美光科技 (MU) 2002 Q4 法說會逐字稿

Good afternoon, ladies and gentlemen, and welcome to the Micron Technology fourth quarter earnings release. At this time, all parties have been placed on a listen-only mode, and the floor will be open for your questions and comments following the presentation.

At this time I would like to turn the floor over to one of your hosts, Mr. Kipp Bedard. The floor is yours, sir.

Thank you very much. I would like to welcome everyone to Micron Technology's fiscal fourth quarter and fiscal year-end 2002 conference call. On the call today are Steve Appleton, Chairman, CEO, and President; Bill Stover, VP Finance and CFO; 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. If you have not had an opportunity to review the earnings press release, it is available on our Web site, again, at www.micron.com.

Our call will be approximately 60 minutes in length. There will be a taped replay of this call available this evening around 5:30 Mountain Daylight Time. You can reach that by dialing 973-341-3080 with a confirmation code of 3497940. This replay will run through Thursday, September 26, 2002, at 5:30 p.m. Mountain Daylight Time. The Webcast replay is available through October 1st, 2002. We encourage to you monitor our Web site at www.micron.com throughout the quarter for the most current information on the company, including information on various financial conferences that we will be attending.

During the course of this call we may make projections or other forward-looking statements regarding future events or the future financial performance of the company 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 Web site. Although we believe that the expectations reflected in the forward-looking statements are reasonable, we can not guarantee future results, levels of activity, performance, or achievement. We are under no duty to update any of the forward-looking statements after the date of the presentation to conform these statements to actual results.

I'd like to now turn the call over to Mr. Bill Stover.

Thanks, Kipp. We have a lot to cover, so I'll jump into our report for the quarter and year ended August 29th. On this first slide you can see comparative information for the fourth quarter of fiscal year 2002 versus the fourth quarter of 2001. Net sales of $748 million in the fourth quarter of '02 was 56 percent higher than a year ago. The operating loss of 468 million was approximately half of the quarterly loss of a year ago. Gross margin improvement was a result of both higher average selling prices and cost reductions.

The press release points out that the fourth quarter had a net realizable value inventory adjustment of $174 million. This inventory write-down is the result of the analysis required under Generally Accepted Accounting Principles. Therein we evaluated the quantities of products and finished goods and work in process, the estimated period of sale, and the estimated sales price in the anticipated period of sale.

We are aware that the company altered its product mix in the last quarter to match customer demand for DDR devices, (inaudible) remaining synchronous DRAM products and work in process and finished goods accounted for approximately three-quarters of the year-end NPV adjustment.

Absent the 174 million write-down and the affects of prior inventory write-downs, cost of goods sold for the fourth quarter would have been lower by an estimated $131 million.

Of the cumulative inventory write-downs, we estimate that approximately $240 million remains in inventory at August 29, and as we had previously characterized, about half of that amount will work through the first quarter of '03.

For the fiscal year ended August 29, net sales totaled 2.6 billion, down from 3.9 billion the prior year. Sales in megabits were approximately 45 percent higher year-over-year, but pricing was down more than 50 percent.

You note on this slide that the income tax benefit for the fiscal year '02 is provided at an indicated rate of less than 10 percent. This is due to establishment of a $348 million valuation allowance against deferred tax assets for U.S. net operating loss carry forwards, as required by Generally Accepted Accounting Principles. The company fully believes it will utilize its operating loss carry-forwards.

I know that many of you would like to calculate an earnings per share absent the effect of the valuation allowance and net realizable value adjustments. I can help with that calculation. The reported fourth quarter loss of 97 cents per diluted share would first be reduced by the valuation allowance effect of 58 cents per share. You would then remove the after-tax NPV effect of 12 cents per share, that amount being calculated from the $131 million net NPV effect on the quarter and our recent tax rate of 44 percent. The result for the fourth quarter - 27-cent loss per diluted share.

Another item that may generate some questions is the relative level of production and sales as measured in megabits. In the fourth quarter, megabits sold increased by slightly more than 40 percent, that percentage increase being measured from the seasonally slow third fiscal quarter. Production growth for the fourth quarter was up 22 percent. The absolute levels of megabit production and sales were approximately equal in the quarter, and accordingly our finished goods inventory level in megabits stayed pretty flat. Megabit inventory levels compared to a year ago, however, were down 19 percent.

Selling, general, and administrative expenses rose to 95 million in the fourth quarter due to costs associated with intellectual property matters, recent class action filings, change in corporate governance regulations, and cooperative development arrangements with universities. We expect SG&A to run between 80 million and 85 million in future quarters.

