美光科技 (MU) 2005 Q4 法說會逐字稿

Good afternoon, ladies and gentlemen.

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

At this time, I would like to welcome everyone to the Micron Technology fourth-quarter and fiscal year-end 2005 conference call.

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

After the speakers' remarks, there will be a question-and-answer period. (OPERATOR INSTRUCTIONS).

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

Sir, you may begin your conference.

Thank you, Jason.

I'd like to also welcome everyone to Micron Technology's fourth-quarter and fiscal year-end 2005 financial release conference call.

On the call today and joining us remotely is Steve Appleton, Chairman, CEO and President.

In the room with me is Mr. Bill Stover, Vice President, Finance and Chief Financial Officer;

Mike Sadler, Vice President of Worldwide Sales; and Mark Durcan, Chief Technical Officer and Vice President of R&D.

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 2005 financial press release, it is available on our website again 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 at 5:30 PM daylight savings time.

You may reach that by dialing 973-341-3080, using a confirmation code of 6146538.

This replay will run through Thursday, October 6, 2005 at 5:30 PM again daylight savings time.

A webcast replay will be available on the Company's website until September 29, 2006.

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

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 after the date of the presentation to conform these statements to actual results.

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

Thanks, Kipp.

I'll first go over summary financial results for the fiscal year and the fourth quarter ended September 1st for those listeners, who may not yet have access to our press release.

For the fiscal year, net sales totaled $4.9 billion, and the Company recorded net income of $188 million or $0.29 per diluted share.

These results broadly reflect the fruit of several years' diversification and R&D investment efforts.

One example of our success is our CMOS imaging business, which grew 200% year over year to become the leading CMOS image sensor provider in the industry and our highest gross margin product line.

The balance sheet reflects significant improvement over just a quarter ago.

Since third-quarter end, Micron has reduced its total debt by over $125 million and reduced total inventory by approximately $50 million.

During our third-quarter conference call, we communicated that we had tolerance for the inventory build and finished goods of PC DRAM.

Our market assessment was validated during the fourth quarter as seasonal demand indeed did allow us to draw down inventories at better pricing than was available earlier in the year.

For the fourth quarter, net sales totaled $1.26 billion, and the Company recorded net income of 43 million or $0.07 per diluted share.

In the quarter, our CMOS sensors and PSRAMs each achieved 40% growth quarter over quarter, and our NAND flash business grew five-fold.

Gross margin for the year came in at 23%.

And as you can see from the chart, the fourth quarter rebounded to 22%.

Gross margin improvement in the fourth quarter was directly attributable to further product mix shifts to image sensors and specialty DRAM and products, including pseudo-static and synchronous DRAM and a relative stability of average selling prices in a period in which the Company achieved significant cost reductions.

The gross margin on TECH Semiconductor products was slightly lower than the overall reported gross margin of 22% for the fourth quarter.

That's a significant improvement in margin on TECH products, which resulted from the lag pricing arrangement and increased average selling prices for the mix of products they supply.

Selling, general and administrative expenses have stayed in the 85 to $90 million band throughout fiscal 2005.

We expect a quarterly run rate in 2006 to be between 85 and $95 million, in part influenced by the effect of stock option expensing, which took effect at the beginning of our fiscal 2006 year.

Research and development expense for the fourth quarter was $151 million, the run rate that held for most of fiscal 2005 as an overall reduction of $150 million comparing 2005 to fiscal 2004.

Future R&D expenses will vary significantly with the number of wafers dedicated to new device development and qualification and likely range in the near-term between 150 and $175 million.

From this slide, you can see that cash flow provided by operations for the fiscal year improved again relative to the prior year.

The improvement reflects our successful diversification into specialty DRAM and CMOS image sensors, and capital spending for fiscal 2005 came in at about 1.3 billion.

This early in fiscal 2006, our forecast for capital spending is a range between 1 billion and 1.5 billion.

As of year end, Micron had cash and investment balances approximating $1.3 billion; total debt declined to 1.17 billion at fiscal year end.

During the fourth quarter, we repaid the remaining $100 million balance of a $210 million subordinated note that was scheduled due in September 2005.

Call protection on our $632 million of convertible notes expired in February 2006.

Those notes, which are due in 2010, are convertible into common stock at a conversion price of $11.79 per share.

I'll turn our commentary over to Mike.

Thanks, Bill.

Heading out of the quarter, we assumed we'd see an early seasonal boost from computers and cell phones, the primary driver of the demand for our semiconductor components.

With a quarter now behind us, I would say that things unfolded just about as expected.

The demand strength did result in increased commodity DRAM prices off the lows observed in the prior quarter.

We saw relative price stability and enhanced gross margins across the balance of the product portfolio.

We had another strong quarter of executing on our goals of growing business in mobile phones, server and consumer electronic markets with a mild reduction in dependence on the commoditized desktop and notebook personal computer markets.

We should expect this theme to continue moving forward.

We achieved sequential megabyte shipment growth of about 30% relative to the prior quarter.

Without a healthy PC-driven demand environment, it would not have been possible for us to achieve this growth in conjunction with the realization of a commodity DRAM price-per-bit increase.

