Micron Technology, Inc. (MU) Q4 2012 Earnings Report, Transcript and Summary

Good afternoon. My name is Karen, and I'll be your conference facilitator today. At this time, I would like to welcome everyone to Micron Technology's Fourth Quarter and Fiscal Year End 2012 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, Karen. I would also like to welcome you to Micron Technology's Fourth Quarter and Fiscal Year End 2012 Financial Release Conference Call. On the call today is Mark Durcan, CEO and Director; Mark Adams, President; and Ron Foster, Chief Financial Officer and Vice President of Finance. 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 fourth quarter and fiscal yearend 2012 financial press release, again, it is available on our website at micron.com. Our call today will be approximately 60 minutes in length. There will be an audio replay of this call, accessed by dialing (404) 537-3406 with a confirmation code of 30042904. This replay will run through Thursday, October 4, 2012, at 5:30 p.m., Mountain Time. A webcast replay will be available on the company's website until September, 2013. 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. 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 our 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 with that, I'd like to turn the call over to Mark Durcan. Mark?

Thanks, Kipp. I'd like to start today with a brief overview of the key developments during the quarter and an update on the proposed Elpida acquisition, followed by a recap of fiscal 2012 and our focus heading into 2013. After that, I'll turn it over to Ron for a financial summary, and we'll close our comments with Mark Adams discussing key developments in our business unit and operations, as well as an update on market condition. Revenue in the quarter was down about 10%, primarily due to a reduction in DRAM bits. You may recall that we had forecast in our previous earnings call that this would be occurring as we transition significant production to more forward gigabit densities. Also impacting this comparison was a significant reduction in inventory for both DRAM and NAND that occurred in our fiscal Q3. Gross margins were pretty flat quarter-over-quarter, with improvements in NAND and NOR and offsetting reductions in DRAM. From a market perspective, NAND flash demand improved during the quarter. And this, coupled with muted industry supply, led to improve ASPs compared to the quarter-to-date estimate we gave you in late June. On the other hand, DRAM market prices deteriorated during the quarter, primarily related to weakness in the PC environment, although our net DRAM ASP, A-S-P, was essentially flat due to mix improvements. NOR revenue remains stable in the quarter, and we're pleased to see NOR gross margins improved by about 5 percentage points, mostly due to cost reductions. Total inventory was down again in fiscal Q4. We're continuing to focus on improving our turns over the next several quarters. In DRAM, 30-nanometer yields are on track, and we should see a nice jump in production and cost improvements over the next few quarters. From a NAND perspective, we had the first full quarter of results following the restructure of our supply and technology agreement with Intel in April. All product coming out of IMFS in Singapore and our Virginia fab now flow to Micron. As a result, we have a much larger base in wafer capacity in our trade NAND trade business, targeting key gross segments including SSDs, smartphones and tablets. Although we would have preferred to be reporting better financial results for fiscal 2012 today, looking back, we had some significant highlights in the year. We achieved increased manufacturing scale and performance as we completed our volume manufacturing ramp of our world-class flash memory facility in Singapore on both 25- and 20-nanometer technology and transitioned Micron's DRAM operations to 30-nanometer DRAM manufacturing. We launched the Hybrid Memory Cube product line, reinforcing our leadership in high performance DRAM technology. We completed the successful implementation of our new global supply chain management system. We achieved a successful outcome in the Rambus antitrust trial. We redefined and expanded our joint venture with Intel. We signed a sponsorship agreement to merge Elpida's operations into Micron and which, upon close, will place Micron as the industry's leading memory pure play and one of the top 5 semiconductor companies in the world. Relative to Elpida, let me elaborate on a few of the items and update you on the timeline to close. The Japanese court and Elpida trustees ran a rigorous and thorough sponsors selection process and determine Micron's sponsorship proposal to be the most attractive alternative. We believe the proposed agreement represents the best solution for Elpida's debtholders, employees, customers and the Japanese semiconductor and electronics industries. For Micron, the deal will provide greater manufacturing and R&D scale, as well as a significant product, customer and cost synergy, all at attractive economics for Micron shareholders. The combined company will be well positioned to grow and thrive, optimizing a potential return for all involved. The financial terms and expected timeline to close have not changed since our announcement in July. On September 11, the U.S. HSR waiting period expired, and depending on the timeline for other government approvals, we still expect to close some time in the first half of 2013. Entering fiscal 2013, we are focused on driving our return to profitability with a continued emphasis on cash flow. We do face some headwinds from the macro environment and from the demand profile in a number of segments. Specific to DRAM, prices are currently soft. While certain amounts of supply has come offline, we still need to see improvement in demand in the market to get things back on track. The upcoming Windows 8 launch and proliferation of DRAM in mobile devices should help get the ball rolling. Although it's hard for us to predict the timing of our recovery, Mark Adams will comment later on this. NAND market prices are currently up quarter-to-date, following the nice ASP recovery over the summer and we don't see any disruptive supply coming online in the near future. We expect strong demand for SSDs and ultrathins in enterprise servers and storage, as well as the proliferation of smartphones and tablets to fuel significant NAND consumption. From an operational standpoint, we continue to drive manufacturing cost reductions and optimize our product mix, focused on the best margin and growth opportunities. We also intend to be very disciplined in our spending profile, both in terms of overhead and CapEx and manage our inventory levels carefully. I'll stop here and turn it over to Ron and Mark before returning for Q&A.