Micron Technology remains well-positioned with cash and investment balances in excess of $1.1 billion. That includes $160 million classified as long-term investments due to their maturity dates. Cash flows from operations have been positive this year, with the most recent quarter generating a little less than $100 million.

I'll turn the presentation over to Mike.

Thank you, Bill. As expected, we benefited from a mild back to school seasonal uptick in PC demand and continued memory content acceleration in PCs.

On the sequential basis, our megabit shipments were up approximately 40 percent relative to FQ3 (ph). As our .13 micron SDRAM process reached mature yields, our megabit production output was up about 20 percent sequentially. We experienced a significant finished goods inventory reduction in terms of weeks of supply, and the mix profile of our inventory changed substantially. We entered the quarter with high levels of DDR inventory relative to then-current demand and exited the quarter with very little DDR inventory. This can be attributed to the rapid market transition from SDRAM to DDR in the PC space.

We were and continue to be well positioned to take advantage of this transition with our 128 megabit and 256 megabit DDR devices. We are continuing the ramp of our .13 version of the 256 megabit DDR chip. The chip size of the .13 version is much smaller to enable lower cost and it operates at higher frequencies for deployment in more advanced applications.

The memory industry continues to be under severe price pressure, as evidenced by the sequential 30 percent decrease in average selling price per megabit that we experienced in fiscal Q4. The bulk of this price degradation occurred on the main memory synchronous DRAM devices as DDR prices were increasing exiting the quarter. This trend has continued in the initial part of fiscal Q1 as DDR demand currently exceeds the industry's supply capability.

Although not the case today, our expectation over the intermediate term that is synchronous DRAM and DDR pricing will be roughly in balance on a per megabit basis, as many SDRAM applications will have lifetimes spanning well beyond 2003.

Earlier in fiscal 2002 we discussed our plans for entering the imaging market with our family of CMOS image sensors. In the just-completed quarter we secured multiple design wins in the mobile phone camera application area, and also received initial production orders from customers. In subsequent quarters, we will continue with designing activity in expected growth areas such as cell phones, PDAs, digital still cameras, PT (ph) peripherals, and automotive applications.

We are continuing to add to our portfolio of memory-intensive products in support of the networking and communications industries. I am pleased to report that we are gaining additional traction in a variety of next generation data network and equipment with our high performance RLDRAM and SRAM product offerings. We are , industry leaders in the networking hardware sector as the primary TCAM source for several of the key high end routing applications. These programs will serve as a strategic launching pad for Micron's TCAM business.

Needless to say the market environment continues to be challenging. We are obsessed with driving costs down in our mainstay DRAM business while making strategic investments to further enhance our competitive position and embrace emerging markets with all of our products. With that I'll turn it back to Kipp.

Thanks Mike, what we'd like to do now is to take questions from callers. Just a reminder, please pick up the handset when asking a question so that we can hear you clearly. With that we would like to take questions.

+++q-and-a

Ladies and gentlemen, the floor is now open for questions. If you do have a question or a comment, please dial the numbers, 1 followed by 4 on your touch-tone telephones at this time. If at any point your question has been answered you may remove yourself from the queue by pressing the pound question. Much once again that's the numbers 1 followed by 4 to bring yourself in the queue. Our first question is Scott Randall of SoundView Technologies.

Thanks very much. I indicate what the like [inaudible] bit production for Micron is next quarter and also for fiscal `03.

For 03 we're looking at a range of 40 to 50 percent depending on markets conditions for Q1 should be low double digits.

What about the industry for fiscal 03 Kipp?

Still looking for '03 in the industry somewhere between 40 and 50 percent. So pretty close to what Micron will be growing next year as well.

Thank you.

You bet, Scott.

Our next question is coming from Dan Nelson of Ray McKenzie.

No question, thank you.

Our next question is from Tim Mahon of CFSB.

Kipp last quarter you mentioned capex for '03 of a billion and a half depending on market conditions, can you give us an update on that?

Estimating '03 we're running between 800 million and 1.2 billion.

Okay. And as a follow-up, can you give us an idea how the DRAM schedule is going at Manassas.

We're clearly, we're still running some the product that we had committed for Toshiba. We have been installing some equipment there in order to transition to Micron's process. Realistically that's going to take you a few months so we're just in the early stages of running what we call prototype lots for the transition.

Thank you.

Thank you. Your next question is Nimal Vallipurum of Dresdner Kleinwort Bensen.

Hi. A couple of questions here. First of all, as far as the DDR market is concerned, it seems like the DDR finished good inventory with Micron as well as the industry, seems to be pretty low than expected. Is that possibly because some of the Windows are having a having hard time transitioning to .13 in making DDR.