Most industry experts are projecting PC unit growth to exceed 10% in calendar 2005.

Our customers are forecasting DRAM demand increases, driven both by unit growth and accelerating memory content per system.

In summary, we are quite pleased with the DRAM demand that is being created by the PC manufacturers and their end customers.

We are in new territory in the computing market with respect to multiple memory interfaces concurrently driving substantial DRAM demand.

In years past, the model called for a rapid wholesale transition from one architecture to the next with very little overlap and demand.

While there appears to be no question that the mainstream is headed towards the DDR2 interface, there continues to be significant demand for both synchronous DRAMs and DDR architectures from the computing market.

As we look out on horizon of processor and chipset roadmaps, this segmented demand should sustain for sometime into the future.

We're pleased with this situation and believe that it plays into our strength as one of the few DRAM providers with an expansive portfolio that covers the three architectures across several densities of components and modules.

These are exciting times for us in the mobile phone arena.

The market is expanding in terms of unit volumes and a wide variety of applications being enabled in mobile handsets -- is creating more demand for semiconductor content from Micron.

The camera-enabled phone penetration rate will approach 50% this year, and we have become the largest image sensor shareholder supporting this boom.

We're now engaged commercially and growing share with every major mobile handset maker in the world.

Our customers are quite pleased with the image quality being enabled by our silicon and the commercial success that is being achieved as a result of having access to our technology and operational infrastructure.

We are proud of our achievements in fiscal 2005 with the imaging products and technology.

As I mentioned, we are now the number one shareholder.

We believe we are the only major provider of CMOS sensors that is currently growing share.

We have demonstrated the industry's smallest pixels at 1.7 Micron and are shipping sensors with 2.2 Micron pixels.

We have established a solid foundation, not only in the mobile phone market but also in consumer products, medical applications, and in the automotive industry.

Our model of drafting image sensor production off the memory process technology is proving to be efficient from both an R&D and capital deployment standpoint.

The family of low-power DRAM products inclusive of cellular RAM is complementary to our image sensors in mobile phone and other handheld applications, such as digital still cameras and personal digital assistants.

As customers strive to simplify their supply networks, we offer them the advantage of one-stop shopping for both imaging and memory needs.

As new applications are being enabled in mobile phones that require large amounts of memory, our ability to offer both memory and imaging chips puts us in a unique position to provide value and competitive advantage to our customers in this space.

As we have mentioned in past sessions, we completed our commercial entry into the mass storage NAND/Flash business and saw our first quarter of significant revenue and margin contribution from these products.

Demand for Flash is explosive and being driven by USB drives, detachable memory cards, solid-state MP3 players, mobile phones and a variety of other consumer electronic products.

The Flash opportunity is large, growing rapidly and is an ideal means for Micron to effectively leverage both the capital and research and development spending taking place in the DRAM business.

From our vantage point, the end markets are quite healthy in terms of end customer demand with both the PC and cell phone markets again growing double-digit percentages this calendar year.

The products that our customers are developing and bringing to market continue to gain momentum in terms of richness of semiconductor content, in particular, content that fits nicely into Micron's product portfolio.

As each quarter goes by, we are making more progress toward enabling a richer mix of Micron revenues into what we believe can be sustainably higher margin markets.

Thanks for your continued support, and I will turn it back to Kipp.

Thanks, Mike.

We would now like to take questions from callers.

Two reminders please, Steve is joining us remotely, so I will be directing more questions than normal.

Secondly, if you're using a speakerphone, please pick up the handset when asking questions so that we can hear you clearly.

With that, we would like to open it to questions.

(OPERATOR INSTRUCTIONS).

Adam Parker, Sanford Bernstein.

Just one point of clarity and a question.

Bill, did you say that the gross margin on TECH was lower than the overall corporate average but above the commodity PC DRAM you produced during the quarter?

Yes, the statement we made was slightly below the indicated 22% overall gross margin for the fourth quarter and certainly higher than the core DRAM.

For the core DRAM.

Okay, great.

Could you provide a capital spending guidance number for 2006 fiscal year, please?

I indicated this early in fiscal year 2006; it's a pretty wideband.

So it's 1 billion to 1.5 billion.

So it will take some time for that to shape up.

(indiscernible) you falling on, on that?

Because given the revenue growth and what seems to be lower kind of capital spending, you are therefore having a capital spending to sales ratio that's declining kind of meaningfully.

Can you talk about this new revenue?

I don't know if Kipp or Steve or whoever wants to answer this, but can you talk about this new sort of revenue diversification?

And what do you think that does to your structural capital intensity over the next 2, 3 years?

Steve, would you like to take that one?

Hey, Kipp, let me jump in here.

You know, Adam, we really have somewhat of a bifurcation in the model, if you will.

Clearly, the capital draft, if you will, of the products that don't need the leading-edge process -- they may need leading-edge talk technology -- it's just not the leading-edge process itself.

Clearly, that will be successful in driving down the capital needs to bring those products to the market.

Of course, the counter to that is that's not necessarily true of the NAND memory, which does continue to drive you towards an advanced process.

