Micron Technology, Inc. (MU) Q1 2018 Earnings Report, Transcript and Summary

Good day Ladies and Gentlemen, and welcome to Micron’s First Quarter 2018 Financial Call. At this time all participants are in a listen-only mode, later we will conduct a question-and-answer session and instructions will be given at that time. [Operator Instructions] As a reminder, today’s conference is being recorded. I would now like to turn the call over to Ms. Shayne Hudson. Ma’am, you may begin.

Thank you, Chelsea and welcome to Micron Technology’s first fiscal quarter 2018 financial conference call. On the call with me today are Sanjay Mehrotra, President and CEO and Ernie Maddock, Chief Financial Officer. Today’s call will be approximately 60 minutes in length. This call, including audio and slides, is being webcast from our Investor Relations website at investors.micron.com. In addition, our website contains the earnings press release, filed a short while ago. Today’s discussion of financial results will be presented on a non-GAAP financial basis unless otherwise specified. A reconciliation of GAAP to non-GAAP financial measures may be found on our website along with a convertible debt and capped call dilution table. As a reminder, the prepared remarks from this call and webcast replay will also be available on our website later today. 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. You can follow us on Twitter, @MicronTech. As a reminder, the matters we will be discussing today include forward-looking statements. These forward-looking statements are subject to risks and uncertainties that may cause actual results to differ materially from statements made today. We refer you to the documents we file with the SEC, specifically our most recent Form 10-K and Form 10-Q for a discussion of risks that may affect our future results. 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 today's date to conform these statements to actual results. Lastly, Micron is planning to host its 2018 Analyst and Investor Event on May 21st in New York City. We'll share further details about this event in the coming months. With that, I will now turn the call over to you, Sanjay.