You're cutting out a little bit but restate the question to make sure we understand. I think what with your asking us, the premise based on people are struggling through the DDR transition is that in part what's causing some the price premium.

Yes. Kipp, thanks a lot.

This is Mike. I can address that partially I believe. I presume that to be the case. That's the feedback that we're getting from some of our customers. Very safe to say that our market share in the DDR area is greater than our overall DRAM megabit market share so logical conclusion some of our competitors are having difficulty on getting DDR Megabits into the marketplace. I could not presume whether it's a process issues, circuit design issues or what have you. Safe so say we're pulling more than or market share in the DDR market space.

If I take that question to the next level, would that make be Micron be the leader in the DDR, can you position, can you tell me where you are like Infinion.

We can tell you a little bit about the percentages and I think instead of will address it as well. In Q4 we shipped about 40 percent of our total bits in DDR and pursuant to customer requests, we'll be around 50 percent in Q1 and I believe it's around 60 or 65 percent bit ends the calendar year.

That's correct.

As Kipp pointed out, we have obviously the majority of our output moving to DDR very rapidly. If you look at our wafer starts, its dominated by DDR. The issue about who last got the large are market share, I noticed some reports about a couple of our competitors about their claim to be the largest share in DDR. It's difficult to respond to that because unfortunately we're the only ones that have to tell the truth. And as a result, we can't sometimes decipher fact from fiction on some of these other comments.

All right. Thanks a lot.

Thank you. Our next question is coming from Jonathan Joseph of Solomon Smith Barney.

Yeah. Mike, per week basis where inventory started at the beginning the quarter and where they ended for total and for.

Jon we're having a real tough time hearing.

Can you hear me better now.

Perfect.

Sorry about that. Can you give us a essential, mike, where inventory started at the ends the quarter and had he they send the I noted DDR was higher and ended up at zero, but in aggregate?

In terms of weeks of supply, Jon, our best estimate for the industry would be entering the quarter in 5 to 7 and a half 8 week range and exiting the quarter in 4 to 6 week range. In either case we were at the lower ends of those ranges. In terms of total megabit inventory.

Of course DDR is still at zero. You saw positive pricing trends on a contract basis? Certainly spot market has been ticking down a little bit as you exited the market. And you did suggest at some point that DDR and SDRAM would be at parity. Can you give us some sense of what you think contract pricing will do over the next quarter and then do you think sufficient DDR supply will come into the market that gap, that premium will close?

I'm going to be real hesitant to make a prediction on what DDR pricing is going to do. That's going to be dependent on our competitors ability to get more products into the marketplace. I can tell you the last price move on DDR was up. We're putting very little DDR product into the spot market because we're scrambling to get DDR material into the hands of our OEM customers. I couldn't give much of a barometer on the spot markets. Based on what our customers are telling us they're projecting that the DDR situation is going to be. Tight probably through the Christmas consumer selling season. So you can draw your own conclusions from there.

A quick follow-on. Are you seeing any build for the Christmas season, when would you expect that. Obviously, Taiwanese MB shipments are doing a little bit better than anticipated. Probably billing inventory for Christmas are your customers telling you that it will be in line, below or above expectation Christmas.

I think the bottom line is our customers don't really know. And depending on which customer, you got different takes. All I can tell you is if we aggregate everything we know it would be typical to slightly muted.

Thanks, Mike.

Our next question is coming from Joe Osha of Merrill Lynch.

Hi, guys. Yeah. Can you talk a little bit about what your production mix looks like by process at this point? And where you expect it to be by the end of the upcoming quarter.

Sure. We're at approximately about 10 percent, .13, today, that will be substantially completed late spring and we're currently producing some wafers on .11.

I'm sorry, you'll be at 100 percent .13 by late spring.

If I can just jump in real quick, this is Steve. You have to talk about output as opposed to wafer starts and to clarify, about a quarter of our wafer starts are the .13 today but that's rapidly going to 50 percent, next quarter and you can imagine we'll probably get it up to 70, 75 percent the quarter after that. Realizing we'll be limited either by the material that's going into .11 or some legacy product that we have to build for customers, because those are the parts qualified and they have a lot longer life. So I think that's a pretty good guide right there.

Just a follow-up. In terms of is that ramp as usual, being led by Boise and the other part following or is it more simultaneous?

In general, I think that's true. There is - there might be a facility that's a little bit ahead of the other ones out there and fairly close to what we're doing here in Boise but in general, that's right. We typically get it led and the others follow on afterwards.

One last question. You did talk about what your sort of organic reduction in cost per megabit was in the press release. Would you care to give us any insight all those various inventory effects for the coming quarter.

