Dan Niles, Lehman Brothers.

Actually maybe one question around the Intel investment, and specifically it talks about deployment of DDR -- DDR2 and 300 mm. Does this change and all? Maybe this is more for Steve. Does this change at all or pull in any thoughts around the 300 mm transition. And I am just going to follow-up.

We had a hard time hearing you, so I'm going to repeat what I thought (multiple speakers) --

I think I heard your question. On the 300 mm, when you say does it change our plans, well, a couple of comments on 300 mm that I will just make in relationship to that question.

One, remember our 300 mm is (indiscernible) and by far the vast majority of 300 mm in the world today is all on the older process known as .13 or maybe something even before that. So our efforts are on really intersecting 300 mm at the advanced process technology node.

Not, as you know, we have been in development of that for a little while now, and that continues. And I think the difference between what we were doing without the relationship with Intel, and what we're doing now, is really how far do take that development and how far do we ramp the facility?

So I think what you could expect is that we will just continue to move the 300 mm technology forward and continue to ramp that facility over the next 18 to 24 months. And obviously the market will determine that how that deployment occurs. But it does assist us in taking it further.

Okay. Maybe for you or for Mike -- there's been a lot of talk, and it seems as though over the last month or two there has been some chatter about prices for DRAM going up, and therefore the usual suspects on the PC side are making the comments about tailing back on the memory per box, or not cutting back but maybe flattening out the increases. It seems as though some of that has come through on some of the market numbers I've seen. What are your thoughts, maybe Steve or Mike, around when the prices actually do become a problem? And are you concerned at all about the memory per box stuff sort of flattening? It seems to be around the high 300s at this point.

I think yes, I would verify or validate your statement that memory content has flattened out somewhat in the past month or so, particularly in the consumer desktop area. If we look at the past year, we're still looking at 40 to 50 percent content growth per platform, whether it be notebook or commercial desktop or consumer PC. My belief is that this is in response to not necessarily the absolute price of memory, but certainly the uncertainty over where the absolute price was headed.

From an absolute pricing standpoint, I believe we have data that suggests that memory content, or the memory price as a percentage of the building material for a PC today, is still in the 5 percent range, something like that, which is not really a problem. And our customers are not telling us it's a problem. If they have concerns at all about it is the uncertainty about where pricing might go. And with the extreme pricing pressure, and what I would consider to be a price war in consumer PC area, there's a tremendous amount of focus on cost right now, and I think -- again, my perception would be that in the low end consumer PC space our customers are kind of taking a breather on memory content until they get some more visibility as to where pricing is headed.

Great. Thank you.

Joe Osha, Merrill Lynch.

Good morning for me; good afternoon for you. You have typically been willing to provide some comments on where you think shipment volume is headed in the upcoming quarter. Would you be willing to do that this time?

Let me just correct you. We haven't given any guidance on bit shipments, but we have been on bit production and that is (multiple speakers) we're starting in the low double-digit range for fiscal Q1 over fiscal Q4.

Do you expect to be able to continue to ship roughly at production rates? That would be trying to get the answer (multiple speakers).

If we can believe our customers we certainly would. I just got fresh projections from our top 10 customers today, and if I aggregate everything it amounts to about 19 or 20 percent bit demand growth calendar Q4 relative to calendar Q3. Even if I were to judge that, I don't think it's going to be any less than 15 percent.

Just to close, some spot price softness or perhaps lack of acceleration here past couple of weeks. I am in Taipei. Some people are attributing that to folks, in particular channel players, just dumping inventory. Can you comment, Mike, on what you've been seeing in the spot market?

I would also verify that we've seen the spot markets relatively quiet in the last couple of weeks, particularly to what it was -- what the environment was like 30 days ago. My perception is that the stock market has traditionally been a feeder into the white box PC manufacturers, and my perception is that the branded PC makers are really in a share taking mode there relative to the white box. All I can tell you is from an OEM customer standpoint, demand remains red-hot from my perspective.

Thanks a lot.

Clark Westmont, Citigroup.