Thank you, Shanye. Good afternoon. Micron's record first quarter results demonstrated the company's continued strong execution, a market environment that reflects the strategic importance of memory and flash storage and healthy supply and demand fundamentals. During the quarter, we continue to enhance our cost competitiveness, achieving yield maturity on both 1X DRAM and 64-layer 3D NAND. We improved our mix of high value solutions, delivering record SSD revenues and further increasing our SSD shares. More recently, we began shipping our first 64-layer NAND consumer SSD. We also introduced the industry’s fastest high-density 32GB NVDIMMN, which combines Micron’s DRAM and NAND to deliver a persistent memory solution that addresses intense data analytics workloads. We have garnered solid interest from enterprise and cloud customers, and customer qualifications are underway. And, we strengthened our talent bench with the recent addition of Manish Bhatia, who leads our global operations. Manish brings extensive experience in managing end-to-end operations and is focused on driving manufacturing and supply chain efficiencies to reduce costs and improve our agility. Finally, we improved our financial foundation with the retirement of $2.4 billion of debt. I’m pleased with our accomplishments and believe our focus on speed and execution better position Micron to deliver value to our customers and capture the increasing number of end-market opportunities. I will now discuss trends and results in each of our major markets. Cloud and traditional enterprise data center trends are continuing to drive robust demand for memory and flash storage solutions. Our Q1 SSD revenue to cloud and enterprise customers increased 50% sequentially. We recovered from the flash component issue discussed in our September earnings call that impacted last quarter’s SSD sales. On the compute side, we had solid sequential DRAM revenue growth into data center markets, driven primarily by enterprise sales. DRAM bit shipments to both cloud and enterprise customers were up by more than 50% year-over-year, underscoring the data center’s growing need for memory and our strong execution in this market. Our 1X nanometer designs have been well received by cloud customers, with more than a quarter of our cloud revenue in Q1 coming from our 1X technology. Fast qualification and production ramp by our cloud customers of new technology node products is a significant benefit, as it diversifies and accelerates our customer traction and market reach during early stages of production deployment of these advanced nodes. The need to access, analyze, and store data extends well beyond the cloud. This is perhaps most apparent in the mobile market. Smartphone capabilities have surpassed simple communication and web browsing, they help us navigate, monitor, and interact with the world around us. New cameras capture precious moments with amazing fidelity, and emerging applications like AR have tremendous promise. This increase in functionality is driving the use of higher-capacity memory solutions and increased storage in mobile devices. These trends, along with our solid execution, drove record mobile revenue in FQ1. We are strengthening our offerings and continue to diversify our portfolio of LPDRAM, MCP, and discrete managed storage solutions to meet the growing needs of our customers. We are accelerating our progress to expand our portfolio of low-power solutions with the release of new products such as our 1X LPDRAM designs. We also shipped initial samples of our 64-layer NAND discrete UFS solution to chipset partners and customers, with very promising results. Home automation and edge computing devices continue to drive strong revenues in consumer and industrial market segments, which require a wide variety of memory and managed storage products. As more edge devices begin to integrate machine learning and intelligence, we see opportunities to provide higher-performance memory and flash storage solutions in these markets. We have also seen rapidly growing demand for our graphics products. The graphics market continues to be fueled by the ever-growing popularity of gaming and eSports. Although smaller in size, recent interest in crypto currency mining has put further pressure on graphics memory supply. Our close customer relationships and leading product portfolio helped drive record graphics revenue, up more than 75% year-over-year. We sampled industry-leading 16Gb-per-second GDDR6 products to key customers and are seeing significant interest in automotive and networking applications that need the high bandwidth this memory provides. We plan to ramp GDDR6 to production in early calendar 2018 for the graphics market, followed by other high-performance applications such as automotive and networking. The rapid innovation in automotive technology towards autonomous driving continues to create significant demand for higher memory capacities and greater performance. We secured a key design win in an important autonomous driving platform this quarter and are focused on replicating our success to retain our leading share in that market. Automotive customers are moving more rapidly to new memory technologies than they have in the past, and our announcement of the fastest 1X LPDDR4 and GDDR6 products for autonomous driving applications will ensure we continue to support this shift to leading-edge technologies. We also set record revenues supplying the networking applications that serve data centers and edge devices, as our reputation for consistency and innovation drives strong ties with networking customers. These diversified growth drivers and structural market trends are generating tremendous opportunity for Micron. We are uniquely positioned in these markets with a broad portfolio of both DRAM and NAND solutions, excellent quality, and comprehensive customer ecosystem engagement. We are focused on developing the right products, deepening our customer relationships, and enriching our revenue mix to capitalize on these opportunities. Turning to manufacturing and technology, our ability to execute our technology roadmap and drive cost competitiveness are foundational to our ongoing success. In terms of wafer manufacturing plans, we still expect to achieve bit output crossover on 64-layer NAND during the second half of FY2018 and expect to achieve bit output crossover on 1X DRAM by the end of calendar 2018. We are outfitting our new back-end factory in Taichung, Taiwan, to ramp assembly and test capacity, and expect meaningful output from the facility before the end of the fiscal year. Our capital investments are tracking with our deployment plans, and we are seeing good traction in improving the efficiency and cost effectiveness of our operations through these investments. Both 1Y DRAM and third-generation 3D NAND development are progressing well, and we remain on track for initial output of both in the second half of calendar 2018. We continued to make good progress with our 3D XPoint technology. Historically, Micron’s efforts on 3D XPoint have been largely focused on technology development and early manufacturing ramp, but given our increased focus on high-value product solutions, we have recently resourced a product development team to address the opportunity ahead of us. Simultaneously, we are working with various players in the ecosystem to assess market and enablement opportunities, and we will provide further details of our views regarding these opportunities during our upcoming analyst event. We will also continue to have the opportunity to sell our 3D XPoint output to our partner as this market develops. Switching to our industry outlook, our supply and demand projections remain consistent with what we shared last quarter. DRAM industry supply bit growth is expected to be about 20% in calendar 2018, and we expect a healthy market environment, driven by the ongoing enterprise data center, cloud, and mobile strength as we just discussed. We expect the industry bit growth for NAND to approach 50% in calendar 2018, as the industry continues to ramp 64-layer designs into volume production. SSD adoption in client computing and data center applications continues to increase and will expand further as more supply becomes available over time. Against that backdrop, projections for our own bit growth remain unchanged, we expect our DRAM bit growth to be slightly below the industry, and we expect our NAND bit growth to be somewhat above the industry for fiscal 2018. During fiscal 2018, we are focusing on technology transitions for both DRAM and NAND, without any additions to our total wafer capacity, and on improving our mix of high-value solutions to enhance our revenue share. For fiscal 2019 and beyond, we continue to assess scenarios for the fab clean room space required to implement technology transitions to future, more advanced DRAM and 3D NAND nodes. I’ll now turn it over to Ernie to provide details on our first quarter results by business unit.