I think one way to think about it is, we think that going forward we have better leverage than what we've been able to do historically. And as we just mentioned we're in a pretty heavy conversion to the .13 and I suppose, while on I'm on that subject, I think it's worth saying, we did a lot of - we made a lot of effort to get ready on the .13, and I think we were - I think our 256 sink came along pretty well, maybe a little bit later we thought. But the conversion to the 256 DDR exacerbated the problem. We weren't where we wanted to be on our designs going into the process. So we had prepared, spent a lot of capital getting ready for the .13. The designs to utilize that capacity historically weren't where they wanted them to be. I think that's why we're making the comment much better leveraged going forward because it turns out now that the our designs for the .13 and the .11 are really doing well. In fact, either equal or ahead of the schedule we set and we will now be able to leverage that capacity more so going forward.

Okay. Thanks very much.

Our next question is coming from Dan Niles, Lehman Brothers.

Great. Thanks a lot. Steve maybe just following up on your comments on .13. From where I can roughly calculate is the .13 is about 25 percent less cost than the .15 you're on now. Obviously you can't give us the absolute cost. And then tied into your answer on the leverage part, it seems as you're moving to .13 and potentially 1.1 a lot fast are than some of the other transitions you had in the past and not just across the US facilities, it's across everything, if I understood your comment. So when you look at it from that perspective.

Can you speak up Dan and tell me what's your question.

My question is in terms of the cost of the .13 relative to the .15, is that roughly 25 percent less. Second thing it seems you're transitions much faster than you normally do in the sense that your international facility as well as your stuff in the US is also moving from where it is today to 100 percent .13 in very short order. Before from what I remember during prior downturns, of the US has led and the international stuff has followed later.

Yeah. We're not going to give you an exact percentage for competitive reasons. It is 50 percent more dye per wafer. That doesn't quite transfer because it's more advanced and the yield is not as good and mature, etcetera until some point later in time. So clearly there is going to be a cost advantage there. All you got to do is run the incomes. 50 percent more dye per wafer.

The rate of conversion, the way I would explain it, relates to what we have done historically getting ready for the .13 design then being slightly delayed from what we wanted. That would naturally - obviously we didn't stop work on developing either advancing the .13 process from where it was or on working on the .11 process getting it ready. Given we were slightly behind where we wanted to be on the .13 we have great leverage on that moving forward, but at the same time the .11 will be ready in comparison to when the .13 was ready in comparison to the .15 if you follow me. So I would naturally expect a somewhat accelerated conversion to the .11 as a result of that.

Maybe following up to Mike's comments earlier in terms of at some point in the future there should be parity between the DDR and SDRAM pricing. Tied in to that question, I guess there are two ways that could happen. The DDRs can move down in pricing or the SDRAM can move up. From what I remember from prior technology conversions, from fab to EDO to SDRAM, the older stuff as people stop producing it that price tends to move higher. What's your thoughts in terms of how this works? As you move more volume to DDR, does the SDRAM drift higher or fall off more...

I don't want to be predicting prices but the current market price for SDRAM we believe to be unsustainably low. Our expectation is that the market price will go up once the inventory is back in balance. Our view some of our competitors are producing more SDRAM or have excess in inventory than would be optimal and as a result of the market price is somewhat unnaturally low. Our expectation would be the SDRAM price is going to go up. Not suggesting the DDR price won't go up. But our expecation is that the SDRAM price will go up in the intermediate term.

Is ODE above SDRAM in terms of what you produce there?

On a per bit basis I believe ODE pricing is higher than SDRAM, that's correct.

Great. Thank you.

Our next question is Jack Geraghty of Gerard Klauer Mattison.

Just a quick question on Lehigh where that is in the great scheme of things.

We've continued to move the facility forward at a relatively slow pace and the worse the market conditions are the slower we move it forward. We need that facility. It's essentially full. We have about 6 to 700 people there. With respect to the wafer fab, though we have not completed the clean room. If you will or finalized it to the later stages or installed any equipment there of course. Our plans of [inaudible] when we are going to need it but obviously we are going to be prudent and that's not spend a whole lot of money on it until the market warrants it.

One other quick question in general, are you finding it pretty easy to get equipment these days in the sense of either delivery and/or pricing.

Yeah a lot easier than it was about 6 months ago. So, my comment, I think, you obviously talk to equipment guys they can give you a better guide. What we're hearing is only confirming what the equipment suppliers have been saying, given the severity of the market right now, despite some reports of activity there that it's really a lot lighter than you would suspect and a lot of companies are trying to reschedule or cancel in order to avoid incurring those costs right now.