I'm wondering if you can flush out the Intel investment a little bit more in terms of maybe you could let us know what their liquidity horizon is and any dilative impacts or share count impacts we need to think about from the deal? And also, if you can also contrast this to the -- they made an investment many years ago at the beginning of the Rambus generation and just tell us how this is structured differently, or similarly, to that.

Addressing a couple points there, the press release identifies that there are 33.9m shares underlying the rights which they purchased today. That's approximately 5.3 percent. The investment was one of a strategic nature and the relationship, of course, is one of working closer together. As we move forward we think that our roadmaps are pretty well line.

With respect to the liquidity question, we haven't had any conversation with them about that. I can't speak for Intel's capital and what they think about how they are going to manage their portfolio moving forward.

Obviously, the intent of the investment was to advance some strategic objectives between the Company's and that's really all we're positioned to comment on.

How about how this -- I forget what the Rambus, that generational investment from Intel, could you remind me of the dollar amount? And did they get shares -- rights to shares in that agreement as well, or was that structured different?

The prior investment was 500 million. You mentioned Rambus; actually there were a number of strategic objectives that were part of that investment as well that dealt with CapEx and advanced technology processes nodes, etc. So it was not all circled around Rambus, but it was a stock price issue and it was about $500m.

Okay, thank you.

Michael Masdea, Credit Suisse First Boston.

On the Intel side, one last thing. Can you give us any idea, and I know you probably can't say too much, but give us any idea of what the milestones are related to? Are they production, are they required allotment to Intel or anything around that you can give us any color on?

Not anymore that we've already done. Obviously we've said that it's related to -- and some of the primary milestones are around 300 mm technology and around DDR2. And that's obviously appropriate because that's in both of our roadmaps and both of our interests as we move forward.

Great. On the Flash side, has the -- strength DRAM had any impact on your Flash business? Have you allocated any capacity over there? To that same point is any of your production increase in the coming quarter due to shipping up capacity as opposed to shrinks, or what have you?

Yes. We're seeing strong demand, particularly in the wireless area, for our Flash products. I mentioned the MCPs. Again, the MCPs are, for us, incorporating a NOR-Flash chip and a pseudo-static RAM chip, and we are increasing -- I will call it wafer allocations towards both the pseudo-etRAM area and the NOR-Flash area in support of the demand strength we're seeing in wireless.

How is that demand or the production increase next quarter broken down between DRAM and non-DRAM? Or was that number you gave us pure DRAM?

We haven't given it specific. It will increase slightly to non-DRAM but we haven't given out any specifics.

Fair enough. Thanks a lot.

Tom Thornhill, PBS.

Could you discuss a little bit about the trend in inventory, the ability to manage with such low inventory and what the prospect of that is going into the next quarter?

I presume you're talking about finished goods inventory.

Yes, in the hubs, in the management of the hubs, I thought we were running pretty thin at about two weeks and it sounds like you pushed it a little lower.

We did. At the end of the quarter we pushed it down below two weeks, which is really unmanageable from our customers' perspective in particular. And we have, in the last three weeks, because we had to basically, we've replenished a couple days worth of inventory, so we got it back up around two weeks, which is still about -- just about the absolute minimum we can run at and still manage our customers' demand. And what we've got modeled in for the quarter is trying to maintain that two weeks throughout the quarter.

Another question -- in the Intel transaction, is there any cross-sharing of technical support on 300 mm, copper, other types of process issues?

We don't really want to talk about any of the specifics along those lines. There is obviously -- well, it's not obvious, but (indiscernible) technical development going on between the companies (indiscernible).

And percent of sales contract versus spot?

In the quarter that we completed our spot sales accounted for approximately 20 to 25 percent of our megabit sales, and that would include what we call the SpecTeck branded material, as well as the spot business that we run through our consumer sales unit, Crucial Technology.

Would you expect that to change going into the quarter we're currently in?

Again, if I can believe our customers, I would expect it not to change at all.

Thank you very much. Great quarter, guys.

John Barton, Wachovia Securities.

On the topic of contract prices, just to go back through a couple of questions in your statement, you said you saw relative strength continuing through the last three weeks of the beginning of this quarter. Obviously spot pricing has weakened a little bit. What are you seeing on the contract front and do you expect to -- how do you expect to see those two price levels converge -- contract coming down to meet spot or vice versa?