Thank you, Sanjay. And thanks to all of you for joining the call today. We had a very strong start to our fiscal year, exceeding guidance across all financial metrics, driven by strong execution, a continued robust market environment, and further progress on our technology migrations. For FQ1, total company revenue was $6.8 billion, up 11% from the prior quarter and up 71% on a year-over-year basis. Non-GAAP gross margin expanded to 55%, up 4 percentage points from FQ4 and 29 percentage points from the first quarter of fiscal 2017. Non-GAAP operating margin was 46%, up from 41% in the prior quarter and up 35 percentage points from the year-ago period. We continue to prudently manage spending with non-GAAP operating expenses totaling $612 million for the quarter, up 2% from FQ4 with both SG&A and R&D remaining relatively flat quarter-on-quarter. Non-GAAP net income increased to 44% of revenue and totaled approximately $3 billion, or $2.45 per share. This performance compares with $2.4 billion or $2.02 per share in Q4, and $335 million or $0.32 per share from the year-ago period. Turning to performance by business unit: The Compute and Networking Business Unit reported FQ1 revenue of $3.2 billion, up 13% sequentially and more than double year-ago levels. Our record performance was driven by increasing server memory content, which drove higher sales to enterprise customers together with strong demand for graphics processing. Operating income was 60%, compared to 56% in FQ4 and 14% in FQ1 2017. Q1 Storage Business Unit revenues increased 7% sequentially to $1.4 billion, driven by strong growth in SSD sales. On a year-over-year basis, revenues were up 61%, driven by increasing market share in SSDs. In fact, sales of SSDs reached record levels in the quarter, with double-digit sequential growth across Consumer, Client and Enterprise and Cloud markets. SBU operating margins increased to 29% from 19% in the prior quarter, and negative 5 percent in FQ1 2017. These results reflect a higher value product mix and continued market acceptance of our TLC 3D NAND based products. The Mobile Business Unit reported $1.4 billion in revenue, up 16% sequentially and up 32% year-over-year. We are seeing strong acceptance of our LPDRAM products and continue to enhance our portfolio of managed NAND offerings. The solid demand environment, combined with the traction we’ve made with our latest-generation products, led to operating income of 37%, up from 31% in FQ4 and 9% in FQ1 2017. The Embedded Business Unit reported revenue of $830 million in FQ1, in line with the prior quarter and up 44% year-over-year. Operating margin was 41%, essentially flat from the prior quarter and up 10 percentage points year-over-year. As Sanjay noted earlier, we continue to see exciting demand trends across each of the underlying embedded markets with evolving end-market requirements ranging from high-performance memory required for autonomous driving, to ultra-high-density storage solutions for edge devices such as video surveillance cameras. We are focused on building upon our existing leadership position to capture these growth opportunities. Turning to results by product line: DRAM represented 67% of overall company revenue in FQ1. Demand for client PCs, solid exposure to new flagship smartphones, and ongoing strength from servers, particularly in cloud and hyperscale data centers, drove DRAM revenue higher during the quarter, up 13% sequentially and up 88% year-over-year. Sequentially, shipment quantities increased in the upper single-digit range, while ASPs increased in the mid-single-digit range. DRAM non-GAAP gross margin was 61.5% in FQ1, up 2 percentage points from the prior quarter and up 33 percentage points from the year-ago quarter. Revenue from trade NAND increased by 2% sequentially and represented 27% of overall company revenue in FQ1. Trade NAND revenue was up 47% year-over year, driven by our strong growth and market share gains in the SSD market and robust demand from the mobile and embedded markets. On a sequential basis, shipment quantities increased in the mid-single-digit range, while ASPs declined in the low single-digit range. Trade NAND non-GAAP gross margin was 49% in FQ1, up 9 percentage points from the prior quarter and up 26 percentage points from the year-ago quarter, reflecting a richer mix of sales into high-value end markets. As Sanjay noted in his prepared remarks, we are making strong progress on the roll-out of our 1X DRAM and 64-layer 3D NAND deployment. The roll-out of these technologies will enable meaningful levels of ongoing cost-per-bit reduction as we make progress throughout fiscal ‘18. For DRAM, our bit output growth will be more heavily weighted to the first half of the fiscal year, while NAND bit output growth will be relatively greater in the second half of the fiscal year. The company generated operating cash flow of $3.6 billion in FQ1, compared to $1.1 billion in the year ago period. During the quarter, we deployed $1.9 billion for capital expenditures, net of partner contributions. We continue to expect FY18 CapEx in the range of $7.5 billion plus or minus 5%, fairly balanced between the first and second halves of the fiscal year. Free cash flow for the quarter was $1.7 billion, compared to negative free cash flow in the year-ago period. We continue to pursue our plans to strengthen our balance sheet and lower debt. During FQ1, we raised $1.4 billion from an equity offering and repurchased or converted $2.4 billion in principal amount of our debt. Total face value of debt was $9.3 billion as of the end of FQ1, and we currently expect to exit FY18 with approximately $8 billion in face value debt. We expect the interest savings from these de-leveraging actions, combined with higher interest income from larger cash balances and the anti-dilutive effects of settling converts for cash, to materially offset the dilutive impacts associated with the equity offering. Exiting FY18, we foresee non-GAAP net interest expense of $25 million $30 million per quarter versus $100 million per quarter in FQ4 of ‘17. We ended the first quarter with cash, marketable investments, and restricted cash of approximately $6.6 billion and continue to see the opportunity to exit FY18 in a positive net cash position. Moving on to guidance for FQ2 2018, on a non-GAAP basis, we expect the following: Revenue in the range of $6.8 million to $7.2 billion; Gross margin in the range of 54% to 58%; Operating expenses between $625 million and $675 million; Operating income ranging between $3.25 billion and $3.45 billion, and; EPS ranging between $2.51 and $2.65 per share, based on 1.241 billion diluted shares. Finally, a word about tax reform, as drafted, the legislation would have no significant impact to our FY18 tax rate, which we continue to expect to be in the mid-single-digit range. In FY19 and beyond, we would expect some impact to our non-GAAP tax rate, with an offsetting benefit of more flexibility in deploying our global cash balances. As further clarity around this legislation develops, we will provide appropriate updates. With that I will turn it back to Sanjay.

