美光科技 (MU) 2011 Q2 法說會逐字稿

Good afternoon.

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

At this time, I would like to welcome everyone to the Micron Technology's second-quarter 2011 financial release conference call.

(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 and welcome to Micron Technology's second-quarter 2011 financial release conference call.

On the call today is Steve Appleton, Chairman and CEO; Ron Foster, Chief Financial Officer and Vice President of Finance; and Mark Adams, Vice President of Worldwide Sales.

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

If you have not had an opportunity to review the second-quarter 2011 financial press release, again it is available on our website at Micron.com.

Our call will be approximately 60 minutes in length.

There will be an audio replay of this call accessed by dialing 706-645-9291 with a confirmation code of 52960648.

This replay will run through Wednesday, March 30, 2011, at 530 PM Mountain time.

A webcast replay will be available on the Company's website until March 2012.

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

Please note the following Safe Harbor statement.

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

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

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

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

These certain factors can be found in the investor relations section of Micron's website.

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

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

And now, I'd like to turn the call over to Mr.

Steve Appleton.

Steve?

Thanks, Kipp.

For those of you that were not able to make the call during the last quarter, I just want to remind everybody that we have a slight format change.

And that is that I'm going to make some introductory comments, and then Ron Foster will cover some of the financial aspects of the release, and then we'll open it up for questions.

So, to start with, I'd like to open on the topic of Japan.

You know, I want to first comment we are thankful that none of our employees were injured during these tragic events, and we hope the best for the people of Japan as they start the recovery process.

But more specifically to the impact on Micron, I know people are trying to determine whether it's a positive or a negative for Micron, and unfortunately, I can't answer that for you, but I will comment on the things that we are evaluating.

First, our wafer fab operates more towards the center of Japan and away from the coast, and we did not sustain any damage, nor did we experience any disruption in our production there.

We do have a design center that's near Tsukuba, just outside of Tokyo, and as you might expect, we have some sales support offices in Tokyo, and they were both shut down for a limited amount of time.

I think the greatest challenge there is they're experiencing what I believe most people are in the area, which is difficult transportation and rolling blackouts, which I'm sure is frustrating to a lot of people.

Now, a number of our suppliers are in Japan, and they do range from chemicals to materials to wafers.

Many of them supply us from operations they have outside of Japan, but also as well from operations they have inside of Japan.

Some of them were not impacted and obviously some of them were.

Now, we have a pretty complex matrix analysis on everything, as you might imagine.

And for the most part, the companies have been very open, and they've been briefing us, really, daily or hourly, depending on what the circumstances are.

So, our information is obviously only as good as theirs, and I think many of them are still trying to determine themselves what the best path back -- what is the best path back to full production.

Now, also as you might expect, we're filling gaps on an hourly basis where we might have holes, and as been noted, I think, in the media, most of us have inventory for the near term, but really we still need a couple of weeks to understand what, if any, regain -- remaining gaps might exist.

So, we're working on it pretty diligently.

We don't have any other information than that.

And we think it's something that we have to stay pretty diligent on and stay on top of, but that's what we are doing.

Now, with respect to fiscal Q2, on operations and technology we had a number of achievements I think are worth noting.

Probably the one to highlight the most is that IMFS is coming along ahead of schedule by about two months.

It may be hard to believe, but we're actually shipping now product from that facility.

And I'll probably take this opportunity to mention capital.

So in our Q1 capital, the private quota was about $570 million.

We made note that this was going to be our big capital for the quarter -- our big quarter for capital.

And it was about $840 million, which is a little bit less than we thought, and it also happens occasionally sometimes things get pushed into the next quarter, just in terms of paying it.

So, Q3 I think will be similar.

But IMFS is going actually so well that we're going to continue to drive pretty hard on [bringing] the capital in there and getting that facility ramped, so I think we're going to be on the upper end of the range that we had given you.

We had previously told you $2.4 billion to $2.9 billion.

I think we'll be on the upper end of that range for the year.

But we felt pretty good about it because, as I said, the operations goal went pretty well, in particular IMFS.

In fact, I think it's also worth noting that 20 nm NAND still looks very good for a ramp in the second half of the year.

Also, what had -- we completed our acquisition of TECH Semiconductor.

We owned most of it.

We completed the acquisition of the rest of it.

So, we now own 100% of it, and it continues to operate pretty well.

Switching over to a few comments on markets and product segments, let me start out with the DRAM market.

I think on our last earnings call, I noted that we did not believe that a winter quarter was going to be similar to the winter of 2007 or 2008.

I think that's proven to be the case, and we're not suggesting that it was a picnic, as some of our competitors proved with reporting that they did with some pretty negative earnings.

But I think the downturn bottomed this last quarter.

Also, we mentioned that we thought it would be at a lower environment for pricing for a quarter or two, but it will begin to get worse.

In fact, it didn't get worse.

And at least from my point, it's proving at least to be somewhat mild.

You know, clearly I think that the new DRAM supply continues to be muted, compared to last cycle.

We also made mention of that, and that's proving to be the case.

It is true that desktop continues to be somewhat weak.

There are a lot of forecasts now that are pointing to something around 10% growth as opposed to some of the higher forecasts earlier, but I would remind everyone that content is still forecast to grow 40%, and if you do the combination, that's still up 50% or in some forecasts I've seen, even greater bit demand, so not too bad, really, for the weaker of the markets that we participate in, and all of the other markets actually look better.

So, on the pricing front, we don't think it's going to go crazy.