So in general, I know Bill made a broad brush comment earlier on. You're still sticking to the plan you have for capital expenditures within that range, generally what you thought you would be doing about 6 months ago or based it back.

Our range is obviously been phased back. We were talking about 1, one and a half billion in the last call, now talking about 800 to 1.2. I would make the caveat we could down that even further if we needed to. We're going to make sure that we keep a pretty good financial balance sheet. If that means we got to cut our capital further well, if it means we have a little more to spend because the prices come back up then we'll spend more. We're focused right now on advancing the technology as opposed to adding new wafers which we think is the intelligent thing to do right now. So that's where we're going to spend our money.

Great. Thanks a lot, Steve.

Next comes John Cross of Morgan Stanley.

Thanks very much. Simple question. On what's the mix of 256 and where that is going?

It's been pretty much static John around 65 percent of our bits shipped and produced now since early spring.

And content per box, can you update on trends there what you might think see going forward.

We got a huge boost in the last 12 months on the consumer platforms which you're well aware of so on a relative basis we see that leveling out. The data that you might see on the screen now shows it's coming down which is no surprise. We do have a significant head room however on the commercial space and as well as on the notebook space. Both of those have seen nice increases in the recent couple of periods we would expect that to continue. One of the other data points we have, of course, is our after market Crucial business selling memory upgrades directly into the retail market if you will. In terms of the sweet spot memory module density there that continues to go up very nicely. We still see memory being a real good bargain for increasing performance of PCs.

What about daily order rates out of Crucial recently?

In terms of orders per day they are up fairly significantly over the last 30 days as opposed to the prior 60 or 90.

And just one big picture question. Steve, perhaps, could you talk about the industry potential for a consolidation. And what you might see here over the next year and what you think kind of break-even pricing might be for the DRAM industry over all?

I'll leave on the break-even pricing that's pretty tough because it's a moving target. In other words we're reducing our costs every quarter that goes by. And obviously that's going to be in comparison with the selling prices are doing. I don't know if I have a good answer to that. With respect to the overall industry in the consolidation, obviously, I think that the wild card here is what does Hynix do or not do. We haven't been engaged with them on any discussions. And as we understand it they have am new board essentially. And they're trying to sort out their problems between the creditors and the government and their new board. We don't have a lot of insight right now. That's the wild card.

If you look at other than that you've got Infinity and Samsung, Micron. And I think the three of us are relatively strong in comparison to the other players. And then you have - we hear all kind of rumors out of Japan what is going to Lpeda (ph) whether stay separate or one hat. Taiwanese, some are trying to stay where they're at and some are trying to build a little bit. I think it's pretty difficult to predict any more consolidation other than what happens I suppose with the wild card on Hynix right now.

Given choice, would it be more interesting to make consolidation happen or let it play out.

What are you really are asking is are we on the acquisition trail on the DRAM business. Primarily we're focused in imposing our own operations. I would say one thing to point out, we spent quite a bit of money in the last 12 years. Our capex budget was relatively large in comparison to others,, maybe even Samsung, on a going forward basis we've positioned the company to be highly leveraged. I'm not talking about debt. I'm talking about the ability to take our current assets and make them more productive.

We've mentioned before we don't need to go out and build new fabs to do 300 mm we have that in place. We don't need to go out and make it to the .13. We need to leverage it on a going forward basis but we're in good shape there so we think we have a lot more to ring out of our current operations and that's where we're focused. If something else shows up as an opportunity of course we'll consider it. I don't think that that's where our focus is right now.

Would you stepping up investment in Netcom and those other product areas at this point.

Yeah. What you're seeing, I think a lot of what you see in our R&D and manufacturing capability investments are to put us in position to build these products that are in design now. Mike mentioned and Bill mentioned a little bit, we've been spending money in what we consider to be more nontraditional computing applications, obviously in the networking space and then in the imaging space and you've got to remember, it's not hugely expensive but relatively costly to have multiple design teams in these products. And you've got an 18 month turn time before we'll start seeing that in any kind of revenues. When these thing start hitting, when we start having a tape out or 2 tape outs a month in these designs and we've had to do things up to now to get into position to be able to build those. Had to spend the money for it but haven't received the benefit of all that R&D activity and I think that's something we pointed out as we're going forward that we're going to be able to leverage.

Next question is Tom Thornhill of UBS Warburg.

This is Razmig in for Tom. What is the flash and SDRAM percentage of sales? Has that, in other words -

Still less than a couple percent of total revenues.

Second question, what was your blended cost for 120 density.

We don't give out our costs.

The average ASP, actually. Of the average selling price is for 120 megabit density.