Thank you, Ernie. As we close out calendar 2017 and look to 2018, we see increasing opportunities for Micron to play a larger role in the technology trends shaping modern life. We will be hosting an analyst conference in May, where we plan to elaborate on our view of these trends and how Micron envisions our technologies shaping the world in the years to come. We believe that our technologies, capabilities, and team talent place us in a unique position in the market. Memory and flash storage are strategic assets that put Micron at the intersection of the biggest growth trends in technology, and we cannot be more excited about our future. Our customers increasingly view us as an essential partner in early design discussions due to the differentiation our solutions can provide. We are focused on increasing this value, and I look forward to sharing the results of that focus throughout 2018. We will now open for questions.

[Operator Instructions] Thank you and our first question comes from the line of Srini Pajjuri with Macquarie Securities. Your line is open.

Thank you and congrats on a great quarter guys. A couple of questions on the NAND side, Sanjay at least among the investor community there seems to be some concerns that there is going to be a flood of supply coming online into the industry in the first half. I am just wondering if you have any thoughts about what your view is about the supply-demand balance as we head into the first half? And then for Ernie, the NAND gross margin improvement is 900 basis points sequentially, I know you mentioned mix healthy there, but I am just curious if you could provide some more color as to exactly what’s driving that and how sustainable that is going forward?

So in terms of the industry demand-supply environment, let me just say that, with respect to the industry supply growth in the calendar year 2018 we have said that the industry supply growth would be approaching 50%. While that is more than what the calendar year ‘17 supply growth is, it is in the range of 35% to 40%. The overall what you have to look at is the demand requirements. In calendar year 2017, of course supply has been tight, there has been pent up demand particularly in the areas of client SSDs and the client PC, notebook PC environments. The march toward high density SSDs in notebook PCs was somewhat slowed down given the overall tightness of supply in calendar year 2017. When I look ahead at calendar year ‘18, I see strong demand trends with respect to attach rate of SSDs and client PCs continue to increase the applications related to cloud and data centers, enterprise data centers continuing to drive higher average capacities usage in cloud and data center applications as well. And certainly mobile applications also the average capacities of NAND content continue to increase. And these are all increasing because of the trends, right? I mean, attach rate in notebook PCs in calendar year ‘17 about 35%, going toward over the course of next several years by 2020 timeframe getting to 75% attach rates. So lot of HDD that is still to be replaced with SSDs and same trends are continuing in all markets for NAND average capacities are continuing to increase. So the demand trends continues to be very robust flash, yes, more supply environment. But we are very focus on continuing to strengthen our product portfolio and increased our share with respect to SSD markets as well as with more managed NAND solutions to address the mobile markets.

And relative to the question about margins, I think there are two pieces of that. One, we are making progress toward a mix -- higher value added mix of solutions and those typically carry the opportunity for higher margins. Also we continue to make progress on the cost side with respect to increasing amount of 3D NAND, as well as TLC and you had a good quarter for cost reduction as well. So it’s a combination of both market facing with the higher value add solutions and a good quarter from an operational perspective relative to costs.

Thank you. And then maybe just one follow-up Ernie on the balance sheet, I think you previously said your gross debt target is about $8 billion, which you said you are going to reach I guess next quarter. Given assuming that the tax bill will pass, is there any change to that target? And then if you can talk about again assuming that the tax bill will pass, what’s your priorities for cash going forward?

Yes, I don’t think we have said that we’d reach that gross debt target in FQ2, we said by the end of our fiscal year. And so I wouldn’t expect that we would achieve that level in fiscal Q2. The tax bill, as we understand it would not necessarily change the priorities, which are always to continue to make sure we have the best technology we can in production and to be able to transition that at a time that make sense for us. And then for fiscal year ‘18, certainly getting the balance sheet in shape relative to these aggregate levels of debt.

Thank you.

Thank you. And our next question comes from the line of Harlan Sur with JPMorgan. Your line is open.

Good afternoon and congratulations on the solid execution. On the DRAM side you’ve got Intel ramping sky lake, you have AMD ramping EPYC on the server CPU side, cloud and hyperscale CapEx spending looks to grow about 30% next year. Given all of these, do you guys anticipate continued momentum and mix in your server and cloud business in DRAM to continue to move higher next year? And then from an industry perspective does the server and cloud segment become a bigger part of the DRAM consumption mix over the next two to three years overtaking mobile?