So to the extent that we continue to expand in that market, it will of course counter the effect that we're getting by having other products that don't need the advanced technology.

I do not think we can as of yet assess exactly what those implications are because it depends on how far each of those product groups expand.

There's no question for us overall if we were solely in NAND, if you will, or if we were solely in the computing DRAM, there is no question that it would drive our capital needs much higher than it will for the model that we have right now.

I see.

One last thing, I know I asked a similar question on the June call, but you know there's a lot of rumblings about Intel and what they might do in the NAND Flash arena.

Do you think what they do might impact your plans?

Is there any sort of apprehension or excitement or anything about what Intel could do in that space?

Yes, well obviously, we cannot speak for them.

I do not think it's going to change our plans in terms of trying to grow our presence in critical mass in that market.

Douglas Freedman, American Technology.

If you guys could expand a little bit on your plans in NAND and how quickly you're looking to ramp in that business?

And you touched a little bit on the capital intensity.

How much of the capital budget do you think is going to be consumed by your NAND expansion versus DRAM and CMOS?

Fortunately, for us, as we stated before, the capacity is somewhat fungible between DRAM and NAND.

So there is somewhere between a 90 to a 95% crossover.

As a result, we don't have to decide where that gets allocated to go forward; we just have to continue to put that in place.

So when you think about what that means for us, effectively, we have the ability to expand NAND on our current path.

Now one thing I should say is that essentially a large percentage of the Boise facility is already being ramped on NAND.

As you know, we have the Virginia facility, which we stated that we're basically 50% on DRAM, and we can allocate that second half any way that we want.

And then of course beyond that, we will just have to see where the market drives us.

But from where we sit today, we actually -- because we are only partially ramped in Boise itself on the NAND -- from where we sit today, are output will continue to grow significantly in the next 12, 18 months.

I guess what I'm trying to figure out is, if you were to invest capital at a faster rate, would you be able to expand your share in the NAND market?

Well, I mean the answer to that is yes.

But when you say invest in a capital rate, that means two things for us -- we can continue to convert current capacity, and we can invest in new capacity.

So those in combination will ultimately determine how fast that grows.

But we currently are converting current capacity.

We are in the discussions right now, as we look at our capital plan, how much new capacity would be put in place for NAND.

Michael Masdea, Credit Suisse First Boston.

This is Randy Abrams for Michael Masdea.

First question, just want to see if you look ahead to the next few months, what are your expectations for the production ramp in DRAM and then also your outlook for growing your non-memory products?

Yes, Randy, this is Kipp.

We're looking at high-single digits for bit production moving forward over the next couple quarters, and that's a combination of both DRAM and NAND bit.

Beyond that, maybe talk about your visibility now or trends in the PC market, how behavior has been from some of your customers over the last couple weeks for both desktops and notebooks.

Sure.

This is Mike speaking.

First of all, we're right in the middle of what we had expected would be a seasonally strong demand period.

And I think things are playing out from a demand standpoint about like we expected.

By the way, our belief is -- if we aggregate our customers' DRAM demands for calendar Q4 timeframe, which roughly coincides with our fiscal Q1, we're looking at about a 17% quarter-over-quarter increase in demand, which again kind of matches what our expectations would be.

Now with respect to the components of demand, we are continuing to see a shift towards stronger notebook sales than desktop sales.

For us, that's really I do not care.

It's essentially the same components that are going into notebooks just on a slightly different module.

But it is the same components that are going in notebooks as desktops.

We are still continuing to see a gradual shift from a DDR interface to DDR2 in terms of the sweet spot memory for PCs.

And just one final question, if you were to take the midpoint of your CapEx budget for fiscal '06, how much difference could you get in DRAM?

And then how much growth could you get say in your non-memory products?

Randy, that would require us to give you a more accurate accounting of where we are going to allocate those wafers, as Steve was describing.

So again, we will give it to you in a combination fashion, where we look to be somewhere in that 45 to 55% bit growth range for fiscal '06 over '05.

Richard Prati, American Technology.

Could you just comment given your diversification efforts into new product areas and decreasing exposure to PC-related, what would your ability over the next 12 to 24 months be to generate free cash flow and remain state-of-the-art and leading-edge on your technology?

Well, as I had mentioned before to Steve, there is two ways to look at technology.

State-of-the-art technology and CMOS imaging is a much different issue than state-of-the-art technology for a computing DRAM.

So you have to look at the way that our business is dividing itself.

And so our capital expenditures for the process requirements for DRAM are clearly going down as we get less exposed to that market.

The capital expenditures for NAND though -- because it needs leading-edge process -- will obviously have to either go up, or we would not keep paying through the advancement of technology.

And so that part of the business still has to have capital investment.

It is really the non -- call it the non-advanced process memory -- and there's lots of it, by the way, whether it's pseudo-static RAM or low-power DRAM or some of these other derivatives.

Those markets are actually growing, and they do not require the same capital investment level either.

Obviously, we are comfortable where we're at with our ability to invest in capital.

And Bill has already giving you the range of the 1 to 1.5 billion.