Thanks, Mark. Our 2012 fiscal year ended on August 30. On our website, we have provided a schedule containing certain key results for the fourth quarter, as well as certain guidance for the next quarter. That material is also presented on a few slides that follow. First, the fiscal 2012, it ended in a net loss of just over $1 billion or $1.04 per share on net sales of $8.2 billion. Fiscal 2011 comparatively showed roughly breakeven net income on sales of $8.8 billion. Total sales for 2012 decreased about 6% compared to the prior year as a result of the generally severe price pressure in the memory industry as a whole. DRAM sales for 2012 reflect a 12% decrease compared to 2011, the product of a 59% increase in DRAM bit sales, offset by a 45% decrease in average selling prices. Bit sales in the premium DRAM sectors increased 62% in 2012 compared to 2011. Trade NAND sales grew 19%, the product of bit sales over 2.5x the 2011 level, offset by a 55% decrease in selling prices. Trade NAND sales were favorably affected by the higher mix of Micron SSDs in 2012, which comprised 14% of trade NAND sales for the year. Sales of NOR products decreased year-over-year in line with the general decline in market demand and the performance of our customer base in particular. On the cost side, DRAM bit cost in 2012 decreased 32% compared to the prior year, roughly consistent with the long-term industry trend. NAND bit cost over the past year decreased 45% compared to the prior year as we leverage our industry-leading process technology. Broadly, annual sales for 2012 were comprised of about 44% NAND, 39% DRAM and 12% NOR. Turning now to the fourth quarter results. We posted a net loss of $243 million or $0.24 per share on net sales of $2 billion. These quarterly results compare to a net loss of $320 million or $0.32 per share on net sales of $2.2 billion for the third quarter. Fourth quarter revenue was sequentially lower in total, as well as for both DRAM and NAND on generally flat pricing for DRAM products and slightly lower pricing for NAND products. Although DRAM pricing was generally down in the fourth quarter, as Mark mentioned, improved mix in our premium segments yielded an overall flat average selling price. Both DRAM and NAND saw lower volumes of bit sales. As a result, when combined with stable bit cost for both DRAM and NAND across the 2 quarters and at better NOR margin performance, consolidated gross margin was fairly stable for the third quarter in a row. Depreciation and amortization in the fourth quarter was $509 million. This decrease, compared to recent historical periods, is due to certain production equipment primarily in the Lehigh and Virginia NAND and 200-millimeter NOR operations that have become fully depreciated. We anticipate depreciation and amortization expense will be approximately $2 billion for fiscal 2013. As a result of the changes to the IM Flash joint venture with Intel that occurred in the third quarter, NAND sales to Intel decreased by half in the fourth quarter, consistent with our previous projections. That volume has now shifted to trade NAND going forward. However, trade NAND bit volume growth for Q4, although consistent with our projection, did not reflect the magnitude of this shift for 2 reasons. First, onetime WIP sales to Intel, which were part of the IM Flash transaction, masked the normal Q3 to Q4 growth rate. And second, some production volume went back into WIP as we refilled the manufacturing pipeline with Micron trade products, so you can better ascertain the underlying growth rate by comparing trade bit growth from Q2 to Q4, which grew 78% over that period. A higher volume of SSD sales in the fourth quarter mitigated the decrease in trade NAND selling prices compared to the third quarter. Price decreases for NAND components enabled increased penetration into the client SSD space. Sales of enterprise SSDs were also up nicely albeit on a lower base. Trade NAND selling prices are up a couple percent quarter-to-date, while trade NAND cost per bit is expected to be down a few percent in the first quarter. Trade NAND bit growth is projected to be flat in the first quarter compared to the fourth quarter as we transition to new technology nodes. DRAM revenue on the fourth quarter decreased 9% compared to the previous quarter as a result of a 9% decrease in bit sales volume, consistent with our prior guidance, due to a mix shift to lower density specialty DRAM, as well as initially lower yields on our new 4-gig DDR3 RAM. Selling prices for specialty DRAM products improved slightly in the fourth quarter, while prices in the PC sector dropped significantly due to weakness in demand. Quarter-to-date, selling prices are down on the high teens when compared to the Q4 average. DRAM cost per bit declined slightly quarter-to-quarter, and we are forecasting bit cost for DRAM in the first quarter to be down in the high single-digit range as we continue to transition to our 4-gig DDR3 product. This transition is also driving bit production up low- to mid-teens in the first quarter. NOR sales increased slightly in the fourth quarter compared to the third quarter and represented 12% of total revenue for Q4. NOR revenue in the first quarter is expected to be in the same range as the fourth quarter. We anticipate NOR cost reductions will continue to outpace selling price reductions in the first quarter. Turning now to business units. The NAND Solutions Group revenue declined from the third quarter due to the onetime impact of selling back-end inventory to Intel during the third quarter that I mentioned earlier. Trade NAND sales and NSG were up in the fourth quarter on increased volume and stable average selling prices. NSG was able to achieve positive operating income from the third quarter level on improved product mix sold in the fourth quarter, including a 67% increase in bits sold in SSD form, as well as lower cost base sales to Intel. Operating income from the DRAM Solutions Group in the fourth quarter primarily reflects the lower DRAM revenue trends I commented on earlier. DSG sales into the personal system sector decreased as we move volumes into higher-margin areas, notably networking, as pricing in the personal systems came under additional pricing pressure, particularly in the latter part of our fourth quarter. DRAM sales into the personal systems market were just 14% of total sales in our fourth quarter. Our Wireless Solutions Group performance continues to reflect weakness of the feature phone market segment and our customer group, in particular. Revenue declined for the third quarter, primarily due to lower NAND sales, as we transitioned to new products and configuration and as customers worked through existing inventory. Wireless DRAM and NAND declined ahead of product transitions, while wireless NOR sales remain steady quarter-to-quarter. The Embedded Solutions Group results continue to reflect growth, particularly in the automotive and industrial sectors. In fact, in the fourth quarter, ESG reported its highest revenue levels since the establishment of our business unit. The revenue growth came from the broad strength across the technology portfolio as NOR, NAND and DRAM revenue increased from the third quarter. In addition to the top line growth, ESG operating income strengthened as the fourth quarter results included better cost performance than the previous quarter and improved factory utilization. In operating expenses, SG&A expense in the fourth quarter was slightly below our projected range as a result of lower costs associated with pending legal matters and lower personnel costs from our variable pay plans. We expect SG&A expense to be between $135 million and $145 million in the first quarter of fiscal 2013. Fourth quarter expense for R&D was right at the high end of our guided range. These expenses vary primarily due to the volume of development wafers processed and the timing of the product quals. R&D expense in the first quarter is expected to be between $220 million and $230 million. The company generated $450 million in cash flow from operating activities in the fourth quarter. Comparing operating cash flows in the third quarter to the fourth quarter, the IM Flash restructure transaction had a favorable impact on Q3 from the $300 million deposit received from Intel in the third quarter, while the fourth quarter operating cash flow was negatively impacted by $45 million of product sales, invoices charged to that deposit. We ended the fiscal year with a cash and investments balance of $2.9 billion. This balance increased $718 million compared to the previous yearend, as we generate $2.1 billion of cash from operations, raised $1.5 billion net new financing and spent $1.9 billion in capital expenditures. Inventory declined $82 million quarter-to-quarter, where finished goods reductions more than offset the WIP increase related to refilling the back-end pipeline after the Intel transaction. We expect expenditures from property plant and equipment for fiscal 2013 to be between $1.6 billion and $1.9 billion, slightly weighted toward the earlier part of the fiscal year. Now I'll turn it over to Mark Adams for his comments. Mark?