In other words, it's not going to go crazy up, but I will tell you that we do see a continued improving pricing environment through the rest of our fiscal year, and I think most of you have seen in the media that some of that's actually materializing now.

On the NAND front, also mentioned in December that we felt we had hit a level where the NAND pricing would be pretty stable.

Of course, it's fluctuated a little bit.

It's gone up a little, it's gone down a little bit.

For the most part, I think it's worthwhile saying that it has pretty much behaved as we would've expected it to, obviously not withstanding the recent spikes due to the Japan effect.

So, wireless, particularly in smartphones in terms of memory consumption, continues to look very good.

I want to also mention that our SSD shipments were up significantly Quarter Two over Quarter One, and we see that trend continuing with Quarter Three is going to be over Quarter Two.

And obviously, a lot of the news around tablets and those kinds of devices are actually pretty positive for us in the NAND arena.

So, all those things added up look like -- all the demand signals going forward look pretty good for both the NAND and the DRAM environment.

On the NOR front, the embedded markets, we have the business groups now, which Ron will just mention in a second.

But the embedded markets for NOR, DRAM, and NAND all look good.

And I think, as you would expect, as we tried to communicate earlier, for the most part we're pretty stable in pricing and demand.

I will say specifically to NOR, we believe we maintained our market share.

In other words, our market-share leadership in NOR.

Obviously, we saw some modest seasonal decline in fiscal Q2, but that's pretty typical.

And I think another highlight is that we sampled our 45 nm NOR products for wireless and embedded, and actually began our 45 nm NAND production ramp.

So, all of that is on a good note.

So, all of this leads me to my final comments, which execution on our product portfolio strategy is working.

We think we're in pretty good shape.

In fact, we think we're in as good a shape as we've ever been coming out of a negative period.

The demand signals from the majority of our customers are strengthening from Q2 to Q3, and I think we're pretty well positioned so that even with stable to moderately improved pricing, we should now start to build margin back into the financials.

So, all in all, we feel pretty good about where we sit.

And with that, I'll turn it over to Ron.

Thanks, Steve.

The Company's second quarter of fiscal 2011 ended on March 3.

As usual, we provide a schedule containing certain key results for the second quarter, as well as estimated metrics for the next quarter.

That material is presented on a few slides that follow, as well as on our website.

The results for the second quarter include a $40 million gain in other operating income associated with the Samsung patent cross-license we discussed last quarter.

A portion of this gain is reflected in the operating results for each of our reportable segments, as you will see later in the presentation.

Recall that the final installment of $35 million under this settlement agreement is expected in the third quarter.

The provision for income taxes in the second quarter is higher than what would normally be expected, primarily as a result of a tax rate change outside of the United States.

This change caused a $16 million one-time write-down of deferred tax assets, and, accordingly, is not expected to have a significant impact on the tax provision going forward.

In addition, the tax provision includes the withholding tax of $7 million related to the Samsung payment under the patent cross-license.

Total idle capacity costs, which are charged directly to cost of goods sold, were $36 million in the second quarter, compared to $59 million in the first quarter.

These costs are primarily attributable to the Singapore fab start-up.

In addition, preproduction costs at IMFS of $15 million were charged to cost of goods sold.

We anticipate that IMFS will begin to inventory production costs in the third quarter as the initial products are qualified for sales to customers.

NAND trade production costs per bit declined faster than NAND trade selling prices, while selling price reductions for DRAM products outpaced the cost reductions.

The result was a compression of the overall Company gross margin from 23% in the first quarter to 19% in the second quarter.

As you recall from our winter analyst day, we estimated DRAM production bit growth to average low teens over the next three to four quarters.

The second quarter actually came in above this range, at 23%.

The higher bit growth in the second quarter, along with a plan to shift product mix to higher-margin specialty DRAM products, drives our estimate of low to mid single digit bit production growth in the third quarter.

We also projected NAND production bit growth to average mid-teens over the next three to four quarters.

The second quarter actually came in above this range as well, at 20%.

We are shifting NAND mix to lower density and higher margin products, including SLC- and NAND-based multichip packages in the third quarter, leading to bit growth in the mid to high single digit range.

Correspondingly, the NAND trade cost per bit decline in the third quarter has moderated in the low to mid single digit range as we shift to these higher-margin products in our mix.

Bit production includes bits purchased from third-party suppliers, which may impact reported volumes and margins.

To give you an idea of the general composition of our product portfolio, we have developed this high-level summary.

As a percentage of consolidated net sales, DRAM comprised in the low 40% range, NAND in the mid-30s, and NOR in the mid to high teens as a percent of revenue.

This quarter, we begin reporting financial results for each of our four business unit segments.

The second quarter results for DRAM solutions group tracked closely with the DRAM product average selling price cost per bit and bit volumes I mentioned earlier.

This resulted in revenue and operating income somewhat lower than prior periods.

The DSG segment includes virtually all of the production volume from Inotera, which was relatively higher cost in the second quarter compared to other DRAM fabs as their production volume ramps.

NAND solutions group trade sales tracked closely to the NAND trends presented earlier, with revenue up quarter to quarter.

NSG sales to Intel from our IM Flash joint venture were just over $200 million in the second quarter, roughly flat compared to the first quarter.

Recall these sales are at long-term negotiated prices approximating cost.

Wireless solutions group sales of products by architecture in the second quarter were NOR, NAND, and DRAM in increasing -- decreasing order of revenue.

That's NOR, then NAND, then DRAM in decreasing order of revenue.

A portion of NAND sold in WSG is supplied from Hynix at market prices.