Yes absolutely, cloud and server, data center enterprise as well as cloud, data centers absolutely will continue to be the biggest growth drivers at large volumes for the DRAM industry and we have very strong penetration in these markets. And actually we do expect to continue to build momentum in these markets going forward.

Great, thanks for that. And Sanjay you talked about semi and fully autonomous wins in the quarter. All of the major auto OEMs and subsystem guys are focused on this, we talk to one of the leading guys that focused on sensors and processors technology required for these types of vehicles, they are talking about 25-30 gigabytes of DRAM and 1 terabyte SSDs per car for level 4 and level 5 fully autonomous. That’s a pretty significant step up in DRAM and NAND content, is that consistent with how you see the content trends in automotive over the next kind of three to five years?

It’s absolutely is. I mean, when you look at autonomous vehicles, they really are level 5, autonomous vehicles in the future are projected to have about 40 gigabyte of DRAM content in them. And when you think about it that is very similar to the average capacity that are associated with servers in the server workstations, right. So these cars will be really very powerful computers in the future and they are not only going to be driving tremendous amount of DRAM content usage, but they will also drive NAND flash usage. They will be generating using data to make millions of real time split second decisions to make sure that the passengers in the autonomous vehicles can be transported effortlessly and safely to their destination. There will be sensors from sonar to cameras that will be generating billions of signals and all of that data will have to be processed, accessed in order to make fast decisions. So you are absolutely right to note that this is really a secular trend here in front of us in terms of driving continued usage of memory and storage. Earlier we talked about cloud and data center applications and again those are growing faster than the rest of the industry, autonomous vehicles will be another big driver in the future.

Thank you.

And Micron is uniquely positioned with the strong portfolio of solutions both with its flash solutions as well as DRAM memory solutions to address these fast growing market trends.

Thanks, Sanjay.

Thank you. And our next question comes from the line of Blayne Curtis with Barclays. Your line is open.

Hey, guys. Thanks for taking my question and clearly congrats as well. Couple of questions, maybe on DRAM, can you just go back to the servers been a huge driver and that's really been without Intel’s clearly still the minority units. Can you just maybe talk about content per server, how much that increased this year and then as you look into next year how much of the driver is clearly as you look into second half and that being more meaningful?

So in terms of content per server I think 2017 if you look at flash attach rate it's about average capacity around 2,500 gigabytes. And when you look at projection over 2018 timeframe growing to anywhere above 3,000 gigabyte average capacity continue to march ahead by 2021 timeframe almost tripling from the 2017 levels well above 8,000 gigabytes as well. So that is on the SSD side, and similarly on the DRAM side, average capacity is continue to increase nicely there as well. In 2017 timeframe about 145 gigabyte per server average capacity estimated and industry reports are showing that by 2021 timeframe going to about 350. So, very strong solid and year-over-year increases not just in near-term, but again as I say these are secular trends here.

Thanks for that. And just on the NAND side, I just want to go back that part of question, obviously you are seeing a across over of 64, how do you think about the cost trajectory here as you then start to ramp 3rd gen as you look to the fiscal year is there any sort of -- what's the slop of that costs curve as one is going up and the other one is going down?

As we introduce 3rd generation 3D NAND, we would continue to expect that that technology of mature yield have or has favorable cost dynamics relative to 64 layer. We don't believe that it will have a significant positive or negative impact in our FY18 results as we will just be implementing that as we get into the last quarter of the fiscal year. So, it won’t be material enough either way to change the fundamental dynamics that we're going to see this year, which is largely driven by our 64 layer TLC NAND.

Thanks, Ernie.

Thank you. And our next question comes from the line of Kevin Cassidy with Stifel. Your line is open.

Hi, thanks for taking my question. You had mentioned when you gave your guidance for CapEx for the year about adding more value added into your DRAMs and Flash. Can you say what percentage right now would you call in the value added group and what's the goal for that?

In the value added solutions category we count our SSDs in that and in fiscal Q1 we had really solid increase in our SSD revenues, we gained shares in the market as well. In the mobile space, managed NAND solution are also and discrete NAND solutions, we consider as value add as well. Our share in those markets today is relatively low, low single-digits, but we see tremendous opportunity as we are continuing to diversify our product portfolio. And as we execute on that product roadmap in the quarters to come, we expect to be gaining -- making substantial progress. And of course in the DRAM side, applications such as automotive, such as graphics, high performance applications they all contribute toward the value add solution. We are not providing any specifics at this point. We'll discuss more details at our Analysts Day in May.

Okay, great. And maybe just as one follow-up is because especially DRAM market had been such a commodity market, are you getting any push back from your customers that they want you to move more towards the commodity?