Thanks, Ron. Today, I'm going to walk through some of our fourth quarter operational highlights, as well as discuss the current market dynamics in memory business. Our NAND Solutions Group recovered from a weak ASP environment in the beginning of the quarter. If you recall, ASPs were down mid-teens this quarter-to-date during the first few weeks of our Q4. Prices recovered during the quarter, and the trade NAND business ended up with slight revenue growth and stable gross margins. In addition, finished goods inventory declined quarter-over-quarter. Micron SSD revenue was up 33% in the quarter with unit shipments up over 50%. We're seeing steady growth of our client SSDs at key OEM partners and continued growth of Crucial blended SSD drives in the channel. Where other channel competitors have struggled to generate a profit, Crucial continues to drive solid financial performance. We are now shipping client SSDs to 5 of the top OEMs in the world and plans to grow our share in the coming year. Client SSD revenue was up close to 30% in Q4. For fiscal year '12, unit shipments more than doubled. Also, part of our growing client portfolio is our new mSATA SSD, which was selected as Best of Show Award winner at this year's Flash Memory Summit as the Most Innovative Flash Consumer Application. On the enterprise side, our SATA drives, the P300 and P400e, continue to be the lead drives for us this quarter in terms of shipments. We also maintained steady progress in qualifying our PCIe drives with leading OEM customers and are now gaining traction to our distribution channels as well. In Q4, Enterprise SSD revenue was up over 50% albeit off of a relatively low base. We're gearing up for some new products this fall, including our enterprise grade SATA drive and our first SAS drive. Both drives are involved in casing now with OEMs. Unique features we provide with these new products showcase our silicon to system strategy including the firmware and hardware management schemes that are only available from a Micron-integrated SSD. Another example of our growing system capability is the acquisition of Virtensys this fiscal year, which had provided Micron with a virtualized appliance. In combination with our P320 PCIe cards, it provides much more storage in the appliance versus mobile server storage, and when we hook these up via PCIe interconnects, you can share the storage across an enterprise environment. We've also made significant strides in the quarter and in the fiscal year with our internal controller development. Our controller strategy is really 2-pronged, using both internal and externally-designed controllers. We're focused at the high end of our system solution for our internally design controllers. However, our approach is modularized and can be scaled down to the lower-end client systems over time. On a NAND technology operation front, we continue to be pleased with 20-nanometer ramp and expect production crossover in the first half of calendar year 2013. MLC represented about 75% to 80% of our wafer production in Q4, with SLC and TLC essentially splitting the remainder. Although fiscal Q1 will only see a small reduction in trade NAND cost per bit, we expect to average down mid to high-single digits over the next several quarters, mostly related to the 20-nanometer execution. As I mentioned earlier, following a tough first half of the calendar year, NAND prices started to recover over the summer, and we have seen continued improvements in market prices subsequent to our quarter end. In terms of NAND inventory, we see the channel balanced about 3 weeks. We are excited about the drivers for NAND, including SSDs, tablets and smartphones and believe the industry bit demand compounded over annually 50% [ph] from 2012 through 2016. In 2013, NSG bit supply forecast are up in the low 50% [ph] range, well below historical levels. We continue to see balance between our current capacity and serving our customer needs. Our DRAM Solutions Group achieved record shipments in networking, storage, graphics and consumer segments this quarter. On the server side, we increased market share with existing customers and won [ph] critical qualification slot that our new customers are [ph] in the high-growth data center markets. We did, however, see some price pressure in the server market as OEMs compete for this growth-oriented data center business. Nonetheless, this server-growing market segment share is helping to drive DRAM bit demand over 50% year-over-year. The networking segment had another strong quarter, exceeding targets for revenue, bit growth and gross margin. In terms of demand in the segment, LTE adoption is progressing well. While the Americas and Japan have strong investments in the infrastructure, so does -- China will continue to grow despite some weakness showing in the European market segments. During Q4, we have seen softness in PC demand, although we were able -- still able to maintain minimal ASP declines in the quarter. That being said, we are seeing price weakness in Q1 and see short-term challenges with channel inventory overhang and OEM customers remaining generally cautious. The demand catalyst for improvement in the PC DRAM segment include the upcoming Windows 8 launch, with both Intel and ARM-based products expected in key OEMs, in particular, with the Ultrathin category. From a technology perspective, our 30-nanometer process nodes saw volume ramp in fiscal Q4 with qualifications across the broad customer base. The timing of our 4-gigabyte transition actually limited our output in fiscal Q4, but we expect to catch up this quarter, which, coupled, with the 30-nanometer technologies, providing significant bit growth and cost reductions. Another key aspect of our cost per bit and output improvement over the next couple of quarters is Inotera's execution. In fiscal Q4, they took steps to remove a production bottleneck during what was