I think that we feel like that's going to get accomplished, what we need to get accomplished in both parts of those businesses.

Steve, I might add that the synchronous business certainly has had a stability and longevity to it and is one of our strongest margin and a very nice cash flow off of that business.

And we anticipate that that is going to continue for sometime period.

John Lau, Jefferies Co.

Focusing more on demand products, what do you think the sweet spot is today, and where do you see that going next year?

I think you guys are on 4 gigabits right now.

So in terms of the new capacity, you mentioned also Boise and Virginia, how's about in the expansion of yopi (ph) fabs in Italy?

Thank you.

I will talk about the market standpoint from where the sweet spot is and where we think it's going, John.

This is Mike.

And I think well, I will defer to Steve on the fabrication strategy with respect to NAND.

Where our primary product today is the 2 gigabit component.

We are actually shipping it in the form of 2 gigabits, 4 gigabits and 8 gigabits via stacking it.

So I would say the sweet spot today in terms of demand would probably be at the 2 gigabit level.

And I think we would expect that to shift to 4 gigabit components next year.

Many of those 4 gigabit components, by the way, will be shipped in the form of a two chip or a four chip stack to make a gigabyte or 2 gigabyte memory stack, if you will.

Did we get your questions answered?

And the fab question?

On the fab -- in particular with respect to Italy -- the Italian operation is actually ramping the imaging business, and that's going well.

But we expect that will continue to be the case.

It will move more -- it will continue to have more of its resources allocated to imaging over the next year, 1.5 year as we continue to finish that ramp.

We think that is the best use of the capacity that we have there.

Does that answer your question?

Great.

Thank you very much.

Glen Yeung, Citigroup.

Thanks, nice quarter.

I have a couple or three real short questions and then two maybe a bit more in-depth.

First one was just in terms of revenues in the quarter.

Was Flash revenue greater than image sensor revenue in the quarter?

No, it did not quite reach that level.

Just by the sound, it sounds kind of close.

Is that a fair assessment?

Yes.

Then, it's kind of a response to a previous question about state of business right now.

One of your competitors has been suggesting that they may be seeing some cancellation from a major U.S.

OEM, and frankly our firm has not been able to confirm that.

But just any thoughts you may have on potential DRAM cancellations in this kind of market?

We're not seeing any cancellations.

As I mentioned, the demand environment is quite robust.

So we feel pretty good about demand.

I would suggest that my observation is that on the supply-side -- the supply-side DDR2 is a little bit ahead of where demand is.

So we're starting to -- we have seen some mild price pressure in the DDR2 area.

And my observation again would be that if we looked at the market right today, DDR2 supply probably exceeds demand.

So that maybe what the other supplies are referring to.

But certainly, we are not experiencing any cancellations to the extent that we are able to meet the commercial requirements of the customers.

And so if I look at inventory right now, would you say that it's pretty comfortable except for maybe a bit of DDR2 of kind of where you guys are?

I can only speak for our case;

I cannot speak for anybody else.

But yes, we are quite comfortable.

And then, I guess the two other sort of bigger questions maybe for Steve -- one is, your thoughts on the supply/demand balance for 2006 in the PC DR (ph) space to begin with.

Then, the second question is, what are your thoughts on consolidation in the industry both in terms of your willingness to consolidate and whether or not you think it is actually going to happen at all?

Well, on the first question about the supply-and-demand balance actually, you know Mike may be a little better situated to answer than I will.

Clearly in terms of the supply side, you know it is going to depend on how much of the capacity that has been forecasted to come online.

It only comes online.

As you know that a lot of that tends to evaporate the more difficult the market or if pricing doesn't come up and they don't quite raise the capital and they make a little bit less investment.

Although, I think it would be a good assumption there will be capacity brought online by some of our competitors.

And then I'm going to defer the demand to Mike, and he can comment about that in a second.

But with respect to the consolidation, you know, there is really only five of us that develop technology anymore.

And consolidation among those five of course is always speculative.

You can read the media reports just like I do in terms of one or two of them.

We don't really know where some of that stands right now as they try to do their internal evaluations of what they should do with their businesses.

I will say though that if there is consolidation opportunities for us to look at, we are definitely going to take a look at them.

We still think the industry will consolidate.

And as a result, we of course -- if you look at our history -- have participated in that in various forms.

If the opportunity presents itself, then we will continue to look at it.

On the demand side next year, I think we've got to two boomers with respect to demand creation occurring next year.

One on the PC side would be the new Microsoft operating system, which is going to be a product I believe in the second half of '06.

We are expecting, based on input from our customers, that they are going to start putting Vista compliance -- from a hardware standpoint, Vista-compliant systems out into the marketplace in the early part of next year.

That really is going to result in we believe an effect of doubling on memory content per system.

You know, those things only occur once every few years.

And when they occur, they typically result in a huge incremental jump in terms of memory content.

So we have got that working for us from a demand standpoint in PCs in '06.

On the Flash side, of course, we had the continued encroachment of solid-state MP3 players on hard disk drive-based MP3 players occurring really throughout the rest of this year and into '06.