And let me just point out in response to your previous question that while we will provide more detail in May, but on a year-over-year basis, certainly we are increasing our value-add solution mix substantially and very pleased with the progress that we are making. Regarding your second question on DRAM, DRAM really is a strategic enabler today in diversity of markets and the mega trends we talked about whether it is data centers, cloud applications being the few for AI engine helping make decisions for various search algorithms as well as various activities in all verticals that are being pursued leveraging AI in mobile space as more and more applications go toward augmented reality type of features in mobile requires more DRAM, DRAM is highly strategic. Even in the future when you look at notebook computers, they will require the form factor and the low power aspect, even in PCs of LPDRAM in future years to come. So you see and of course we just talked about automotive as well. So now DRAM is addressing diversity of markets, I mean, several large diverse markets product portfolio meeting the needs of these markets is becoming differentiated and when that happens that always gives you stronger opportunities to drive profitable growth in that market. And our customers when we have an engagement with them, when we have dialogues with them they are talking to us how memory, DRAM memory is really now helping them solve the bottleneck in their applications. So DRAM today is very different, as I said many times before from the DRAM in the PC era only or when it was about just PC and mobile, today it is about many more applications of DRAM. And really providing a very strategic enabling role in creating all these applications that are truly transforming the world right before arrives.

Great, thank you.

Thank you. And our next question comes from the line of David Wong with Wells Fargo. Your line is open.

Thanks very much. I’, not sure that you said it, but on the November quarter, how much of your total NAND production was 3D NAND? And as you transition to 64-layer in the February and May quarter next year, did your NAND production drop or grow sequentially?

So about 80% of our output for the quarter just completed was on 3D and over the course of the balance of fiscal ‘18 we would expect that number to grow to about 95%. We expect that we’ll have some measure of bit growth as we said each and every quarter we said it would be in the prepared remarks little more heavily weighted to the back half of the year.

But you never let will grow in the first half?

Yes.

Okay, great.

Yes, that output for the quarter was up in the low single-digit range.

Excellent, thanks very much.

Thank you. And our next question comes from the line of C.J. Muse with Evercore. Your line is open.

Yes, good afternoon. Thank you for taking my question. I guess first question on DRAM, you talked about 20% bit supply for the industry in ‘18, which would suggest continued tightness throughout the whole year. So curious how customers are reacting to that reality, what kind of visibility are you seeing to pricing, as well as extension in contracts?

We work very closely with our customers and we certainly work hard in terms of gaining visibility to their future requirements of DRAM and how their applications are shaping up so that we can make sure that our technology and product roadmap is adjusting their needs. And this varies from customer-to-customer. Some customers tend to be more on a month-by-month basis, some more quarter and some certainly longer term engagements as well. And you are right to note, that yes, I mean, the industry supply growing about 20% and demand likely to be somewhat higher, we do expect a healthy DRAM industry supply and demand environment and continue to work very closely to drive strategic growth of our revenue mix going forward.

Very helpful. And I guess as my follow-up on the NAND side you are talking around 50% bit growth or approaching that for that industry one or two of the other players in the market are expecting lower type of bit growth. So it would appear that you are making assumptions around the 64-layer ramp that might be a little bit more robust than your competitors. So curious how are you thinking about that ramp? And clearly I think you would agree the risk is probably lower than 50%, how would you kind of I guess put a probability around where you think it truly ends up?

So of course in terms of projecting, our industry bit growth estimates here. We of course are taking into consideration our estimate of the ramp of 64-layer technology in the industry and our assumptions around the ramp of wafer production as well as yield ramp of that technology. And along with other mix of technologies in the industry as well, but you are right to note that the dominant factor of the supply bit growth in the industry will be with 64-layers here. And we continually look at our estimates and we review it based on all the intelligence that we may have, as well as we may collect from third-party reports we refine our models on an ongoing basis, and if we have any changes on this we will share them with you.

Thank you very much.

Thank you. And our next question comes from the line of Joe More with Morgan Stanley. Your line is open.

Great, thank you very much. I wonder I know you don’t want to get too far ahead yourself on this, but when you get to sort of net cash neutral and you get to investment grade, what’s next from a prioritization of free cash flow. And how do you just qualitatively think about cash return versus maintaining cash for a more strategic usage?

I think that we will continue to prioritize the supporting our latest generation technology in production as we talked about for fiscal ‘18 clearly focused on getting aggregate debt down in that $8 billion range. Those things will lead to we believe overtime improved ratings. But I think how we think about deploying cash is not wholly dependent upon the achievement of any particular ratings grade. We just think these are things that will be worked in parallel. And at the appropriate time I think we are very open to thinking about broader uses of cash including shareholder return programs. So those do have a place in the hierarchy of thinking about uses of our cash. But until we get to the point where we have completed or substantively completed our work on this technology deployment and regaining cost competitiveness and also getting our aggregate debt levels to the -- to a level that we are more comfortable with certainly those will continue to remain our priority.