a slow demand period. Moving forward, the throughput and percentage of wafers on leading-edge technologies from Inotera will improve. In aggregate, we expect crossover output on 30-nanometer in the current quarter, putting us in a much stronger competitive position. We are also progressing with a 30-nanometer design strength [ph] , which will give us additional cost reductions over the next couple of quarters. Our 20-nanometer process node will commence in calendar year 2013. All in all, we see quarterly cost per bit reductions averaging mid- to high-single digits over the next several quarters. DRAM market prices were generally stable coming into the quarter, although things began to weaken in July and we're seeing that trend continue as we head into fiscal year 2013. In terms of DRAM inventory, we believe that the channel is currently running about 6 to 8 weeks on average. Some of the short term weakness in PC DRAM seem attributable to a pause prior to the Windows 8 launch. We anticipate things will improve with the new OS and in combination with Ultrathin form factors, although timing is difficult to predict. Outside of PC, we're generally pleased with DRAM demand, with several segments driving bit demand well over 40% year-over-year. That said, continued mild or even lower industry supply bit growth might be the required catalyst to get DRAM ASPs back to profitable levels in the future. Current industry bit forecasts are high 20s -- or 20% for 2012 and similar for 2013 although there's no certain possibility of further supply or CapEx cuts in the industry given current ASP weakness. Revenues in our Wireless Solutions Group declined with both DRAM and NAND as we continued [ph] to work on aligning our product portfolios to customer demand. NOR sales stabilized and were essentially flat quarter-over-quarter. As we rebuild our wireless business, we remain focused on profitability. Despite the top line revenue decline, we see better operating performance in WSG, which will be our focus when we prepare for a combination with Elpida's wireless business. Fiscal Q4 saw the introduction and rapid growth with the smartphone open market in China. This is expected to represent the majority of the rapidly growing Chinese smartphone market, which is forecasted to grow from 50 units -- 50 million units in 2011 to 600 million units in 2014 due to the rapid transition from feature phones to smartphones. During fiscal Q4, this emerging market adversely affected our wireless revenue due to abrupt NAND MCP density changes. However, in the long run, we anticipate this modest growth to be a huge opportunity for Micron due to our future portfolio alignment. One example is the ramp-up of our 4-gigabit low-power DDR2 mobile DRAM, which is growing in present significantly in a portion of the market. We're also excited to announce the official qualification of our 1-gigabit phase change memory-based MCP product with 2 key mobile customers. We expect volume shipments to commence in fiscal 2013. I also want to take this opportunity to welcome Mike Rayfield, who recently joined Micron as our new Vice President of the Wireless Solutions Group. Mike has an impressive background in the wireless business. I know he is excited to hit the ground running and help position this segment for growth and profitability in the future. Our Embedded Solutions Group had another solid performance with record revenue in gross margins driven by strong quarter-over-quarter shipment growth in DRAM, NAND and NOR and across all regions outside of Europe. We have strong design wins across all technologies and clearly strengthened the networking in automotive sectors during the second half of the year. In automotive, where we are now ramping the MMC products for the segment, which is expected to drive significant revenue in fiscal 2013. We've also completed qualification on over 150 platforms at 60-chipset partners during the last year. These qualifications span all segments with particular strength in consumer networking, ensuring that Micron embedded memory solutions were supported and pre-validated with chipset solutions to enable our customers fastest time in the market. We are leading the NOR industry from a manufacturing and technology perspective with 45-nanometer now ramping in mass volume on 200 millimeter and then the recent introduction of 45-nanometer, 300-millimeter NOR in our Virginia fab. Moving into 2013, we will continue to focus on growing our share in key embedded segments leveraging our product portfolio, technology leadership and manufacturing scale. Our operational performance in fiscal 2013 is to continue the deployment of advanced technology, which I described earlier in the business units discussion, as well as improved cost efficiency in our fabs, optimize our inventory and enhance supply chain management. Our broad product portfolio extending scale and global manufacturing presence requires a greater focus on supply chain management. The same goes for inventory management. We are tasking our business units and our sales teams to increase turns and quarterly sales linearity for high-volume products, while we're also ensuring we have the right levels of strategic inventory for premium margin-specific products. While we have experienced ultimate market conditions over the last 12 months, we are optimistic that we can see improved money market looking out over the next 12 months. While DRAM prices have indeed weakened, we have seen a nice recovery in NAND ASPs and our overall NAND demand while we're benefiting from slower industry supply. Our NOR margins improved in Q4 and this business continues to generate substantial free cash flow. We remain focused on optimizing our product portfolio, as well as leveraging our technology leadership and manufacturing efficiency to enhance our financial performance in 2013. Now I'll turn it back -- the call back over to Kipp.