Again for competitive reasons, we don't want to give specifics. But for the 128 Meg, SDRAM you've primarily been if the 2 dollar range for most the quarter and recently a little bit of pressure on that for the high one dollar range. For the 128 meg DDR, it started in the 2 dollar range in the beginning the quarter and ended up exiting at around 3 or high 2s. And the 256 Meg for the most part is double that, in most cases.

Last question. Spot volume versus contract volume for the quarter.

We were just in the spot market what we characterize as spot, we were 20 percent of our Meg bits we moved into spot market.

Sure. Thank you.

Our next question is coming from Doug Lee of Banc of America Securities.

This is Sumeet (ph) Donda (ph) for Doug Lee. A couple of questions. I think you gave this out but I might have missed it. The production - the expected production bit growth for the next quarter and the full fiscal year could you repeat that.

For fiscal Q1 looking at production to be up low double digits and for '03 depending on market conditions 40 to 50 percent for Micron in both cases.

The second question, on the inventory writedown, now the writedown this quarter I'm assuming includes roughly 44 million from past or the benefit this quarter includes 44 million from the past quarters is that accurate?

The numbers again. Correct it was 174 million dollar year-end quarter-end writedown and the net effect for the quarter including the benefit from prior writedowns is 132.

And for the forthcoming quarter that expectation is 120 million and that includes the benefit from past writedowns too.

Correct. Approximately 240 previously written down amount would be expected to flee through in Q1 of '03.

One final question. Any indications of an inventory build in terms of either SDRAM or DDR parts at any PC OEMs.

No. We don't see it. With the inventory management programs the OEM's have no reason to build any inventory unless they would speculate on price moves and I don't see any indication of that occurring in this environment.

Okay. Great. Thank you.

Thank you your next question is coming from Anish (ph) Goyle of from Newberger Berman.

Yeah. A couple of short questions. What is DDR percent of our total production today and where do you expect it in the next 2, 3 quarters.

Sure. As I mentioned before. We shipped about 40 percent in our fiscal Q4, that will be 50 percent in fiscal Q1, before that 60, 65 percent at the end of the calendar year.

What do you expect in depreciation in fiscal '03.

Looking at 1.3 billion I believe.

Up about 100 million from the current year.

Will it be right to assume that you are gross margin positive after excluding all the charges in your DDR product.

The net inventory, the inventory writedowns were principally on synchronous product. Which gives you an indication that there were very small adjustments on DDR.

So, I come out at about negative 5 to negative 10 percent on gross margin line. After adjusting it for writedowns, what does that mean? Does that mean you are breaking even in DDR or gross margin positive? Because seems like pricing for DDR is much better than synchronous.

We're not going to get you more specific than you are right now. One of the challenges of course is that it's a continual slope. That's numbers change all the time. As soon as we say one thing we're going to be updating you the next time. In general I think you're thinking about it the right away. Again just reiterate that we've got some pretty significant cost reductions coming up for the next several quarters will certainly help us on that side.

If I look it on a slightly different way, cost of goods this quarter was about 780 million dollars. What will change your cost of goods next quarter besides the change in depreciation.

What will, if I understand your question right, what will - what impacts or what effects are there that we are anticipating on cost of goods sold in Q1 over Q4.

Right. Besides the change in depreciation.

Despite the changed depreciation. Certainly a higher influence of .13 would do that. Any effects of changes in ASPs would certainly have an impact.

How will the change in ASP change your cost of goods line.

Sorry, I was jumping to the margins side of that. Which I assume is where you wanted to head eventually.

I want to keep the conversation on cost of goods. If I look at the cost of goods on a sequential basis, besides the change in depreciation, what else will change your cost of goods next quarter.

Pretty direct result of whatever quantity you want to estimate being sold in that period and the cost reductions coming through manufacturing.

Okay. Thank you.

You bet.

Thank you. Our next question is coming from Charles Boucher of Bear Stearns.

I think just have one question up here. Last quarter you said you were starting to see some increase in bookings from some of your communications networks OEMs. Did that continue and how does that situation look today.

It has continued and although it haven't increased dramatically and for the most part stabilized but lower levels than we saw, in mid-2000 time frame.

Sure. But the improvement you saw last quarter maybe got modestly better and you're seeing some activity there.

It basically stopped and yeah, it's stable.

Our next question is coming from Ben Lynch of Deutsche Bank.

General question. Given the increase we've been having in DDR pricing and let's say signs that that's going to continue at least in the near term what are PC OEMs saying about DDR content given we've seen in the past some sensitivity to SDRAM pricing and the other question I have on DDR is, various DRAM manufacturers are able to swing wafers between DDR and SDRAM to customize difference stages in the process. Where is Micron? Are you able to do that? Do you have to fix SDRAM at the start of the process or more towards interconnect stages or latter ends the front end process.