And I think that is going to result in a real boost in demand that some of the third-party market research firms are not really accounting for either.

So those are two huge events I think occurring in '06.

They are going to be big demand drivers.

And our expectation is that there is going to be some toggling from a capacity utilization standpoint -- some toggling of capacity from Flash to DRAM frankly speaking based on opportunistic profitability of the two.

And my belief is that those two big demand events next year as well as the ability from a supply side to toggle capacity utilization should result in a pretty good market environment for us.

Jim Covello, Goldman Sachs.

Quick question on the image sensor business, Micron is obviously doing a terrific job on the market share side.

Some of your Korean competitors are making more noise about being more active in image sensors and the next-generation technologies.

Do you think they are a credible threat, especially to your margins?

Or do you think you'll be able to effectively beat them at the next-generation technologies as you have with the current generation?

Thanks.

Steve, do you want to get that?

If I could jump in, Mike, real quick.

In terms of the competitive nature on the development side, there's no question that -- by the way one of them was probably in the sensor business before we were, and I think we've gone past them by quite a far margin.

So I think we're doing a great job there on the development side.

Remember that as I mentioned earlier, the technology is not necessarily associated with kind of executing on a process roadmap; although, certainly, it is part of it.

But it's the coupling of the development and how you think about the capability of the sensor itself.

We have been able to so far I think keep a pretty good lead.

Having said that, there is no question that the larger the market grows, the more people it will attract.

Now fortunately, a lot of that technology is not simple to come up with and not simple to duplicate.

So we think that we are in a great position as we move forward in order to continue to be one of the leaders.

And what is interesting of course is that our sensors are in really all the handset manufacturers today, including some of our competitors.

I do think in general though, over time, what you will see is the companies that have tightly coupled, really think about it as design and development activity, tightly coupled with their capabilities and manufacturing and process -- that those companies will continue to be more successful over time because of the ability to optimize the performance of the device.

So clearly, one or two of them will continue to be competitors, and they will be more successful I think in this space.

But that doesn't necessarily mean it's more successful against us.

Because there are still quite a few players that we think don't have the right model long-term.

Terrific.

And then if I can ask a quick follow-up question on the commodity DRAM space and relative to NAND, the allocation of DRAM over to NAND's ability to kind of soak up some of the excess capacity coming online, if we fast forward -- if we go back to the beginning part of '05, I would say that we have had more DRAM allocated over to NAND than we thought.

We have had better PC unit growth than we thought and still the commodity DRAM pricing was kind of weaker than a lot of people expected.

Do you think that that could potentially be the same scenario for '06?

Outside of vista, is there a different dynamic there?

Mike, do you want to take that?

Well, we've got the -- outside of Vista, there are no once in a 3 or 4 year type of events that are occurring.

So we've got the Vista occurrence next year.

Other than that, we've just got typical unit growth rate and of course content persistent growth that would normally be associated with new applications coming to PC.

It increased communication bandwidth and so forth.

So, by the way, from a demand standpoint, we think those events are going to be sufficient to drive 50 percent-ish type of demand bit growth from the PC arena.

Really with respect to what that means in terms of the supply/demand balance, as you know, it is impossible to predict.

But the comforting factor for us is that for the first time really since the beginnings of the DRAM industry, we've got this other very, very large market -- that being the NAND Flash market -- that utilizes essentially the same tools as well as essentially the same kind of technology that is used for DRAM.

That gives us as DRAM manufacturers, the ability to turn that knob if we get to a point where we are in an oversupply situation on DRAM.

So I think that is a different dynamic really that it really has never existed in the DRAM industry today.

It will be interesting to see how it plays out.

Shawn Webster, JP Morgan.

This is Shawn calling for Chris.

You talked a little bit about your DDR2, the DDR2 pricing environment.

Can you expand a little bit into DDR and SDRAM?

And then, I have a follow-up.

Sure.

Just in general on the DRAM pricing front, the current environment is that DDR pricing is generally flat.

DDR2 pricing I mentioned; we're seeing some price pressure.

So I think the direction or the near-term direction on DDR2 is probably going to be price decreases.

Synchronous DRAM pricing is generally flat as well.

And our other DRAM products, just kind of the byproducts of the DRAM, would be a low-power mobile DRAM pseudo-static, and that's generally flattish as well.

So actually, in summary, everything generally flat with the exception of DDR2, where we are seeing some price pressure.

And on your wafers, I think last quarter you mentioned that you were at a wafer start rate of 60,000 wafers.

And you thought it would increase a little bit.

Can you talk about what your wafer outs or wafer starts were this quarter and where you expect them to be by the end of fiscal '06?

That is correct.

We hit our targets.

We're actually shipping a low 60,000 wafer outs per week, and that will be up a couple percent in fiscal Q1.

Can you talk about where it will be at the end of fiscal '06?

We will update you as we get into the next conference call.

Tim Luke, Lehman Brothers.

That's appropriate;

I'm calling from Korea.

I was just wondering, if you could have commented with respect to how you'd seen linearity through the quarter in general.

And then just remind us how that usually works as you move through the November quarter.

From a shipment standpoint?