I would -- I'm not going to speculate on where pricing is going to go. I can tell you what has happened recently. I'd say the rate of increase in the contract area has moderated somewhat over the last couple of negotiation periods, which would span back 30 days. But I'm not really going to speculate on what might happen going forward.

Fair enough. If my memory serves me correctly, you were talking to have about 50 percent of your starts at .11 micron by the end of the year. I believe it was actually the calendar year. If that still on track and if not where are we headed there?

That is still on track. If you go back to the graph that we had at the analyst conference in the earlier part of the year, we had pretty much had a trendline that we had shown you as our target. If you were to plot where we are today it would be actually slightly above that trendline, but not much. We're between 35 and 40 percent today of our starts on .11 and we're on target to hit the 50 percent by the end of the year.

Thank you very much.

Andrew Root, Goldman Sachs.

A quick question on the CapEx break up for next year. Can you give the mix that's going to be spent on the 300 mm versus process technology versus other?

We have shown some slides in the past, and out of the approximately $1.1b we're looking at about $16m specifically -- excuse me, 16 percent specifically dedicated to 300 mm and it was around I believe about 55 percent in fab in general. Expect that to have some changes over the year as we shape it up.

That level of 300 mm spending contemplates getting to approximately what type of wafer out number?

We're not going to give any guidance on our wafer ramp in 300 mm. There is a lot of competitors that listen. We feel like we're in good shape with it. We're going to defer questions on that.

That is fair enough. Let's go back to one of the conversations relating to bit production in the quarter. Mike, I think you indicated you thought your customers would be a minimum of 15 percent growth this quarter in terms of demand -- 19 to 20 percent what their forecast is. Should we expect that with what you think you have running through your fabs and what kind of bit growth you can produce, you will be able to maintain share this quarter or do you think you will lose a little share to some other producers?

I think we've given you enough pieces to kind of fill in the puzzle on your own. Kipp, refresh me on what we said on bit growth.

We were looking at low double digits on bit growth, and I guess the question is do you think that the supply in general is growing faster, and can it meet that high-teens expectation from customers.

And then, I'm just curious on the inventory question. If you're finished goods inventory is down a little bit again, and there's still confidence about your ability to do what the customers need, does that mean your work in process raw material inventory is up again this quarter?

Those two things would be unrelated. We've run with, based on other production targets, that's unrelated to what we do in finished goods.

That question is really last quarter, finished goods was down as cited, but total inventory was up. So the total inventory line of the balance sheet would be up or down this quarter?

You need to be a little bit careful looking at the dollar value in the work in process by way of historical net realizable value inventory adjustments. As Kipp has indicated, we're managing whip to optimize cycle time and deliveries to our customers.

Thanks.

Matt Gabel, Calypso Capital.

I was wondering if you could put more color on the sequential decrease in your costs.

Just historical, give you a reference. Essentially we've standard (ph) our targets. Fiscal Q2 of last year we brought costs down approximately 10 percent. Fiscal Q3 was approximately 20 percent, fiscal Q4 was approximately another 10 percent. We're looking at Q1 to be slightly better than that.

Okay. Thank you. Nice quarter.

Eric Rothdeutsch, Friedman Billings.

Can you give an idea of outside of the DDR2 production you expect to ramp of 300 mm, when you would expect to have a material amount of 300 mm wafer starts for just DDR, say 10 percent or more of your wafer starts?

First of all, the ramp of the product on 300 mm will essentially all be the .11 technology and the DDR technology. So all of that capacity goes into the advanced process node on DDR. And it's really just a continuum. I don't know your definition of material, and so we've already said that the initial ramp was up to about 5-600 wafers per week. And we're just going to continue on through that over the next 18 to 24 months.

So you would be ramping at just DDR outside of the Intel relationship as well?