Thank you, Mark. We will now like to take questions from callers. [Operator Instructions]

[Operator Instructions] Our first question comes from the line of James Schneider from Goldman Sachs.

I guess, first of all, on the NAND side, I think you referred to some industry bit growth estimates for 2013 in the low 50% range. I was wondering, based on what you see today, do you think we're going to end up for the industry on the high or low end side of that? And where do you expect Micron to come in, in terms of that industry bit growth, higher or lower?

I think the answer to your question is probably both on the low end, both the industry and Micron.

Okay, fair enough. And then could you maybe talk about in the Wireless Solutions Group what are the steps you're going to take to start turning around the profitability in that segment? I think you obviously referred to improving gross margins in the NOR Flash space. But can you talk to maybe some of the actions you might look to take, either operationally or in terms of pulling apart portfolio there?

Sure. I think it's important to note that the business -- our wireless business has evolved from both the acquisition of Numonyx and kind of where this segment was. And then we had a pretty high heavily concentrated business in the feature phone market. We're -- we've been moving our product development efforts, as well as product roadmaps towards the smartphone piece of the business and our advancement in terms of low-power DRAM will allow us to get more aligned with the broader demand cycles in the mobile phone market. So I think from a product perspective, we're well aligned going forward. We also have, in taking a look at our spending across NAND, DRAM and NOR, to make sure we're focused on these future products and making some test choices on some of the existing or historical legacy products that have not generated profit. So between the expenditure side and some product alignment and, of course, our new leadership, we feel pretty good about the business going forward.

That's great. And the last question for me will be, I think you gave a number in terms of the SSD sales for fiscal '12. Do you care to make any kind of guess about what that number could be like on a -- or your basis for 2013?

You know us, James, we try not to predict anything like that or forecast it, but nice try.

And our next question comes from the line of C.J. Muse from Barclays.

I guess first one on the DRAM side, can you talk about what you're seeing there in terms of utilization rate, particularly from the Tier 2 guys? Are you starting to see them ratchet back or they're still lingering?

Really, we just read the same stuff in the press that you guys do, and it's hard for us to comment beyond that sort of speculation.

Okay. I guess, thinking a little bit longer term, can you share with us what the mix was, PC, server mobility, et cetera, in August? And where you think that could move to 12 months from now and how we should think about the implications to gross margins for that business?

Sure. On a revenue basis, we were exposed to the personal systems by about 15%. This is of total revenues, had a pretty strong server business, low double digits; networking and storage, mid-teens; mobile, low-double digits; and then AIMM was around 10%.