Great. And then I wonder the disparity between the 60% operating margins in the compute business and mobile which is still quite high, I guess to meet sort of you can see that PC DRAM still quite a bit more profitable than mobile. How do you think about that going forward in terms of your allocation of mix obviously mobile gives you more stability down the road versus the upside that you can get from PCs, how you are thinking about that balance?

Yes, don’t forget that the Compute and Networking Business Unit includes a wide array of products, which includes both clients but also includes data center, mobile -- not mobile but networking and other products as well. So it isn’t just a statement about the client compute environment and many of those sub segments within the Compute and Networking Business Unit we would also consider strategically important these are relationships where we have the opportunity to deploy both DRAM and NAND. And certainly mobile continues to be important. So I wouldn’t necessarily presume that the margin performance out of our Compute and Networking Business Unit is largely driven by the performance of clients.

Thank you very much.

Thank you. Our next question comes from the line of Romit Shah with Nomura/Instinet. Your line is open.

Yes, thank you. Sanjay since some of your equipment, semi cap equipment suppliers have -- it just seems like over the last three months have been raising their spending forecast for 2018 driven impart by higher DRAM investments. So my question is what’s the risk that your competitors perhaps are going more aggressively for share and that the 20% growth assumption you laid out for next year may actually end proving conservative?

I think the gain I would estimate today for DRAM as well are based on all the information from various sources and reports that we gather and we have put that here. I do think that even when you look at that estimates by other suppliers they are all pointing to fairly close tightly aligned industry supply growth estimate here. And -- but this is our estimate, we will as I said before for NAND same thing, we routinely look at this very, very closely. And just also remember that the technology transitions in the industry today certainly are becoming increasingly complex and they have greater capital intensity, as well as with the advance technology nodes you are actually getting less bits on a per wafer basis even at mature yields compared to prior generation nodes. Again given the scaling challenges increased complexity. So, all of these trends are actually increased technology complexity and increased capital intensity have a moderating effect on the supply growth.

Thank you. And then just on the February quarter, I don't know if you guys look at your individual periods on a seasonal basis, but it seems like the February quarter typically is a down quarter and you are guiding to sequential growth off of a better than expected November period. What's driving the better than normal seasonality for you, is it pricing or other factors as well?

Yes, I think Sanjay mentioned earlier the diversification of the DRAM demand stream and certainly that is also true on the NAND side. And in the context of that diversified demand stream, for example, automobile tend to be far less seasonal than the client PCs and if you look at datacenter deployments among all of the key players there are certainly a specific pattern for customer. But when you put them all together, there is no macro seasonality pattern that emerges when you look at that. So, as we've transitioned to these diversified demand drivers the concept of seasonality I think has been redefined and is perhaps less impactive on a quarter-be-quarter basis than it may have been in the past.

Interesting, thank you.

Thank you. And our next question comes from the line of Jagadish Iyer with Summit Redstone. Your line is open.

Thanks for taking my questions. Sanjay, two questions, first, if NAND ASPs continue to decline from rising supply, do you expect cost reduction to keep pace with the ASP decline as we look at fiscal ‘18? And then I have a follow-up.

I mean, we are certainly not projecting pricing for fiscal year ‘18 here. But again what I'll point to you is that if you look at the history of NAND, increased technology capabilities, technology advances which leads to cost reductions ultimately enable opening up of many new market application and drive the elasticity in many of these markets as well and drive the demand environment. So, I consider it as healthy if the cost declines are greater than the price declines in the industry, that's a healthy industry environment. And while we don't get into the projection of again when we look at all the demand drivers that are ahead of us, I spoke about it repeatedly in the call today multitude of demand drivers and in each of these large market segments again whether it is about could data center, enterprise data center, mobile, smartphones, client notebook, automotive applications all of these are continuing to drive strong demand trends for NAND flash application.

Yes, I think the one thing I would add to that as well as if you think about our statements around industry bit growth which are approaching 50%; those would be associated typically with fairly healthy levels of cost reduction. We're not going to get into the specifics of what that might be. But it is important to remember that that bit growth usually would be accompanied by some good cost reduction as well.

Okay, fair enough. Then on the 3D XPoint, I was wondering how much are you year marking for spending in fiscal ‘18? And has there been customers who have identified and how should we think generally on a big picture about pricing there? Thank you.

So this is Ernie. The 3D XPoint CapEx and any associated R&D expenses, the CapEx piece is embedded in our non-volatile estimates for the year, which we said was between 35% and 45% of our total spending. And as we look at that product we’re going to have to see how the markets develop, as Sanjay mentioned before we provide any broader perspective on that.

Thank you.

Thank you. Our next question comes from the line of Steven Fox with Cross Research. Your line is open.

Thanks good afternoon. Two quick questions for me, first of all you mentioned the market share gains in SSDs and I assume some of that was related to fixing the component issue, but can you sort of give us some more color on where you think specifically you had most success in gaining share? And then secondly, I know you are not going to talk specifically about pricing for the upcoming quarter, but if you could maybe qualify some more how you get to your sales guidance in terms of pricing and mix that would be helpful. Thanks.