I think your question -- just so we're not mixing things up here -- we had mentioned DDR2 as part of the relationship with Intel and the 300 mm technology, but those are not necessarily tied together in the same facility. We would expect our DDR2 production to predominately be in the other facilities, in the early stages anyway, because obviously it is (indiscernible) and Mike showed a graph earlier as to what he thought the ramp of that would be in the industry. So that's really separate from the 300 mm development.

Just a clarification there -- does the Intel relationship also include DDR memory or is it only DDR2?

The difficulty is that there are a number of milestones that relate to some targets and objectives, and whether DDR falls into one of those categories is -- I suppose it could. It would depending on how you want to look at the milestone. We're not disclosing each one of those are and how they relate. So I think the way to look at it is what we had in the press release, which is predominantly related to 300 mm technology and advancing the DDR2 production.

Thank you very much.

Elliott Glaser, Dupasglow & Company.

You have an outstanding increase in revenues sequentially of 20 percent, but your account receivables are up 30 percent. Can you give us a little insight on that?

I can confirm for you that the customer base was consistent, that there was no increase in days sales outstanding, so it's just reflective of ASP growth at the end of the quarter.

So why were the accounts receivables up more than increase in sales?

You're looking just at the weighting of ASPs late in the quarter, which --

Okay, the decline in spot market ASPs is 7 percent since the end of the quarter. That would take care of the account receivable problem the next quarter?

What we can confirm for you is that there is no accounts receivable problems. There's no change, there's nothing unusual in the accounts receivable at all.

Okay, thanks a lot.

Mona Eraiba, Rosetta Group.

Could you discuss the write-offs that you took during the quarter?

You may be speaking to other expense (multiple speakers). There was a $22m other expense incurred in the fourth quarter where we are recognizing that we have -- hold a number of tools out of our facilities around the world, and as we see a relatively soft equipment resale market, we have just been conservative with regard to our expectations on fair value of those tools. So $22m charge there.

Okay. Thank you.

Jeff O'Reilly, J.P. Morgan.

It seem like you guys are talking more aggressively about ramping a more fuller product (indiscernible) DDR2, including 256 Mb. I am just kind of curious of what your philosophy is on that potentially versus your competitors.

Repeat the very end of that again -- how what compared to our customers?

It seems like most of your competitors are focusing on initial 512 Mb for the DDR2 ramp up and you guys are seeming like you are coming out initially with a 256. I am just kind of curious of your view of that versus probably your competitors' view.

I can tell you our view, and I don't want to speak for our competitors. But the initial product that we put into production is a 256 Mb chip and we believe that most effectively addresses the needs of the market in terms of module density from 128 Mb to 512 MB. It's absolutely no question, it's the ideal density chip to be putting into production to support the market of the outset. We're following up with a 286 Mb, with a 512 Meg and a 1 Gig chip, which, again, will get us to -- enable the higher density modules all up to 4 GB. So without question, what we're doing is the right thing to do in terms of meeting the needs of the marketplace.

Mike Rappaport , Rice Volker.

Good afternoon. It's nice to see things starting to come back together again. I had a question I'm surprised nobody asked, and that's what are the terms of the Intel rights? (multiple speakers)

Could you speak up a little bit?

Can you tell us what the terms are on the Intel rights? (multiple speakers) over $13 per right, so we could gauge how much they're spending to buy the stock and how much they're spending to do the joint development.

Maybe the easiest way to answer the question is that there really is not a significance to the structure as rights, as opposed to thinking of them as the actual underlying shares. So the 33.9 million common shares is the right way to look at in exchange for the 450 million. No significant to the rights, per se.

So there is not a meaningful strike price or a meaningful payment to exercise the rights?

No.

Could you also tell us what the depreciation should be for Q1?

The actual number for Q4 came in at $299m and the quarters through fiscal year '04 will probably run between $310 and $315.

Thank you very much.

Adam Parker, Sanford Bernstein.

A couple of questions. If you look at the previous cycles in the DRAM industry, it appears you had some -- clearly some sustained periods of shortages and now it appears that through manufacturing enhancements, or whatever, that it's possible foe supply to be altered somewhat more rapidly, and therefore in some of the recent years you have had very short periods of shortages. I'm just wondering if you think that is some sort of secular trend, that maybe sustained shortages are less likely going forward? Or is this just a function of this particular down cycle?