And care to talk about where you see that going 12 years -- 12 months from now?

Can I just say just going to repeat my answer from the last question? It's really hard, especially we're talking revenues. You have to also predict not only your growth rate in bits, but also what ASPs are going to be doing, and that's just pretty difficult to determine.

Sure, very fair. And then last question for me on the NAND side. Can you walk through the roadmap there in 20-nanometer? When we should start to see -- you talked about, I guess, mid-to-high cost down each quarter. Will that commence in calendar Q1, Q2? How should we think about that?

Go ahead, Mark.

I think the answer is we're really focused on hitting more of the production crossover towards the end of the first half of 2013. So then you'll see some of that costs flow through to our NAND costs. So today, we're shipping the product but expect to continued increased volumes that allow us to realize big cost savings.

If I can add, I can just refer you as well to your guidance sheet. You can see there for Q1 '13, we're suggesting down a few percent in NAND and then a more aggressive cost down over the next few quarters after that.

And we also have a question from the line of Daniel Amir from Lazard.

Can you a bit expand how you see kind of the PC market playing out in the next year given Windows 8, Ultrabooks, some of the change that you've done also to the DRAM franchise? And then I have one follow-up.

Well, I think we commented earlier that, currently, we're in a soft PC DRAM market condition, and we do have some belief that between Windows 8 and a further enhanced delivery of ultrathin products from across the OEM base, can infuse some higher demand signals. It's hard to call. It doesn't seem like it's going to be in the real short term, but we think over the fiscal year, we see some balance there in terms of the markets, some recovery in the PC space. But that's really those 2 demand drivers offset, obviously, by smartphone and tablet growth and how that impacts the overall long-term piece of the business.

Okay. And then in terms of -- how should we look at kind of the best guess would be on cost-downs in the DRAM space here in the next year? I don't know how much you want to comment on potential Elpida, but maybe without Elpida.

We're looking at around 30% to 40% on DRAM.

And our next question comes from the line of Steven Chin from UBS.

First one I had relates to your mobile DRAM products. I know it's still a relatively modest business for you currently. But referring from your perspective, what was the current average content for mobile DRAM in the smartphones that you guys have exposure to? And also, is your ability to produce mobile DRAM, is that going to be the -- is it the limiter to how big you guys can grow your MCP business?

Sure. I can start with the first part of that with the raw data, then maybe Mark can jump in with some of the spots around how the market's set a bit. Today, on average of all handsets, you're looking at about 315 megabytes per phone, average handset and about 3.5 gigabytes of NAND. In the smartphone segment itself, you're currently looking at smartphones in about the total 600, low 700 megabytes per handset and more in the low-30s to 40 -- low-30s gigabytes in NAND in the smartphone segment. The average megabytes per handset next year by third-party analysts are expected to grow again over 100% and NAND up around 50%.

Okay, perfect. And just looking at your solid-state drive business. Obviously, you do have a number of different products that are helping to expand portfolio, both on the client side and enterprise side. Care to take any guesses then as to what kind of margin profile the overall business will look like in a years' time, just given the different enterprise pieces, whether it's PCIe or the flash drive?

You guys are really making it easy for me to answer. I keep saying, sorry, too hard to predict.

Okay. And then just one last question. In terms of capital equipment reuse, can you remind us, given the latest generation of NAND, the lithography equipment that you have, what would be the smallest node that it can be reused on for DRAM?

This is Mark. There's really a lot of fungibility and we have really state-of-the-art lithography equipment in all our manufacturing fabs. Depending on the manufacturing technology we use, that equipment is generally fungible down to 15 nanometers on either technology.

And our next question comes from the line of David Wong from Wells Fargo.

In terms of your technology, you were talking about NAND transitions down to 20 nanometers. What's your visibility at the moment for future NAND transitions? And when do you expect to have to have EUV to keep progressing?

Well, we -- as I just mentioned, David, we can take the existing immersion -- high NA capacity on -- down to the 15-nanometer range. Obviously, we'd love to have a low-cost EUV solution available. If the manufacturing costs were there today, we'd use it today, but we can continue to migrate our technology without it, and we just continue to monitor that. We'll be adopters as soon as we see the return.

Right. And on Elpida, if I understand correctly, in addition to the regulatory agreement, you still need a final decision from the court for the deal to progress under the negotiated term. Is that correct? And if so, is there a deadline for when the court will make that determination?

We still see this deal closing the first half of 2013, and we see that the -- from our perspective, the bottleneck in that process is likely to be the regulatory approval. The court process will run its course, and we think that the regulatory approval will overlap or extend beyond the end of that process.

And our next question comes from the line of Ryan Goodman from the CLSA.

I have a question on the DRAM side of things. In the past, you've talked about specialty DRAM prices as trailing your commodity trends with lower highs and then higher loads, so sort of a smoother trend over time. But then if we look back over the past year, there really wasn't that boost early in the year in prices when commodity was doing well. Last quarter, it sounds like you've kind of tracked in line with flattish trends and then looking ahead, you're talking about down in the high teens. So I guess, how should we think about this business going forward relative -- how it will trend relative to the commodity trends?