Is that--?

Really, and also from an order development perspective too.

Well, we typically see strength around the end of our customers' fiscal periods in demand.

I do not know whether they are in the mode of deal cutting or what have you.

You'd have to ask them how it works.

But typically, we see demand strength or very, very small surge -- micro-surges is in demand, if you will, around the end of our quarters, our customers' fiscal periods -- which by the way don't coincide with our fiscal period.

So for example, in the quarter that we're in currently, September would typically be a strong period for our customers.

And I think that is occurring just about as expected.

I was also just wondering with respect to the CMOS image sensor area whether you could just clarify what the pricing outlooks had been.

Was that an area which is also expected to be fairly stable?

Or how have you seen that?

For us, the pricing on CMOS sensors has been quite stable.

I would add that our customers are telling us that the pricing environment, particularly on the VGA sensors and 1 megapixel sensors, the pricing environment is becoming quite more competitive.

My perception is that our superior image quality has enabled us to obtain a significantly higher price than the industry average.

And it's also enabling us to -- both to hold market shares, while maintaining a significantly higher price.

Lastly with respect to inventory, having gone down this quarter, do you feel that it is at a level where you would expect it to move up from here?

Or do you think there's further opportunities for it to move down based on what you see in terms of production and demand?

Generally feeling it is at a level necessary to support our customer base, other than what Mike indicated as some DDR2.

Mike?

Yes, I think that is accurate.

You know, it's really difficult to predict.

There's not much predictability in our customer demand from week to week.

So it would be really speculative for me to say what's going to happen to our inventory between now and the end of the quarter, let alone the end of next month.

But our inventory level today is at a level that we feel is necessary in order to provide adequate service to the customers.

I don't foresee anything usual on the horizon.

David Wong, A.G. Edwards.

First, just a quick clarification because I think you said you're expecting high-single digit bit production growth over the next couple quarters.

Do you mean per quarter?

Do you have single digit growth in each of the next couple of quarters as opposed to in total?

Yes, that is correct.

And sorry, can you guys quantify for us -- you did say that inventory had come down a bit today.

Can you give us the weeks of inventory ending the quarter?

In the range of a little over 3 weeks worth of finished goods inventory.

And my final question, average selling prices in the quarter firstly your PC DRAM and then overall?

And secondly, where does today's prices -- roughly how do those compare with your average prices through the quarter just finished?

I don't have the absolute selling price in the prior quarter.

I apologize.

Today's selling prices on DDR would be about the same.

On DDR2 may be slightly lower than where they were last quarter.

On synchronous DRAM about the same.

And if Bill has the absolute pricing --

That characterization is solid compared to the averages we picked up for the quarter.

So if I look at our average pricing, which I have in front of me now, our average pricing last quarter for example on a -- you know, we're high-volume parts, 256 megabyte DDR in the neighborhood of $3 per unit. 512 megabyte DDR in the neighborhood of $7 per unit.

Right, then it's about the same now you're saying?

Roughly, yes.

Krishna Shankar, JMP Securities.

Congratulations on a nice quarter.

Can you rank quarter your gross margins by product line, especially the DRAM, CMOS (indiscernible)?

Can you just give us the line total of your gross margins by product line?

Yes, it hasn't changed much from last quarter -- image sensor, specialty DRAM, Flash, and then PC DRAM.

And then in terms of the dramatic improvement in gross margin compared to revenue growth, would you characterize that just as NAND/Flash getting to sort of critical mass and covering all fixed costs?

What would be the key reason for the dramatic improvement in the gross margin percentage this quarter?

There is a significant amount of that that we hoped comes across with regards of the cost improvements we're making across our manufacturing lines.

Indicated that our 300 millimeter operations moved up significantly and are having very nice yields throughout manufacturing and improved noticeably -- and did acknowledge that there was significant improvement on the TECH Semiconductor gross margin as well.

So 20 to 25% of our product coming from that joint venture had noticeable improvement in margin.

So do you mean a stable, flattish pricing environment?

Would you say that 40 to 50% of incremental revenue growth could translate to the gross profit line here going forward over the next quarter or so?

Let me answer that in a way that we do have been in all of our product categories continued cost reductions really every quarter through the next year.

So in a flat pricing environment, then, yes, you would expect additional contribution.

And then on NAND Flash, why don't you accelerate capacity expansion?

Why don't you accelerate your capital spending, given that you have good success in ramping up your -- it's a high growth market.

And why don't we see that experiencing higher growth in terms of your production investment in NAND Flash?

Well, we actually -- one thing that we can't forget about is that we have pretty high growth in a couple of other areas outside of NAND that are not computing DRAM.

So in fact, we have had to allocate some of the resources that maybe that you would classify as being able to go to NAND to these other products.

Because you just heard from the ranking from Kipp on the gross margin side; we've actually still got two other products that are higher gross margins than the NAND, even though as you just described, the NAND environment is pretty good.

So we are not going to sacrifice that when we are also trying to make sure that we keep a leadership position in a couple of these other markets.

So to the extent that we can accelerate, we will.

But the other thing that you have to keep in mind is we do have a customer base that we have to service.