I think really you have to look at it in context of a couple of things occurring in the cycle that maybe didn't occur to us before.

You really have to back up about probably 2.5 years, where we did have a shift in the average bit demand per year. Obviously, at that point in time, obviously the majority of the product was going into the computing environment. So as we saw the PC demand profile, the growth rates slow, you also saw that corresponding impact on us in the DRAM business. And so we really went from an average of about 75 percent bit demand growth per year to what you've seen in recent years, which is more along the lines of 50 percent per year. And so there was a fundamental shift.

Now in combination with that, of course, we've had a relatively weak economy now for quite some time, which, obviously, depressed the demand side. And that is something that has been new for us really in the DRAM industry and I think that really all of the players are continuing to try to adjust to that. There's no question that we've had further consolidation since that time period. And we still have effects from capacity that's built-up in prior periods.

I really don't think that the ability to change, to fundamentally change supply, the timing of that, has really changed at all. I think what we've seen is that there were facilities that were capable, that had already been built, that were able to come online by simply installing equipment instead of having to build new facilities. In the same breath, though, we've had lots of facilities come off-line as well. So that's probably been -- those two things in combination are really what's been affected. I don't think we had a shift in the ability to change capacity that quickly.

Okay. Maybe just a separate question, but sort of related to a media or long-term trend. Can talk about -- it has been five or six quarters since this non-binding MAU with Hynix fell through. Can you talk about what you really think is going to happen with Hynix here in the next few quarters, given that I think two or three years ago everyone thought that this consolidation would play out with Hynix and it clearly hasn't? Some think they even gained share in the second quarter because of their market-share in China. What do you expect to happen to Hynix in the next few quarters?

That is a pretty tough question. I don't want to speak for Hynix and obviously at this point in time we're on the outside looking in. Clearly, our objective was to -- our objective in the entire trade case was to stop this irrational behavior of subsidies that were going on. That's what it was focused at. Obviously we're in negotiations to try to acquire Hynix and we thought that would be a good transaction for the company. That didn't materialize. And we, at the same time, didn't -- we wanted to make sure there were not artificial market manipulations.

You can speak to what happened, and that is Hynix has had a decrease in market-share from where they used to be. There was three of us that were all pretty close in size. I guess Infenion and Hynix are now arguing about who is bigger. And the market-share that you've seen gains that Hynix claims to have made still is at a much lower level than what they had historically achieved.

So, yes, I think that many companies face the ability to invest in technology and capital moving forward. I have not seen any fundamental shift as to why that wouldn't still be the challenge for a couple of these companies. And really, whether it's an erosion over time or whether they are able to access the markets for additional capital will determine where they end up in market-shares. But I do think we have seen a fair amount of change among the players in market-share. And that's -- all I can say is that's where it's at today without really knowing where it is going.

Ben Lynch , Deutsche Bank.

Two questions. First one is, when you look at your 200 mm capacity on a sort of facility basis, how much of this do you think is upgradeable to 110 nm, and then how much to 90 nm beyond that?

First of all, it's all upgradeable to 110 or 90 nm. In fact, if you look at the percentage of starts that we have going to the 110, our anticipation is that it will be deployed around the world. All the facilities are doing work in some form or shape on under 10 nm. So that is definitely going to make the transition.

I think the question on 90 nm is which facilities do believe are more optimum to go to 90 nm as opposed to split the investment into 300 mm. And right now, it looks like it pretty much will be (indiscernible) can make the transition to 90 nm. We obviously have to make the decision to do that and flow the capital to do that, but I don't think the limitation is there.

The 200 mm today is still very cost-effective for us. And as we go through time, we will just continue and evaluate that.

So the background to the question is, some of your facilities are quite old. With issues just like the cleanliness of the clean rooms, which plays a role also, the availability of 200 mm litho tools, and tools in general from vendors, given that a lot of your peers are going to be producing 90 nm on 300 mm wafers. So that's where I'm coming from.