This supply is transitioning to Micron factories in the coming quarters, which we expect to improve WSG margin going forward.

The embedded solutions group sales of products by architecture in the second quarter were, in rank order, NOR, DRAM, and NAND in decreasing order of revenue.

Most of the inventory that was marked to market in the Numonyx acquisition has now flowed through.

In addition, similar to my comments relating to WSG, ESG also acquired a portion of their NAND products from Hynix and are also in the process of transitioning to Micron manufactured products.

Now turning to operating expenses, R&D expense for the quarter benefited from development cost sharing of approximately $56 million from joint development programs.

R&D expense, as well as the cost sharing, was reduced by the reimbursement of certain costs incurred under government programs in the second quarter.

R&D expense is expected to increase in the third quarter to the $210 million to $220 million range, primarily due to a higher volume of prequalification wafers processed, notably 42 and 3X nm DRAM and 20 nm NAND.

The Company generated $809 million in cash flow from operating activities in the second quarter.

As previously reported, in the second quarter we spent $159 million to buy out the other shareholders of TECH Semiconductor, as Steve mentioned.

The accounting for this acquisition resulted in a $67 million increase to additional capital in our equity section.

The cash and short-term investments balance at the end of the second quarter was $2.2 billion.

In addition, we have restricted cash of $338 million separately presented on the balance sheet.

Now I'll close and turn it back to Kipp.

Thanks, Ron, and with that, we'd now like to take questions from callers.

Just a reminder.

If you are using a speakerphone, please pick up the handset when asking a question so we can hear you clearly.

With that, please open up the line.

(Operator Instructions).

Glen Yeung, Citi.

Steve, in your prepared comments you said you were seeing demand signals improving, and I wonder if you could just give us a little bit more detail about that, one, and then, two, let us know if you've contemplated the impact of Japan in those statements.

Well, let me first talk about Japan for a second, and Mark Adams is here so I might as well let him answer the question on the demand signals.

We haven't contemplated, in thinking about Japan, whether that helps us or hurts us because, obviously, there -- a competitor or two of ours, no doubt, has some supply issues related to that because they're in the area where it was impacted.

Then, of course, you already know the counter to that, which is if customers can't get some components, then they are obviously be challenged in needing other components.

So, that's hard for us to know.

I remember -- I don't know if you remember, but probably not, the Sumitomo chemical plant explosion, the mold compound which was 90% supplied out of that part of the world.

It's probably worth noting, I think, that what we saw in that case and probably what we'll see in this case is that most of the supply chain, most manufacturing operations, have the ability to have an accordion effect, so in other words, we can run at near full capacity, but run a lot lower cycle time.

And so, you can expand and contract your production to try to deal with shortages that come through at various times.

So I think you'll probably see that through most of the manufacturing world where they will expand and contract, and they'll run faster cycle times or shorter cycle times rather than replenishing and trying to squeeze more product down the line.

For the most part, I think if the effects are relatively short-term, in other words a couple of months, a few months, then you probably won't see a whole lot of anything in terms of what happens in the manufacturing operations around the world.

And if the effects are longer term, we clearly are going to have some impact.

On those kinds of things, we just don't know.

And our information is only as good as probably yours, how that might impact our customers.

And let me ask Adams to comment on the general demand signals from the [rest of the world].

And my comments will be obviously pre-Japan, just as Steve addressed Japan.

What we saw coming out of Christmas, which again normally is our weakest demand quarter, we saw pretty stable demand throughout the quarter, and in fact, post-Chinese New Year, we saw kind of an uptick across a number of our key volume segments.

The PC business, in general, was better than expected, and we're also seeing density inside the box increase back to even about where we were back in the [de-speccing] phase of 2010 in the spring and summer of that.

Our density per unit is back up, and so demand in the PC commodity space is pretty strong for us.

The server market, as well, from Micron was very healthy.

As a matter of fact, the percentage of overall specialty bits against commodity, it was up significantly this quarter for Micron.

So, pretty strong demand there, and I would also point out that though server growth overall is not -- in the 6% to 7% range year over year in terms of unit growth, the density per server is dramatically up.

It's around 50% to 60% per box.

So, good growth there and good demand from our customers in that area.

Obviously, the Numonyx acquisition has opened additional customer opportunities around the wireless segment for us, and they're certainly manifesting themselves both across NOR applications and then DRAM, MCP solutions with NOR and NAND solutions.

So, I see pretty good demand through the quarter, which enabled us to form pretty well at the topline.

That's great.

And maybe just a quick follow-up because, Steve, I wanted to clarify something you said earlier.

With respect to your ability to manage your production in light of an issue, I think you said maybe a couple of months you could handle it.

I just want to be clear on -- just some precision there.

Is it something like about eight weeks where you say like, all right, I can manage the fabs at this point for about eight weeks and I've got to start thinking a little more carefully after that about shortages, or do you have a number like that in mind?

Let me first comment that precision in this environment is just unachievable, number one.

Number two, what I was referencing was that -- actually, you made a comment that probably it would be good to add some additional input on.

Number one, most of the companies, us and others, carry a couple of months of inventory for most of our processes and products and so forth.

So we all still have that.

So, I think the reference point that for a couple of months it's unlikely most companies see much impact is probably true, at least in the semi industry.

I can't speak for other industries, but at least in ours.

But secondly, even when you hit shortages at some point in time of supplies, you're able to dial your manufacturing operations to actually just run faster cycle times.

Because the way most semiconductor fabs operate, at least in the [mini] world, is they're at full capacity because they want maximum utilization of the capital.

