Micron Technology, Inc. (MU) Q4 2016 Earnings Report, Transcript and Summary

Good afternoon. My name is Abigail and I will be your conference facilitator today. At this time, I would like to welcome everyone to the Micron Technology’s Fourth Quarter 2016 Financial Release Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks, there will be a question-and-answer period. [Operator Instructions] Thank you. It is now my pleasure to turn the floor over to your host, Ivan Donaldson. Sir, you may begin your conference.

Thank you. And welcome to Micron Technology’s fourth quarter and 2016 fiscal year financial release conference call. On the call today with me are Mark Durcan, CEO and Director; and Ernie Maddock, Chief Financial Officer. This conference call, including audio and slides, is also being webcast from our Investor Relations website at investors.micron.com. In addition, our website contains an earnings press release filed a short while ago and supplemental information including quarterly operational and financial metrics and guidance, GAAP to non-GAAP reconciliations, slides used during today’s conference call, and a convertible debt and capped call dilution table. Today’s call will be approximately 60 minutes in length. A webcast replay will be available on our website for one year. 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 also follow us on Twitter @MicronTech. As a reminder, the matters we will be discussing today, includes forward-looking statements based on the environment as we currently see it. 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 the Company files with the SEC, specifically our most recent Form 10-K and Form 10-Q for a complete discussion of these important risk factors and other 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 call to conform these statements to actual results. I’ll now turn the call over to Mark.

Thank you, Ivan. For fiscal Q4 2016, Micron posted total revenue of $3.2 billion with gross margin of 18% and a non-GAAP net loss of $56 million or $0.05 per share. Both revenue and gross margin were at the high end of the guidance range and together generated earnings per share that exceeded guidance. Operating cash flow was $896 million. On the last call, we reported that while we were seeing signs of improvement in selected markets, we were uncertain of the extent and the duration of the improvements. During the quarter, we continue to see market conditions improve driven by both supply and demand fundamentals. Industry inventory appears lean and the current market outlook for both NAND and DRAM continues to be positive. Currently prices are increasing in number of segments, stabilizing in others and we’re seeing fewer segments showing residual declines. This environment frames our thinking and the guidance that Ernie will provide a little later on. We made significant progress on a number of key initiatives in the quarter including the completion of the mobile product qualifications that we’ve been discussing in our early 1X nanometer DRAM deployment. We’re also seeing the benefits of the 3D NAND yield ramp, which generated stronger than expected NAND bit growth in the quarter. I’m going to provide a high level overview of each of the business units and Ernie will follow with more details on each business’s financial performance. In our compute and networking business unit, we experienced accelerating revenue growth in the face of moderating pricing environment. Revenue growth was driven by continued increases of 20-nanometer shipments as well as a significant transition to DDR4 and the cloud and client segments where shipments exceeded DDR3 for the first time. We saw ongoing strength to graphics driven by GDDR5 and 5-X products, which supported NVIDIA’s introduction of the 5-X based TITAN X, the highest performance consumer graphics card in the market. In the enterprise segment, Micron was recognized at the Flash Memory Summit as the innovation leader in non-volatile DIMMs, an important solution of high performance servers. Turning to our mobile business unit, China is becoming a more significant growth driver in mobile devices. Most of these devices are differentiated by large amounts of memory including up to 6 gigabytes low power DRAM. Additionally, the top of the line Apple and Samsung models now ship with 256 gigabytes of NAND and we see mid range Chinese handsets heading in the same direction in order to compete. Although there is a moderating pace of smartphone unit growth, significant content increases of both NAND and DRAM will generate strong overall bit growth. During the quarter, we introduced four new mid to high density mobile 3D NAND products with positive reviews from the press and from our customer base. Mobile DRAM and eMCP price declines moderated in the quarter with more recent signs of stabilization and increases. Revenue growth in our embedded business unit has been driven by a record automotive business and increased performance in the consumer and connected homes segments as well as the strengthening industrial multimarket business during the quarter. We have maintained our leadership position in automotive through strong customer support and the introduction of new products that meet the fast growing memory content requirements for infotainment, ADAS and instrument cluster applications. High temperature DRAM and automotive grade NAND products are enabling a strong design win pipeline. The introduction of several new industrial solutions has positioned us to grow our share within IMM and retain our high market share in machine to machine communication modules as the industry migrates from 2G to 4G and LTE. We’ve also been successful in achieving strong market share positions in consumer applications from action cameras and home automation to high volume markets including digital TV and set top boxes. Finally, our storage business unit is continuing to transition its product line to our leading edge 3D NAND technology. We’ve seen positive response from customers this quarter related to our refreshed SSD portfolio which offers a breadth of solutions to meet diverse application requirements. We’re finalizing qualification with eight major OEMs on our 1,100 client drive and are engaging with cloud customers on future designs. In the enterprise, we’re seeing companies begin to retrofit HDD infrastructures with SAS SSDs as they modernize existing investments. Our customers and demand executives like financial services and energy are moving stores closer to the server. We have seen significant interest in our PCIe NVMe drives in these applications due to their ability to deliver very consistent high performance. We’re actively extending TLC 3D NAND across our cloud and enterprise portfolio and expect to introduce our first TLC 3D SATA drive for these markets later this year followed by a server focused version early next year. We expect to see positive supply and demand dynamics in the SSD market for 2017. Turning to the memory industry more generally, we’ve seen further evidence that DRAM wafer output is declining as a result of lost throughput related to the 20-nanometer and 1X nanometer conversions. Absent some replacement of these wafers, we could see industry supply growth as low as mid-teens in 2017. As some of lost wafer output is replaced, industry supply growth could be in the high-teens percent range. This compares to our long-term bit demand growth forecast in the low to mid 20% range. For NAND, we estimate 2017 industry bit growth in the high 30% to low 40% range, which is in line with 2016. This compares to our long-term bit demand growth forecast in the low to mid 40% range. Despite significant investments in 3D conversions across the industry, we believe that 2017 supply growth will be relatively balanced with demand given the disruption in the fab output related to these conversions. For our operational priorities for fiscal 2017, or our operational priorities for fiscal 2017 continue to be targeted at ramping advanced technologies with an added focus on building out a more robust product portfolio, in particular for non-volatile memory products. For DRAM, we begin the 1X nanometer ramp and expect to have meaningful output by the middle of 2017. At our winter Analyst Day, we shared two-year bit growth targets for DRAM and we are on track to meet to these targets. We were slightly below the range in fiscal 2017 (sic) [2016] and expect to be above the range in fiscal 2017. All of the DRAM bit growth will come from technology transitions and we currently have no plans to add wafer capacity. We expect our fiscal 2017 DRAM cost per bit to decline 20% to 25%, as a result of completing the 20-nanometer migration and initiating the 1X nanometer conversion. Our cash cost for bit declines will be meaningfully higher than this range. For NAND, we will continue to focus on ramping our Gen 1 3D as well as TLC. As we updated at our last Analyst Meeting, 3D bit output will exceed 2D output in the current quarter ahead of our initial target. We will also be working on deploying second generation 64 layer 3D technology. Similar to DRAM, our 2016 NAND bit growth came in below the two-year CAGR shared at our analyst event and we therefore expect 2017 to be above that range. We are forecasting our fiscal 2017 NAND cost per bit to decline 20% to 25% as a result of the 3D and TLC conversions. This cost per bit includes the impact of our planned build out of SSDs, eMCPs and Managed NAND solutions all of which carry additional bill of materials costs but will also enable a richer ASP mix. On a like-for-like basis, cost per bit will decline more substantially. Relative to 3D Xpoint technology, we are working with market enablers on adoption of our QuantX solutions and continue to believe this innovative technology will be an important contributor to Micron’s future success with initial revenue later in 2017. Now, I would like to turn the mic over to Ernie.