And we have the integrity of the Company and an obligation to the Company to make sure that we look out for them as well.

And as a result, we are not going to do anything too drastic.

We can make changes over time.

But we are going to continue to meet those commitments and make this transition as we are able to as we go through time.

And my final question on the competitive environment, it seems that most amount of capacity in sort of the commodity DRAM areas, which has led to the pricing weakness has come out of Taiwan.

Can you characterize the capital spending outlook and sort of the behavior of some of the Taiwan competitors, who seem to be adding the most of capacity in the commodity DRAM area?

Yes, I think the premise that you just outlined is true; that most of the capacity has -- it's either coming online or being forecasted to come online, really I'd say in some of these larger numbers, mid to late '06.

Now, we don't know if that materializes.

But typically, that is tied to how difficult the market is.

So I cannot speak for those companies.

They have announced plans to bring capacity online.

And we will just have to wait and see what actually happens.

Ben Lynch, Deutsche Bank.

Congrats on the particularly strong performance in the CMOS seamless sensor and NAND.

I guess over the last 2 quarters or so, cumulatively, this business has sort of doubled I think 45% and then 40% growth.

Could you just try and break that growth down in terms of the cell phone market, CMOS seamless sensor penetration, your market share.

I am just trying to deconstruct that growth a little bit if possible, please.

Ben, if I understand your question, you are referring specifically to our growth in the image sensor business.

You would like Mike maybe to detail the different market segments, the growth rates in those and then our position in that?

Yes, that would be great, please.

The mobile phone market, Ben, is continuing to be the primary driver of our business.

We have had some -- actually quite a bit of success in winning designs in the automotive business and in a consumer business, actually in the medical business as well.

But frankly speaking, those have not accounted for significant revenues to date.

So most of the revenue growth you're seeing is coming from mobile phones.

Some commercial success in PC camera products and in digital still camera products as well.

But the bulk of the revenue growth is coming from mobile phones.

And I believe the seeds that we're planting in the automotive industry are going to bear fruit, significant fruit for us probably in the '07 timeframe.

I think, Mike, you said this year, full year, maybe 50% penetration, so I guess it's possibly even above that at the moment.

And also if you could confirm that.

And also what do you estimate Micron's share in the current quarter, please?

We think there is about a one-in-three chance if you buy a camera phone, it's got a Micron sensor in it.

The other question I have is, it's difficult we're trying to construct a few different things from various elements.

You gave us the CMOS image sensor and Flash businesses have grown.

But is 15% the revenues?

I guess you were sort of hinting that.

CMOS image sensors is the bigger part of that.

Let's say it's 8, 9%, and it has doubled effectively in the last 2 quarters.

In those 2 quarters, your overall revenues have sort of been flattish.

So let's say a couple of quarters ago, it was 4, 4.5%.

I know it is clearly bigger than Flash back then.

Ahead (ph) of just notes from a year ago in Q4 that at the time, Flash and CMOS image sensor or other stuff as you called it was 10 to 15% of total wafer starts.

I guess I am trying to understand the dynamic between the wafer start allocation versus now you've finally got to a point where these two businesses seem to be a large part of your revenues.

Could you help us understand was that just sort of maybe particularly in NAND getting yields to acceptable levels?

And I know there is a bit of a lead lag between wafer starts and wafer outs, etc.

But if you could just help me understand a little bit of that please.

Hey, Kipp, let me jump in here real quick, and then you guys can add comments afterwards.

You know, the thing that we have tried to really communicate I think over the last couple of years is that as we grow these businesses, we have to keep in mind that the DRAM market, in particular the computing DRAM market is not standing still.

In other words, it's still growing at this 50% rate per year.

So what you are seeing is a discussion about revenues.

I think if you were to look in terms of output, clearly our output was shifting to the non-computing DRAM.

But that is being countered now by the ramp of the 300-millimeter facility.

So I actually had thought about this prior to the call in terms of how this would look in terms of break out.

Because on the surface of things, even though we are I think higher percentage going to these other product areas than we were as we talked about it a quarter ago or two or three quarters ago.

But because of the 300-millimeter output coming online and then re-contributing the bits going into the DRAM space, it somewhat balanced it out, even though these others have been growing quite rapidly for us.

So I think in terms of growth rate, there is no question that the growth rate in things like CMOS imagers have been outstripping everything to date.

And the reason I say to date is, remember, the Flash market of course is growing at a pretty high rate as well.

And really how that shakes out is still to be determined a little bit in terms of our resource allocation.

But both of those will grow at a higher rate going forward than you are going to see the DRAM computing is for an obvious reason.

Because we're reaching the levels in 300 millimeter that we said we would reach going back a year, 1.5 year ago.

And of course, the growth rate now in that space is going to slow now that we are reaching those levels.

And it will primarily be based upon yield improvements and process node transitions, etc.

And as a result, we probably will start to have somewhat of a re-acceleration now in the percentage of our business that is attributable to the CMOS imaging and NAND product line.

It sounds like there's probably a few things that I guess if you take them all together, yes, it can work out.

Thank you very much for the explanation.