And the way that we achieve that is we make sure that every piece of equipment has material waiting for it to run.

And what happens when you're starting a wafer fab, by the way, and we're experiencing IMFS, when you start up a wafer fab that doesn't have that much material in mind or when you run into a situation where maybe you have some kind of shortage on something that you were expecting to get, but you actually just speed the cycle times up in those fabs.

So the output isn't quite optimized, but it's not like you take the same percentage hit as the shortage of material, and that can last for some period of time.

And then, it's only after that that you then have to start dealing with true shortages.

And of course, we have no idea whether that's going to exist or not.

Doug Freedman, Gleacher & Company.

Ron, would it be possible for us to get the rank of gross margins -- DRAM, NAND, what was the highest?

And then, if I could focus in a little bit on it, it sounds like you're changing strategy there on NAND, looking to go to some higher-value product.

Can you give us the gross margin effects that that should have?

Yes.

Rough -- I'll give you a rough ordering.

Specialty DRAM always tends to be at the head of our list.

Trade NAND is doing quite well for us and probably next in rank.

NOR clearly, and the vast majority of our NOR products, and then core DRAM in sort of that rank order.

Just elaborating a little bit, though, on your line of questioning.

We are seeing market opportunities, as Mark mentioned in his comments, to shift to higher-margin products in the specialty end of things, both on the DRAM side and on our NAND product side in the third quarter, and we are opportunistically shifting some of our mix which is affecting our bit growth and cost per bit computations that we typically give you.

But I just wanted to make sure you understood that we're doing this strategically to help improve margin.

And it's the right thing, given the balance of mix we need for our customers.

Could you help walk us through a little bit of -- you mentioned pretty quickly there in your prepared remarks about the accounting treatment for this startup of Singapore and your startup costs.

Can you walk us through in a little bit more detail and offer some color on how we should think about those startup costs accounting playing out over the next couple of quarters?

Sure.

There's a couple of pieces this quarter.

There is true idle costs, which were about $36 million in the quarter for the Company.

The vast majority of that was IMFS.

And then, there's another $15 million which we characterize as pre-production startup costs, and that's resulting from the fact that we're now actually starting wafers, as Steve mentioned, and beginning to move up the production curve.

So, both those pieces are expensed directly to cost of sales, and so you can compare them to the prior quarter's idle costs that were about $59 million.

But bear in mind that those -- that $15 million of costs relates to variable materials, et cetera, that are used to start our production ramp in IMFS.

So, what's going to happen here beginning in the third quarter is, as we qualify production, which we expect to do, as I mentioned, in the third quarter or beginning of third quarter, we will start inventorying some of those costs, and they will go into our inventory on our balance sheet and they will move out when we ship the qualified products.

So we will not be directly expensing all of the costs as we begin to qualify product, which I would view as beginning to happen in the third quarter and then significantly in our fiscal fourth quarter.

But just so that I understand correctly, so then when that inventory cost flows out of the operating model, it will be at a lower than corporate average gross margin or normal gross margin?

Gross margin presumes an average selling price, so I need to just address it on the cost side.

We will be -- we use a moving average cost system, if you will, so we'll be -- and we're already beginning to ramp costs.

So we'll be moving it into our inventory at an average cost trended over time, if you will, and that will be in our inventory cost structure.

We do not tend to inventory the full cost when you have a significantly higher ramp cost.

We inventory it at a somewhat lower level.

So, (multiple speakers) -- as that goes out, it will depend upon the average selling price, but it's -- we would expect to have a typical relationship to our normal flow of business.

Shawn Webster, Macquarie.

I guess, first of all, on end demand going into Q3, can you give us a little color on what you're hearing from your OEMs in terms of what you're expecting for calendar Q2 bit demand for DRAM and maybe even which end markets you're seeing the most strength in going into fiscal Q3?

From a demand standpoint, I think what we're seeing quarter over quarter is in the mid teens, more or less, and you know, we see no reason, per Steve's earlier comments, that there will be a lot of variability around that number in general, and then, the NAND market continues to look pretty strong around, obviously, the smartphone market, the tablet segment, as well as SSDs.

So, mid-teens DRAM bit demand growth.

Is there any particular end markets, whether it's networking, server, core PC, that's the strongest for you?

I think for us, again this is all kind of on a proportion basis.

The server/networking market showed significant strength over at the end of our Q2 and have continued to show strength going into Q3.

So we're -- continue to be very bullish on those markets, and as I mentioned earlier, despite the fact that the PC numbers are down off of maybe last year's 2011 projection, the density per box is making up for that loss, and we see -- so PC business is still fairly robust for us as well.

Okay, great.

And then, do you have -- can you share with us your expectations for cost per bit reduction in Q3 for NAND and DRAM?

Yes, Shawn, we're looking at down high single digits on DRAM and down low to mid single digits on NAND.

One final one, in terms of the production.

So you had a really strong quarter in Q2 on your production.

You're expecting that to slow down in Q3.

Can you flesh out what you think the next couple of quarters are directionally, excluding the Japan impact, on what you are expecting for sequential bit production?

Not much impact from our Japanese production facilities, and we'd be back to that guidance that Mark set at the analyst meeting, which is kind of low double digits in -- for over the next three to four quarters.

On the DRAM side?

That's correct, and about mid-teens on the NAND.

Kate Kotlarsky, Goldman Sachs.

I wanted to ask a question on content per box.

I think Mark talked about content per box starting to increase nicely.