I think I'll the latter one first. Those competitors that can jump back and forth, if you will from DDR so SDR are paying a significant penalty on the SDR.

The guys say it's five to ten percent. Maybe that's significant but they say it's five to ten percent.

Well in the game we play, every percentage point is very significant so you can draw your own conclusions from there. With respect to the DDR memory content -

You're saying you guys fixed it pretty early on in the process.

At .15 and at .13 we fixed it at the wafer start level.

Less flexibility but perhaps lower cost.

That's correct.

Okay.

On DDR memory content standpoint on the consumer platforms, well above 256 megabytes per system, that's probably a function of declining returns, if you will, on a cost performance standpoint with respect to the consumer platforms. On commercial platforms we're seeing a significant increase whether it be on DDR or synchronous DRAM platforms.

The other general question, given what's happened since in the DRAM market, what's your view on the chance you had in the past to maybe merge or acquire Hynix are you glad that didn't happen because things got worse or are you sort of the same view you had on that deal as you did back then.

I think that the way to look at it is what's the value proposition for Micron and what's the value proposition for our shareholders. At the time that we were in negotiation with Hynix it was a value proposition we thought would be beneficial to us. The markets has obviously changed. Value of Micron has changed in that time period and also the assets in general in the DRAM industry. It's a moving target. We believe at the time it would have been the right thing and had we been able to achieve that I think we would have moved forward with that as we had planned and it would have been a benefit to us.

But do you think you would be any happier today as senior management team in the company if you were also in the process of trying to interested great Hynix.

You're trying to speculate on what impact that would have had on the market. It's difficult for us to predict that. Clearly, I think consolidation we would hope will stabilize the market as we move forward so we don't have a lot of irrational behavior in the market. It's difficult for us to predict how that would play out.

Obviously I said we're focused today internally as opposed to externally on an acquisition. Any opportunities that surface we'll look at like we have in the past. I'm neutral basically. I'm not saying that we are happy now that it didn't happen or we would be happier if it did happen at the time. We had plans and models to make it work. And if it would surface again whether Hynix or any other company, we would have to rerun our models and see if it made sense for us to do.

Great. One last quick question. Any sort of layman's update on the Department of Justice investigations, quite legalistic in your press release.

Not really. We obviously are cooperating with them. They're in their investigation. We don't know what their own time schedule is for completing that and we're going to continue to cooperate as they request.

Okay. Thank you very much.

Our next question is coming from Bill Deselum of Davidson Investment Advisors.

Thank you. I wanted to. Circle back around to the DDR and the supply demand issues taking place there. To what degree to the competitors lack of capital expenditures in some cases may be influencing their inability to get supply out or just simply a function of it being a tougher design?

I think that it's clear there is an impact there. I think that in particular if you look at it from a front end perspective on the wafer fab side there's really not that big of an impact. Building a wafer that is DDR or synchronous DRAM, essentially very similar. Small differences but not much.

Where it really takes effect is in the back end. If you start looking at changes in packaging or testing, in particular testing, and remember that the back end is the cost of the back end in comparison to the front end is significantly higher than it used to be. So rather than a 10 to one ratio back in the old days its more like a 2 to one ratio, so it's pretty expensive. As a result, I think that's where you're seeing it get squeezed. And the inability of a lot of these companies to buy back end equipment because they don't have the cash to do it is hurting them. That's on the capex side.

Clearly there are yield issues as well. Micron had to go through its learning curve which we did a long time ago. These other companies are trying to make hard transitions to the DDR, obviously because of the pricing environment are probably facing what we face in the early transition, because it is a much more complicated device to get the yield look you want to and I think that's what some of these other suppliers are facing.

As a follow up, I'm assuming that Infinion and Samsung also doing okay with DDR? Is that a fair perception.

I think it's clear Samsung is, I'm not as familiar with Infinion. Backup they're a good company and they have obviously performed pretty well so I assume they're doing fine witness as well.

In the rest of the pack, I would throw Hynix, and the Japanese and the Taiwan niece, are any of those competitors having any success with DDR as far as you can determine.

Based again on the information we're getting back from our customer base, our sales and Samsung are clearly out ahead of anybody else with respect to DDR penetration.

That's helpful. Thank you both.

Our next question is coming from Alec Berman (ph) of Pangea (ph) Capital.