Well, I think that certainly there are certain segments that are experiencing significant growth but there's also new dynamics going on in the market segment, and I'm thinking specifically about server where we've seen tremendous application growth, but also on the low end of that segment, some level of commoditization with generic platforms and non-branded solutions where, even in the likes of Google, now being one of the largest server manufacturers in the world. So I think we're seeing pieces -- parts of that segment under some price pressure due to commoditization of it. But overall, we remain very bullish about the server business and I think it will be continue to behave as such. And the other side of that equation is that there wasn't really a big rebound as you might have suggested in the first half of the year. We didn't see that. It might have slowed, but it still -- DRAM didn't really have a recovery in such that it's just got a longer, prolonged pricing environment, especially the lag and now kind of in line with where it normally is.

Okay. And then just a question on the NAND side of things. We've seen a kind of a reshuffling in the competitive landscape, at least on the custom embedded side of things where -- with some of the larger buyers out there have shifted their supplier base. Just curious what you guys are seeing in terms of how that impacted the opportunities you have? Has this opened up anything? Or are you seeing increased competition on fronts where you haven't in the past?

Really, Ryan, we haven't seen much of a change. The opportunities to increase our market share are available to us from various customers. And as we -- as Mark noted, we're excited to get our hands on additional trade NAND bits that we can go service these guys in a higher manner.

And our next question comes from the line of Doug Freedman from RBC Capital Market.

Mark, you made a mention that you guys have the fungibility on the equipment front to go to lower nodes. How about transitioning from DRAM to NAND? And when I look at what's happening in the marketplace, it sounds like the NAND market is clearly in better space shape than the DRAM market, but yet your output is increasing fastest on the DRAM side. How do we reconcile that?

Well, Doug, I think that flexibility is definitely there and that's certainly something we have the ability to do as we migrate to future nodes. The opportunity to take DRAM capacity and move it to NAND is there. It's cost effective and efficient. It's not something you want to do with high frequency, but it is something you can do in Japan. Roadmaps in technology node transitions to efficiently take DRAM capacity and migrate it to NAND. And for that matter, it's the opposite of your thesis, but you can do it the other way around as well.

I guess my question is, really looking at next quarter, you're getting production up in DRAM faster in bit terms than your NAND side and I'm trying to understand why you're doing -- why you wouldn't have inverted that this quarter, given the condition in the marketplace.

Simple answer, Doug, it's just the timing of the strengths. We've got a real nice, steep ramp here on 30-nanometer DRAM in Q1. And our real steep ramp in NAND bits going to 20-nanometer really happened in Q2 and 3.

And I guess as sort of a follow-up, how should we think about the status of your inventory? And whether you -- inventory did come down a little bit this quarter. Is that the right level of inventory? And how should we think about your ability to model inventory for the next year, so to speak? Is there an opportunity to run your factories a little leaner so that there's higher turns? How should we think about that?

Doug, as I commented in my opening discussion, we will continue to look at inventory. Over the last couple of quarters, I think the team has done a pretty good of job on that front. And then we also, as we take a look at some of these more system-level solutions, enterprise is a steeper example. It's -- when we take capacity from just selling raw components to building solutions-oriented products, it might limit some of the future reduction potential. But we will continue to look at it, and we feel pretty confident we are putting the right controls in place to make sure inventory stays at the appropriate level for the business.

And my last question for you, can you guys give us an update on new technologies? I mean, we talked at the last Analyst Day about a whole laundry list of potential candidates for the next generation technology. Where do we stand on that and mainly timeframe?

Well, we continue to make good progress on a number of different technologies targeted at really slightly different applications. Obviously, we're in volume manufacturing now on phase change. So I don't know if you count that as an emerging memory anymore because we really don't -- we count it as here today and going into volume applications in 2013. But beyond that, some of the ROE RAM technologies and other more advanced next-generation memories, we're making good progress on them. I don't see any of them having a significant impact in 2013 or really in 2014 with the possible exception of some volume on vertical NAND technologies, if you include those in that bucket.

And our next question comes from the line of Alex Gauna from JMP Securities.

If I heard you right, you talked about what mobile DRAM in and being down in the mobile category. And I was wondering, is that a phenomenon of your customers working down inventory during what should otherwise be seasonal strength? Is it ASPs? Or is it simply your customer mix and maybe not having the best share positioning as we go into the holiday season?

I think it's really the latter area. We're working ourselves through some customer concentration, customer mix issues that don't really reflect the larger in potential as we transition. But that's what's causing some of the impact to the financials.

How long would it take you to maybe realign that customer concentration out? Can you -- what kind of window should we look forward to?