In terms of SSD share gain during the quarter, we believe we gained share both in client applications as well as enterprise and cloud data center applications. Really our products did very well there and at this point our offerings are with SATA drives and next year we will bring out our NVMe solution sometime during calendar year ‘18. And that should help us expand our portfolio and give us opportunities to continue to gain share in this market. And regarding our guidance that we have given you, of course we have certain assumptions related to total demand and our pricing assumptions are baked in there as well. But obviously for competitive reasons we do not get into discussion of pricing on the call.

Understood, sorry give it a shot. Thank you, congratulation.

Thank you, and our next question comes from the line of John Pitzer with Credit Suisse. Your line is open.

Yes, good afternoon guys, congratulations on the strong results. Ernie maybe I can ask the gross margin question around NAND a little bit differently. I am just kind of curious given that you guys had significant gross margin upside on pricing down low single-digit sequentially under a scenario where pricing continues to decline let’s say at a low single-digit rate sequentially, is there enough mix and cost opportunity for you to maintain and/or grow gross margins from here. How should we think about the parameters that you being able to maintain or grow gross margins relative to your cost and mix availability from here?

Yes, so if we’re framing the scenario in the future as we continue to see sort of marginally declining pricing environment couple of points per quarter or so I think there is ample room for gross margin expansion both as a result of our cost reduction as well as the ongoing transition to high value solutions, which would include redirecting -- continuing to redirect component sales into higher value added solutions as well as some of our partner contracts wind down over the course of our fiscal ‘18 and into 2019. So I think in that pricing scenario that you have set forth, we believe there is reasonable room for gross margin expansion.

That’s helpful. And a similar question on the DRAM front, you mentioned in your prepared comments that you expect more DRAM bit growth in the first half of your fiscal year than the second half. As we transition into the second half how much -- how many levers do you have on the mix side in DRAM to be able to progress sort of gross margins higher from here?

Yes, I think again the same general story with respect to continuing to move toward higher value add solutions. You heard Sanjay speak too for example increasing the breadth of our managed NAND offerings, if you look at server that certainly a very strong growth pattern here as well as automotive. So I think there are a number of segments where we would continue to expect to be able to optimize our bit output toward, that would give us the best gross margin opportunity. And of course even though our bit output growth is going to be lowering DRAM than NAND we would still expect to deliver some meaningful cost reduction there as I articulated in the prepared remarks.

Perfect, thanks guys. Congratulations.

Thanks.

Thank you.

Thank you. And we will take our final question from the line of Mark Newman with Bernstein. Your line is open.

Hi, thanks for squeezing me in. I had a question on CapEx and capacity. So Micron you stayed top of the CapEx quite a bit this year from previous years, but still quite significant under spending some of the ride was particularly Samsung. My question actually is about, how is Micron thinking about the need for fabs longer term, this is more of a kind of longer term question, because both Samsung and Hynix are building brand new greenfield fabs as we speak. What is the status right now for Micron in both DRAM and NAND in terms of how much space you have available to you in DRAM and NAND? And I understand your goal is to keep wafer capacity roughly flat and grow gigabytes from technology migration. At what point do you need more fab space to keep that spread capacity roughly flat as of course more layers requires more fab space and of course as you shrink DRAM you also need more fab space as well. So do you have any plans for new fabs what would be the plan and how you are thinking about that? Thanks.

So like you noted in order to realize the technology transition you of course do need space in your clean room to deploy the tools that are required to implement the advance technology nodes. So our fabs have had space here in terms of deploying the technology transitions for example in Singapore going from 32-layer to 64-layer and in the past in Singapore fab for NAND going from planer to 3D nodes as well. And similarly clean room space is something we look at very carefully for DRAMs as well because as I just said, I mean, more clean room space is needed for technology transition. So as I said in my prepared remarks, I mean, we do continue to assess the scenarios that are needed for future technology transitions and have continue to assess our clean room space requirements. Fiscal year ‘18, we are not looking at any new wafer capacity additions, we are not looking at any new -- any meaningful new clean room space required to implement our technology transition plans in fiscal year ‘18. As I said in my remarks going beyond that we continue to assess various scenarios to implement the requirements for technology and transitions both in DRAM as well as in NAND. No plans firmed up yet, and as and when we look at it we will share these things. Remember any time you build a new clean room it takes several months to build -- construct the building and it takes few months than to roll in equipment into the new clean room and to deploy the equipment, qualify it and build products. So these are things that are to be looked at for the longer term horizon and we are evaluating those scenarios.

Okay, thanks very much.

Thank you. And this concludes today’s question-and-answer session. Ladies and gentlemen, thank you for participating in today’s conference. This does conclude the program and you may all disconnect. Everyone have a great day.