Well, I would defuse at least one statement you made which isn't true, and that's that all of our facilities are old. The facilities have continued to go through modifications and upgrades over time. In fact, the very first facility we ever had has been upgraded, is connected to one of the most modern facilities that we've ever built, and running the most advanced processes. So that's not true. The facilities that we have are all very advanced.

And the reason that that's the case today, which may not be true for others or some of our competitors, is because if you look at Micron's history, we've made it a point to bring all of our facilities to the advanced technology process mode, where that's not true with some other companies. We've moved all of them forward to be relatively equivalent in their capability and the technology. And as a result -- you really have to think of the facilities as new.

In fact, I gave a tour today for some university students on the site that we have here in Boise, and we walked by the original fab that I started in 1983. And as I looked at it, it's not the same fab. The point is, it's not the same fab. It's completely been changed, it's been -- everything that was in there at that time, of course, is completely gone; and it's got new structures, and new clean rooms, etc. So I think that's pretty reflective of what we do around the world today.

And just a quick question follow up question please is have you guys given any indication of what you think your supply growth will be full year this fiscal year or calendar year, please -- '04 calendar year?

You have to have a range of 40 to 50 percent.

Nimal Vallipuram, Dresdner Kleinwort Wasserstein.

First of all, congratulations on your cost reduction. You've done an excellent job there. I think question is for Mike. I don't want to ask this question again; I guess it has been asked before. I just want to get a better understanding as what is going on in the marketplace (indiscernible). Do you see a permanent structural change between spot and the contract market in the sense that the OEM customers are gaining more market-share and the spot market becoming less of some consequence as such that in your price negotiations with your OEM customers that you probably would not be using the spot price as the leading indicator? Can you discuss that situation, the changes that have been happening there?

My observation, Nimal, about the branded PCs taking share are really -- I wasn't attempting to highlight what I -- a macro trend. It's something I believe has been occurring over the last couple of months, or maybe couple of quarters. We don't use the spot market price as a leading indicator; we use it as an indicator, along with a number of other variables, in determining or in arriving at a price for our OEM customers.

With respect to whether it's be going to become less relevant for us going forward, I really don't know.

So I suppose my follow up question is that I guess someone tried to ask that question the same way, is that as we see it, for the last couple of weeks, or three weeks, the spot prices have been well below the contract prices. We have seen the situation happen again and again in the past so many years. However, invariably the contract prices have always followed the spot prices. I guess my question is that if something happened in the marketplace in such a way that that probably would not be the case going forward here?

I can tell you -- again, I think you're asking me to speculate on what may be having going forward, and I'm not going to do that. In the current situation, if you are -- the spot market price, at least the reported or the published spot market price, is lower than the price that we're getting from our OEM customers today. And if you are asking are OEM customers using that as leverage to try and obtain a lower price from us, the specific answer to that question is no, they're not. And if they were to, at this point in time anyway, I have many OEM customers that want more product than I can give them. So that would be what I would do. Now, is that going to change going forward? I really don't know.

Thanks, Mike. Thanks a lot.

Craig Berger, Smith Barney.

I was just wondering if you could discuss your lower cost or market reserves and what you expect to sell next quarter, or I guess fiscal Q1.

You saw, or possibly if you have access to the press release, you will note that we had a $4m new net realizable value charge in the fourth quarter related to Flash product. We have about $65m that remains in inventory at the end of the fiscal year. And consistent with past periods, we would expect about half of that to come through the current Q1.

So half of the $65m?

Yes.

Perfect. Thank you.

Thank you very much. With that, we would like to thank everyone for participating in the call today. If you would please bear with me I need to repeat the Safe Harbor protection language.

During the course of this call we may have made forward-looking statements regarding the Company and the industry. These particular forward-looking statements, and all other statements that may have been made on this call that are not historical fact, are subject to a number of risks and uncertainties and actual results may differ materially. For information on the important factors that may cause actual results to differ materially please refer to our filings with the SEC, including the Company's most recent 10-Q and 10-K.

Thank you for joining us.

Ladies and gentlemen, we thank you for your participation in today's Micron Technology fourth quarter and fiscal year end 2003 conference call. You may disconnect your lines at this time and have a pleasant evening.