I would just say this, without trying to predict too much in the future, Alex, I think that we've been working on this for some time. And remember, some of the product technology that we have that goes in that space, the value of commodity line of the wireless space, we can also use in other segments, namely embedded. So as we're focusing our customer engagement development efforts on some of the better pieces of the wireless business, we're able to use some of that capacity and drive ESG, for example, to higher -- their highest quarter since they've been formed as a BU. So I don't want to predict the future. I will just say that this is a known process that we're going through in the company that obviously, hasn't yet hit the financials but we're pretty encouraged by things like low-power DRAM and our 30-nanometer nodes and the uptick that we've seen in wireless, in addition to our longer-term prospects of a combined Micron and Elpida wireless business.

Okay. And then if I could ask, SSD seems to be performing pretty well but the DRAM, as it pertains to the PC industry, is struggling. How can you reconcile that? Is the SSD business experiencing some of the similar weakness in DRAM? Or has your share gain been so great that it masks it?

I think when you look at SSDs, the client SSD business is coming from such a lower penetration rate that the -- uptick giving more pricing add and the value consumers are placing on it continues to improve. And the core DRAM growth, obviously, doesn't in line with the same traffic, so it's kind of in a much larger share of the platform.

Okay. One more, if I could. You mentioned Windows 8 being a catalyst eventually. It sounds like you've not yet seen some much of the boost. Can you talk about your expectations around the build cycle there and maybe about average density as it pertains to now that we have Windows RT and we have Windows X86?

We believe that the builds have largely taken place. And as Mark mentioned in his commentary, PC OEMs are being extremely conservative. They don't want to have a high risk of inventory obsolescence. And in terms of the second part of your question, remind me again?

I guess I was wondering about what this does to average density, especially now that we've got to deal with the factor of the Windows RT versions out there.

Yes, let me answer it maybe 2 ways. If you go on to a lot of the online website sales organizations for the various large OEMs, you're seeing a fairly large concentration now at 6 and 8 gigabytes. That being said, when you look at the third-party data, they're only suggesting about a 20% to 25% bit growth from the PC area for next year.

And our next question comes from the line of Dean Grumlose from Stifel, Nicolaus. Dean Grumlose - Stifel, Nicolaus & Co., Inc., Research Division: This is Dean Grumlose calling in for Kevin Cassidy. I have a question and a follow-up. As we head into next fiscal year, can you comment on your expected change in the NAND Flash wafer output?

We don't expect a large change in the number of wafers currently. Dean Grumlose - Stifel, Nicolaus & Co., Inc., Research Division: Okay. And as a follow-up, do you have a projection on how your mix of SLC versus MLC and TLC may change, heading into next quarter?

It won't change all that much from where we are today. We're running basically high 70% MLC and the balance is split pretty evenly between SLC and TLC.

And our next question comes from the line of Jagadish Iyer from Piper Jaffray.

I was just wondering, Mark, can you give us some color in terms of the past quarter, in terms of how the sequential ASP trends were between the 2 SSD segments? And how much more value can you add to strengthen the ASP in that segment? And I have a follow-up.

Sure. Well, I think, as the track of cost per gigabytes and ASPs have gone throughout our 2012 year, the demand cycle for client SSDs went up significantly, and that's what's showing up in our financials as far as the client SSD performance. When I look at how much more value is there, I think the value around that is driven by the industry's ability to remain competitive on the ASP front to drive a good level of demand on the longer term PC builds. I think that seems to be in a pretty good place today. So I think there's -- with increased or newer ability to drive cost down over the next 6 to 12 months, we would anticipate a continued demand growth there. On the enterprise side, there's a lot more ability, obviously, to differentiate both in terms of form factor, as well as performance and reliability features, that we think is already in place but will continue to grow as the development -- as we talked about earlier in our controllers and firmware and error correction, those types of things fit into our product portfolio, which will again allow us to drive more differentiation and value.

Follow-up on this. Can you give us some color on the 2013 CapEx between the DRAM, NAND and the NOR segment? And does it include Elpida at this point of time? Or should we model something going to be incrementally for Elpida once the acquisition gets closed?

Yes. So I think in terms of NAND and DRAM, roughly balanced. And that number that we floated out there, $1.6 billion to $1.9 billion, we continue to work on that. And as we really tune our investment for 2013, just on what's strictly needed to drive particular product introductions and efficiency in our technology implementation. Elpida is not included in that number. Obviously, depending on when we close Elpida, the numbers will change. But I think you should expect that to the extent we are investing in Elpida, we'll be trading those opportunities off against internal opportunities to make sure that we'll put the capital in the highest return on investment.

If I could just add one quick follow-up. Can you just characterize how much of it was supply cost in the DRAM between the first half of this year and the second half of this year?

Probably need to ask the people who actually made the cut, to be honest with it.

With that, 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 a number of risks and uncertainties and actual results may differ materially. For information on the important factors that may cause actual results to differ materially, please refer to our filings with the SEC, including the company's most recent 10-Q and 10-K. Thank you for joining us.

Thank you. This concludes today's Micron Technology Fourth Quarter and Fiscal Year End 2012 Financial Release Conference Call. You may now disconnect.