Hi. How are you doing? I want to get more of a strategic. 300 millimeter, strategically thinking bit maybe give us an update in your thinking, if you look at out there, you have some sort of Taiwanese type players who are just smaller companies than you who are trying to aggressively move to is. Samsung is going forward with it. Infinion had it for awhile. I guess, maybe articulate your thinking on the whole pros and consequence of it for the industry for you, why you're taking a more cautious approach. Your more recent thinking on that issue.

300 millimeter, our thinking has remained relatively consistent. There are - there's two reasons to advance a 300 millimeter. One of them is to just get more bits out and at some point in time, that translates in the second item which you hope would be a more cost-effective device. Obviously has the industry moves forward they'll be driven towards 300 millimeter because most of the equipment will be more compatible as you go down to smaller and smaller geometry. We don't want to make a move on the production level until we think that it will be a on a cost per bit basis more effective than we're currently doing on the 8 inch. If you look today for a relatively average speck for a 300 millimeter bare wafer you're going to spend about $350. If you want tighter specs, probably more on the order of what we would require, then the net cost is going to go up from there.

It's really not cost competitive when you consider buying an 8 inch wafer for about 55 dollars. That's always going to play into our thinking. Having said that, we are going to continue to do our 300 millimeter development. We already mentioned that we have a pilot line that we'll put in Virginia. That's happening as we speak. And we'll continue that to make sure we're ready to convert when the economics look like we should convert.

Something to point out which I mentioned earlier. One the reasons do you is to expand capacity. We think we're relatively intelligent about how we do it. You can read a lot of reports about what market share companies think they we're going to have, what expansion they think they are going to get in the next quarter or 2. This market penalizes you greatly on the return basis if you expand when they're not willing to pay the fair value the silicon, I guess the way to put it.

If you look at what Micron has done, it we've gone through a transition because we have acquired an asset like we did with TI, we are really making huge advantages in the cost per bit as we make that transition. We historically have not expanded in new wafers, when the market is going to penalize us the most and that's the exactly the situation we're in today. So adding new silicon today makes no sense. If you're going 300 millimeters to add new silicon that's a mistake in the environment that we are in because you're not going to get the return for the investment required. If you advance your process in order to get a lot of cost reductions, that makes more sense, but in that particular case the economics don't play out yet.

Does it ever make sense to - will it ever make sense when you decide to become more aggressive to retrofit old lines or is that not feasible, you have to start with a new template. Can you go ahead and do that or partially take a line and a fab and turn it to 300 millimeter. Is that ever a logical intermediate step or would that not be logical.

It makes a lot of sense. If you look at Micron's history, we build 8 inch today in the very first fab I stood in personally building 4 inch. So it can make sense. There are limitations, I will tell you now that we have a particular fab or 2 that will not make the transition. In general, I think the majority of our operations can make the transition. When we talk about making the investment for 300 millimeter there are a couple of key components. There is the facility itself and then there is the equipment in the facility.

The facility itself as we have proven can last a long time, you have to do some modifications through time in order to make what happen but the facility itself in general can remain pretty useful. We have both Lehigh and the facility in Virginia, in addition to the facilities that we're currently running on our process and the most of those are convertible. The key is, do you put equipment in those facilities that's going to depreciate at 3.7 year average rate. That's what you don't want to do. You don't want to put equipment in these facilities whether they're old or new or retrofitable or not when the market is not going to generate the return for the cost of that equipment. That's where you have to think intelligently about what you do. Our facilities, as I mentioned, most of them are convertible and we will anticipate, we'll take current facilities and convert them and obviously make use of the new facilities when the market demands.

Thanks a lot.

Time for one more question.

Our last question will be coming from Eric Ross.

Eric Ross from Investec. Two part. I didn't quite get the percentage of revenues going into servers and some the activity you're seeing. The second question flash as a percentage of revenues and the market dynamics you're seeing there.

On the server-side, Eric, I don't have any precision in these numbers but approaching, maybe a third to a half of our megabit shipments for the majors going into servers. Again, I don't have it at my fingertips but we probably could get more precision on that.

Increase or decrease or are you seeing a little bit of strengths or flat there.

It would be a stretch to say that we're seeing relative strengths in the server business to be honest with you.

So things are worse there?

Don't know enough off the top of my head to say they're worse but not a big surge in demand.

As far as flash revenues.

Very low single digits are in terms of percentage of our revenues.

Activity there.

Flat.

Flat. Great. Thank you.

You bet. Thank you. We would like to thank everyone for participating on the call today. If you will bear with me I need repeat the safe harbor protective language. 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 is he including the company's 10-K and 10-Q. Thank you very much for joining us.

Ladies and gentlemen this does conclude today's teleconference. You may disconnect at this time. Thanks for participating. Have a good day.