And I wanted to get your thoughts on how you were thinking about content growth for the remainder of the year, given that pricing is now stabilizing and potentially moving up.

Do you think that might actually have an adverse impact on content growth?

Thank you.

Kate, this is Mark Adams.

I think that's unlikely.

We're not talking about the dramatic uptick in price that we saw back in 2009, going into 2010.

Remember something.

As our cost in DRAM goes down, even as we think pricing will be more favorable in DRAM over the next coming quarters, we don't think the impact on the bill of materials is anywhere as significant as it was back to $2.50 per gigabit at the OEM.

So we don't anticipate any downward trend in density per unit going throughout the rest of our fiscal year.

Then maybe just one other question on IMFS.

You talked about the ramp progressing faster than you had expected.

Are you guys still targeting that 60,000, 65,000 wafer starts per month by the end of the year or has that number now gone up?

No.

We're still targeting that.

Actually, even if we decide to go beyond that, it's unlikely we could have it happen before the end of the year because just the ordering cycle time and the equipment.

So, that's a good target to use for now, and as we've noted before, we're just going to watch the markets as we go through time.

But we're trying to get to that 60K as the first benchmark.

And then, just a final question.

Can you remind us when the first quarter is when your output share of IMSF is aligned with your ownership share?

Because I know there is a 12-month lag there.

Kate, this is Ron.

It depends upon the timing of capital injections, which happen every several months.

So -- and there is an approximately 12-month lag effect per the agreement in terms of the share changing.

So here this summer, we'll have the effect of the first capital calls where our partner didn't fully participate, and it will sequence from there.

But right now, we're at about 78% ownership and our output share is about 53%, and that will ratchet up in significant steps here starting in the summer.

Daniel Berenbaum, Auriga USA.

Can you talk a little bit about maybe guidance in the wireless and embedded segment that I'm just sort of parsing through the data sheet here, and can you then talk about -- the wireless profitability obviously seems fairly low.

Can you talk about how that business trends, both in terms of revenue and profitability, over the course of maybe the next year, 1.5 years or so relative to the embedded group.

We're going to stay away from any guidance at the BU level.

If you'd like Mark Adams to talk a little bit about the wireless segment in general, we're happy to do that.

But we're going to stay away from any projections and guidance on the BU level.

I might just add there's a couple of things I mentioned in my comments, and that is that we are still flowing through higher cost purchase accounting inventory which is significantly hitting WSG, and I mentioned that's significantly through as of this quarter end, and now it's going to tend to be a longer trend line and not have as big an effect on our financials.

So, there was an impact on both WSG and BSG businesses in terms of the purchase accounting effects.

I also mentioned that there is Hynix material volume that is being purchased at market price, and we're in the process of transitioning over some of that NAND volume, and as that happens in the coming quarters, it will improve the margin, everything else being equal.

So those are two things to keep in mind that are currently impacting that will change here in the next coming quarters.

Okay, and then, maybe just to follow up in the other segment, is that basically all imaging or is there other -- is there another component to the other segment?

That's things that aren't of significant enough size to warrant a segment treatment by the normal tests.

But it does include Aptina Imaging, display, solar activities, that sort of thing.

John Pitzer, Credit Suisse.

Steve, getting back to the issue of Japan, is there a concern that memory doesn't become sort of the bottleneck component of the system, like -- it's something else.

In that situation, do we have to worry about inventory being built in memory in the calendar second quarter, which will have an adverse impact on pricing?

How should we think about that?

I didn't hear the first part of your -- can you repeat the first part of the question?

What would have an adverse impact?

Let's assume that memory isn't the gating factor for getting systems shipped out the door; it's another component.

Is there a risk that we see an increase in memory inventory for the industry?

How do we think about that dynamic?

Well, it's -- that's a tough one to gauge because, clearly, there's going to be impact on supply of memory, I think.

If you look at where some of the large fabs are, despite the fact that some of them didn't go down hard, those of us in the memory business know that all you have to do is have a couple of hours of a problem and it takes you several weeks to recover.

So, there's going to be an impact on the supply side, and to the extent that there is something that happens that impacts it on the demand side, which I noted in my opening comments, that's just impossible for us to predict.

We just don't know how that sorts out, and the pluses and minuses on it.

And actually, I don't think it's -- as I mentioned earlier, it's probably not a phenomenon in the next 30 to 60 days.

It's something after that, if in fact there even is an effect.

And then, I guess, Ron, on the operating margin side by business unit, where does the Samsung settlement kind of get accounted for in those buckets?

Yes, it's actually spread because it's a broad-based license among all the businesses.

And then, guys, my last question just on DRAM mix.

Can you help me understand PC, server, and mobile in the quarter just reported?

And, I guess, when do you expect to be able to ship server-ready DRAM from Inotera and help me understand how that might impact the mix if servers starts to grow?

We are looking at qualifying the Inotera output sometime in the second half of the year.

In terms of a breakout of sales, we are running around 25% to 30% of our bits in PC, around 25%.

Maybe a little bit lower last quarter.

It should ramp up a little bit this quarter, but in the mid-20%.

Server, we generally run about twice our market share.

So if we're running 15% to 18% worldwide market share, we're generally running in the 30s in terms of server market share worldwide.

And networking, we do about the same thing.

We also outrun our market share in those markets also.

And Kipp, anything meaningful in mobile DRAM yet?

Coming out of Inotera?

Not out of Inotera, out of Micron itself.

Is it a big enough percentage of the DRAM business to start breaking out?

They just began running mobile wafer starts this month at Inotera.