Scott Stephens (ph), Ivory Capital.

Can you talk about just where you are in terms of your cost structure per NAND presently?

You talk about the quintupling of the revenue, but obviously the production has not quintupling.

So talk about kind of where you are in your cost structure and compared to competitors and where you will be kind of next year?

It is hard for us to know of course where we are at compared to competitors.

Let me just primarily characterize it in terms of where we sit today.

Absolutely unequivocally, our cost structure on NAND will continue to decline I think fairly significantly going forward for two reasons.

One because we are still relatively small in scale building the NAND today; although, we still think our cost structure is pretty decent.

But it doesn't account for that big of our output yet.

And in addition to that, we originally inserted ourselves in the 90 nm 2 gig, which was where the market was at the time we introduced it, and we thought it would be.

But we're sampling the 72 nm, and it actually looks pretty good as well.

And of course, that will translate into a cost reduction also.

So really at the end of the day, we have a lot of runway in front of us in our cost structure in that device.

And actually, we think it's still even pretty good today.

And again, I can speak where our competition is, but we are pretty confident we are going to make significant strides in where we sit.

Okay.

And then Kipp, can you talk about just -- you have obviously generated some free cash and talked about where you guys will get in terms of how you view it going forward.

And talk about what a good construct of return on invested capital might be for going forward?

And how you guys will -- how you think of return on invested capital, how we could calculate that and how that is connected to it to the free cash flow?

I can take the first part of that, and then maybe Bill can address how we might calculate ROIC.

But I think you're referring to the fact that Steve alluded to earlier that we have got several product lines that are growing very rapidly that we would consider to be much less capital intensive than say the advanced process needed for NAND Flash and DRAM.

So from that construct and in an environment where you have somewhat benign pricing, obviously, we believe our cash profile can look significantly different than it has over the last 5 years for example.

And Bill, would you like to take a swat at the ROIC calculation that perhaps we look at it internally?

Well, if you are just looking at it in a EBITDA reference, in the circumstance that for sometime period yet, we have got no tax rate to take into consideration.

And look at the combination of debt and equity as your denominator, you got a high-level view.

And as Kipp has indicated, for over a year, 1.5 year now, we have had a very, very significant emphasis within the Company, where individual resource allocation decisions are being driven by an ROIC analysis.

And we do believe that this quarter is a good indication that that is starting to pay off.

Bill Dezellem, Titan Capital Management.

We have a couple of questions.

First of all, relative to the cost improvements that you see in the next couple of quarters, are there any factors that could or should lead to more than the normal trending down of cost were you see a real step function increase?

And then secondarily, relative to your participation in the NAND market, where or how or naught does Leehigh (ph) fit into your capacity expansion plan for the NAND market?

This is Steve.

Let me start with the Leehigh question; then, I want turn it back to the other guys to comment on the other.

We clearly have kept Leehigh in our roadmap for expansion.

Because it's a 300-millimeter capable facility.

As you know, we have made a significant investment there, and we continue to keep that facility either in a great state or actually we been bringing it forward if you will through time.

And so that facility is available and ready at the time we decide that we need it.

Now, we have not made that decision today, obviously.

But we continue to review that on an ongoing basis.

(multiple speakers) If we were to choose to move forward way, would -- is there enough capital allocated in the budget in that 1 billion to 1.5 billion for fiscal '06 that would include that?

Or would that be (multiple speakers) --

Well, actually even if we were to move forward on it, remember the lead-time of the equipment, etc., it would not affect the fiscal '06 budget much anyhow.

And Bill, on the cost side, as you know, we generally give a guidance of expecting around a 10% cost reduction per quarter.

Certainly quarters like this, we can have even better than that type of performance.

In general, the items that will continue to drive costs for us will be improved 300 millimeter, in general continued improvement in yields and of course shrinks that we are working on all the time.

Andrew Root, Ode (ph) Asset Management.

I get easily confused, so I just wanted to clarify something. 45 to 55% bit growth for fiscal '06, is that total memory bit, so then you will then decide how to split it between DRAM and NAND?

That is correct.

So presumably, NAND grows faster.

DRAM grows at something slower than the midpoint of that?

And that is separate and apart from imaging unit growth.

And then the second part was on image sensors.

Did you say that you had wins with all the Tier-1 OEMs?

Or that you are already shipping in volume to all the tier OEMs?

We certainly have wins with all of them.

And if I'm not mistaken, we are shipping in volume to all of them.

If we are not as of today, we will be by the end of our fiscal quarter.

Nadeem Janmohamed, Lehman Brothers.

My question has been answered.

Great.

Thank you very much.

We'd like to thank everyone for participating on the call today.

If you will please bear with me, I need to repeat the Safe Harbor protection language.

During the course of this call, we may have made forward-looking statements regarding the Company and the industry.

These particular forward-looking statements and all other statements that may have been made on this call that are not historical facts are subject to the number of risks and uncertainties, and actual results may differ materially.

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

Thank you again for joining us.

Ladies and gentlemen, this does conclude today's teleconference.

We ask that you disconnect all lines.

And have a great day.