Perfect.

Thanks, guys.

Vijay Rakesh, Sterne, Agee.

Just trying to figure out when you look at Japan, going back to that, how much of the wafer supply on your DRAM, NAND, or NOR comes out of Japan?

We have pretty minimal exposure to wafer supply specifically out of Japan.

We carry about five different suppliers.

Different competitors that we have do have a much larger, as Steve mentioned in his opening comments, exposure to Japanese facilities.

Yes, even the supply that we have from one of the larger Japanese guys actually comes from a U.S.

plant.

Got it.

Okay.

And on the NAND side, you talked about the Singapore fab.

How far is it -- how much is it prioritized now?

And the second part is, you mentioned 65,000 wafers additive incremental to the second half.

If it is at SLC, that obviously will be much lighter on the total bit growth that you're seeing, right?

I couldn't -- you're breaking up a little bit.

We're not going to give you the exact number of outs today, but suffice it to say we're about two months ahead of what we thought the ramp schedule would be.

And if you'd like to repeat the second half of that, we'll cover it.

I'm just wondering, if we assume 65,000 wafer starts incremental from the Singapore fab, if it's -- a lot of it is SLC, the bit growth probably is not as strong overall, right?

Are you asking the relation between SLC and MLC?

SLC and your wafer starts.

If it's all MLC, 65,000 wafer starts, that will be a lot of bit growth, but if it's SLC, probably not as much bit growth, right?

That's correct.

You have the right ratio.

MLC would certainly increase that, as would TLC.

Great.

I'm just trying to figure how the supply on the NAND side would look in the second half, so.

It will be increasing for us.

Uche Orji, UBS.

Let me just start off by asking you about SSD.

I think in your prepared remarks, you talked about SSD growing really fast.

The main -- from a consumer standpoint, I'm not quite sure I know -- outside of Apple, I don't see many SSD-based notebooks in the marketplace.

But can you just talk about what's been driving the demand there and how sustainable that is?

I mean, we had all expected that NAND price -- NAND flash prices need to go down significantly to drive SSD demand, but that seems to be happening.

So any comment as to what's driving that, given how resilient NAND flash bytes have been, would just be helpful.

Thank you.

I think one underlying piece of the SSD solution was the underlying controller development that had to go on over the last couple of years to get that market going.

So, we talk a lot about price per gigabyte and the value proposition to the consumer.

In fact, the notebook segment is still driving that growth, by and large, although enterprise is starting to play out pretty nicely, as I said, because of the technology and development around intelligent controllers and firmware development around making these things more reliable from an enterprise requirements standpoint.

So, the markets are pretty strong.

The OEM segments are good.

I would also suggest that the aftermarket integrator and VAR market has proven to be very successful for us so far, through the Lexar and Crucial channels.

So, it's a combination of some OEM desktop preconfigured machines, as well as some aftermarket that have combined for some pretty impressive growth, at least from what we see internally as well as the upside forecast for the back half of the year.

I think enterprise is now becoming more real in terms of the opportunity for us as an industry to put these devices in the backend of major companies and have the reliability and performance benefits that we have counted all along.

That's helpful.

And from what you're seeing, obviously that seems to be a trend that will continue through the rest of the year.

That's how one will read all these trends you just mentioned?

Yes, I think -- we're very optimistic about the market potential that we've seen so far, both in the OEM segment and the channel, so, yes, we don't see any reason that would stop.

Actually, again, I think the reason we're bullish on the enterprise, I think the enterprise game is just starting because the technology around the flash itself, which Micron has invested in over the last couple of years, is mature enough to bring to the enterprise.

When you were making comments about NAND in your prepared remarks, you seemed very positive in terms of how you expect pricing to develop through the rest of the year.

I mean, embedded in that, obviously, is that you have thoughts on supply from the rest of the industry.

So any comments you can make as to what you think the supply/demand dynamics are for NAND flash through the rest of this year?

And also, within that, you talked about tablets a lot.

There are concerns about the non-Apple iPad tablets not selling through.

If that were to be really the case, does that affect your expectation for NAND price development through the rest of this year?

I think how we look at the NAND market is that you've really got three pretty significant application drivers lining up at once, and that's why, I think, in general the market has been bullish on NAND going into 2011.

We haven't changed our position because, in fact, when you break it down the smartphone market is still driving significant NAND demand.

The SSD market is really playing out pretty nicely this year.

And we feel bullish as it does, to my earlier comments.

And then, the tablet market really has been a very sudden surprise from a NAND perspective year over year when you think about where it was just 12 months ago.

So, when you line that up, given the NAND supply picture for 2011, we're still bullish on NAND in general.

I think you asked about -- the last part, if I understood correctly, was with regards to if tablets don't manifest themselves to be as strong as they look right now.

Of course, that potentially could impact demand, but we think NAND is in a pretty good place.

It would have to be pretty dramatic.

And remember, of the three categories, the tablets would be the lower of the consuming categories.

Sure.

And just lastly, if prices were to hold steady from where they are now, what will be the change to ASPs for NAND, DRAM, and NOR by the end of this quarter?

We have flat DRAM quarter to date and we have NAND up a couple percent.

Kevin Cassidy, Stifel Nicolaus.

Thanks for taking my question.

On your strategy for moving to more specialty DRAM and also to higher-margin NAND flash, how quickly can that reverse?

Is this a permanent change or -- I guess, maybe, it's along the lines of questions that you always had about converting from NAND to DRAM.

Can you do the same with the specialty going back to core DRAM?

This is Mark again.

Steve in prior calls has mentioned that the NAND business looks like the DRAM business from early generation of the DRAM and the market behavior, and in fact, I think we're seeing that.

I think -- if you look at what I would call generation one of NAND, it was primarily photography, USB, with a little MP3.

Now when you think about it, NAND is the effective storage platform for smartphones, for tablets, and moving into notebooks and even enterprise storage.

So the maturity of NAND and the role it's playing in storage will allow us to treat that in terms of more specialty value add that Micron can bring in addition to the silicon itself, and so I think that's been a good thing.

You can see -- we feel pretty confident that our mix will continue to improve over time.

We've talked a little bit about why we've stayed so focused on two-bit technology in addition to our leadership position around the 25 nm offering and why we stayed with two bit.

Because of performance and all the specifications we are trying to drive with our customer applications, we continue to believe that regardless of how we transition, and we are transitioning to three bit, the value add is around the NAND, not just the NAND itself.

So, that will continue for Micron.

On the DRAM side of the business, it's really one around the opportunities we've cultivated in the past.

Very strong server share, very strong networking share, and those are two good ones that continue to grow nicely for us from a density per unit and, really, just overall units out the door.

So, we're positioned well in those segments, and I would continue to think those are opportunities for us to grow.

Maybe if I could turn to Inotera, is their supply chain very similar to your internal supply chain, or do they have exposure to Japan also?

Well, in some ways I think they are similar in areas, but they actually, I think, have less exposure to Japan than we do, just because traditionally we have had good relationships with Japanese companies over several decades.

And they, as I said, they actually have less reliance upon Japanese suppliers than Micron does.

Great.

Thanks.

Congratulations.

Manish Goyal, CREF Investments.

I have a question on mobile DRAMs.

Can you maybe give some color where you see your internal capacity or your wafer allocation going forward in mobile DRAM, and specifically what densities you are targeting?

Manish, we're not going to give out any specifics.

We certainly don't want to give our strategy out to competitors who are certainly most listening to the call.

So, if you'd like to answer -- ask a different question, feel free.

Perhaps, can you give some color as to how much mobile DRAM do you procure from the open market today?

How much what from the open market?

Manish, you are referencing the Hynix relationship?

Is that the question?

Yes.

Yes, because that's -- they're trying to match that up traditionally with what the Numonyx business was.

I actually don't know right off hand myself.

We buy a little bit of outside DRAM that goes into mobile applications, but that's -- we're transitioning out of that, as I mentioned, along with the NAND, over the next several quarters.

It's not a lot.

So, I'm just trying to understand that over the last several quarters or years, you've talked a lot about mobile DRAM, but we've not really seen a whole lot of progress.

So, where and when is the inflection point and what will traverse that?

I think -- a couple of things to point out.

One is that mobile DRAM, in terms of the consumption of it, has obviously predominantly been, and is occurring, in terms of densities in the smartphone arena.

And if you look at what's happened historically, it hasn't been that long ago since they started converting to a NAND DRAM combination going into the MCP, which has been addressing the higher density needs in the wireless space itself.

Previous to that, it was either relatively low density, and by the way, we are one of the largest pseudo-static RAM suppliers in the world, which is really just a DRAM with a static [front] interface.

We are one of largest suppliers in the world and have been for a long time, and almost all that product was going into the wireless space.

So, I guess I would characterize it a little bit differently than you did.

I think what you're specifically referencing is what they characterize as kind of the low-power, high-density DRAM stuff.

And it hasn't been in the wireless space in large amounts for all that long a time.

Betsy Van Hees, Wedbush.

First and foremost, congratulations on the quarter.

I think this is kind of an historic moment for you guys.

Isn't this the first time that you've ever been profitable during a DRAM downturn?

Well, it's -- yes.

First of all, thanks.

Actually, we were able to basically break even in 1990, 1991 in a mild downturn that occurred then.

That's the last time that it's happened, and even then, we weren't really profitable at the time.

We were just breaking even, but we stayed positive, and that's the first time it's happened.

And I think the other thing that's, I guess, worth noting is that clearly we just came out -- we think we're bottoming the downturn and we're able to maintain -- I guess this is our sixth quarter of [our own] profitability, and we think it improves from here throughout the year.

When we look at DRAM, I would imagine, with the gross margins around 19%, that that was probably pretty much the significant drag on margins.

And as we look at the mix going forward, how can we look at -- I know you guys don't like to give any type of guidance in terms of gross margins, but how can we look at that in terms of getting back to maybe where you were in fiscal Q2 of last year?

How is that going to help us?

How can you help us in our modeling?

A couple of things.

One, remember that along the lines of what Mark said and we said earlier is that we won't need rising pricing in order to build margin in.

We're going to continue to drive down our costs per quarter, so even flat pricing, our margin will expand from where it's at.

Secondly, I don't think there's any question that, as Mark Adams already noted, that the PC space in terms of the margin, as you already noted and highlighted, was a drag on us in our margin.

And we actually -- I think we're in an improving environment in that space, too.

So, that combined with I think we would also acknowledge that we haven't fully optimized the DRAM operations around the world yet, and there is still is some progress to be made there.

I guess the way I'd characterize it is there's some runway left there in addition to our normal cost reduction.

So, when you add all those up, I guess I think that's how you ought to think about it in terms of modeling what's going forward.

Thanks, that was helpful.

Then I want to take another run at the mix of MLC versus SLC.

If we look at your production today, can you give us the percentage that is SLC, and as you exit the calendar year, what do you expect the percent of SLC mix to be?

I'll tell you what we are today, but we probably won't give you the exit because we will continue to optimize for the market.

We run about 95% MLC today, about 5% SLC.

Okay, thank you very much, and once again, congratulations.

Bobby Gujavarty, Deutsche Bank.

I was wondering if one reason for your confidence in content for box growth, is it really enterprise clients?

Do you guys see that refresh continuing, and I think that's a higher content per box than a consumer box.

Could you comment on that?

Yes, I think you're spot on.

I think we think of the two -- the enterprise refresh as definitely driving that growth and the overall market condition.

Great.

And just one thing, if you can help me make sense of this.

I noticed in your breakout sheet, you had trade NAND ASPs were down 12, but in the press release, you mentioned that NAND ASPs were down four.

I assumed your trade NAND always has a better ASP performance than your overall NAND.

Maybe I misunderstood that.

Could you help explain the discrepancy?

This is Ron.

The trade NAND ASPs are, in fact, the movement of our markets in the NAND marketplace, and the total includes our at-cost shipments to Intel, which are moving with our cost-down movement relative to the products that we ship to them.

So when you look at the total, you've got the combination of those two movements going on.

That's why we're also focused going forward on giving you a trade NAND view of what's happening with our trade market, and also the cost per bit related to our trade NAND production, because there is a significant change -- difference in terms of our mix of products versus Intel, for example.

We mentioned that we have significant SLC mix we're shifting to in the third quarter.

We've got MCP volumes, which also report in our NAND sales volumes, and so, all that factors into the trade ASP numbers we are giving you for Micron, and the average is the mix of those two.

It's the cost on Intel shipments and the trade pricing on Micron shipments.

Alex Gauna, JMP Securities.

Thanks for taking my question.

I believe it already got asked, but it wasn't clear to me.

In terms of you moving more to SLC, how large is that addressable market?

How well can it absorb your increased supply and how is it that we've seen the pricing move with you guys at only about 5% of production in SLC that you weren't there for it in terms of these past quarters and the run-up into it?

I think the question around SLC for us is in those certain high-performing markets that really warrant both the opportunity costs of making the product around the performance we're be asked to hit in terms of specifications from our customers, so markets like high-end enterprise SSD performance, we'll be looking at how SSD -- that architecture -- I mean, sorry, SLC plays in that architecture density in a combination with an SLC/MLC architecture.

Those types of things we're evaluating from a technology standpoint, so it's really basically the opportunity costs to manufacture it versus what we can garner in the market to justify that production switch.

Are you aware of any of your competition making similar moves or should we feel pretty good about the ability for this market to absorb your moves?

Yes, we see other players looking at it because, again, when you look at the enterprise storage market, this is not a glorified USB.

This is something that's very sophisticated and much different than the pure NAND players understand.

And so, why we haven't been so big on the three-bit per cell campaign up until today is because performance, reliability, and endurance around the enterprise sector is going to make a winner from a solution standpoint, and we're very focused on not just the NAND, as I said earlier, but all the technology that goes around it in making a world-class product, and that's going to be more on the SLC type of architecture or high-performing controllers with a combination thereof.

And do you think -- just one last one on this topic.

Do you think that you can get those SLC gross margins towards what you're experiencing in the specialty DRAM at the top of the stack?

Especially in enterprise.

If you look at -- Alex, and you probably know this better than I do, but if you look at the projections for SSDs in general, while the revenue is very heavily weighted towards the desktop client segment, the overall projected margin profitability by analysts suggest that the enterprise margin is going to be dramatically higher and almost get to a 50% of the overall category margin.

I think we have time for one more question.

Win Cramer, Avian Securities.

Just one quick question on DRAM production.

Can you talk about as a percentage what is produced at 5X nanometer and 4X nanometer?

We're just looking at the data to make sure we're accurate for you.

This is Ron.

Looks like at 5X nanometer, we're running in the neighborhood of 85%.

Okay, but can you talk about the transition schedule -- the 4X and possibly 3X as well?

Yes, we're -- obviously, we're in transition to 4X now.

It's running in production.

I think we would expect probably a majority of that could crossover towards the end of this year, sometimes second half of the year would be the majority, and then we also happen to be currently running 3X.

But there is association with that with qualifications and getting it throughout our wafer fab network, and clearly, that wouldn't occur in terms of large volumes probably until we get into next year.

Is that end of the fiscal year or calendar year?

With respect to what piece?

The 4X.

You said you'll be there by the end of the year.

Yes, fiscal year.

Thank you.

Then my last question for you, you hinted towards the fact that you are making progress with TLC.

Is that going to become more of a staple of your business or still two-bit MLC all the way?

Yes, I think, as I was suggesting earlier, the challenge for NAND manufacturers in general is how do you keep the endurance and performance at an acceptable level for the application of where the parts are going?

Historically, TLC has been pretty low-end performance, cycling well below two bit per cell.

And so, we'll have to [true] it in the industry, and then again, I mentioned again that Micron has invested pretty heavily around controller and firmware development to enable this, is that that has to become more of an intelligent controller for things like error correction that allows us to gain the performance levels needed for things like SSD and things like smartphones and the mobile wireless market in general.

So, we believe that our roadmap will continue to evolve, utilizing TLC as we've grown in this area, but we've only believed that we're getting there today.

We didn't think we missed a big market opportunity in the past.

You bet.

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

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

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

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

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

Thank you.

Thank you.

This concludes today's Micron Technologies second-quarter 2011 financial release conference call.

You may now disconnect.