Thank you, Mark. We had a solid end to our fiscal year as our continued focus on execution coupled with a more favorable business environment resulted in improved financial performance. I’ll begin my comments today with some technology and business unit details followed by an overview of the Company’s results for the quarter and guidance for our first fiscal quarter 2017. DRAM represented 60% of our total revenue with the following segmentation. Mobile represented about 25%, similar to the prior quarter; the PC segment was in the upper 20% range, up slightly from the prior quarter; server business was in the high teens percent range, down from the prior quarter; and specialty DRAM, which includes networking, graphics, auto and other embedded technologies, was in the low 30% range, up from the prior quarter and primarily driven by an increase in graphics which represented more than 10% for the quarter. In our non-volatile memory business, trade revenue represented 31% of total revenue with the following segmentation. Consumer, which includes memory cards, USB and components, represented about 50%, down slightly from the prior quarter; mobile was in the high teens percent range, up from 13% in the prior quarter as we began to see the impact of our completed customer quals. As a remainder eMCPs are primarily in the Mobile segment. SSDs represented 13% similar to last quarter and the automotive, industrial, multi market and other embedded applications were in the high-teens percent range similar to prior quarter. Turning to performance by business unit, the compute & networking business unit delivered fiscal Q4 revenue of $1.25 billion, up 14% sequentially, primarily driven by 20-nanometer shipment growth accompanied by strengthening demand in a moderating pricing environment across all segments. The non-GAAP operating loss was $7 million or less than 1% of revenue. In the enterprise segment, conversions to DDR4 are now largely complete and we are focused on enabling our solutions for the next generation server platform. We are seeing continued interest in both higher density RDIMMs and LRDIMMs as well as NVDIMM solutions. The cloud component of the enterprise segment is growing quickly driven by continued growth in density across both DDR3 and DDR4 solutions. In graphics, we saw a growth well above typical seasonality, driven by our GDDR5 and GDDR5-X products. There are additional G5-X based product announcements from NVIDIA in both the graphics and workstation segments during the quarter. In networking, we enjoyed double digit revenue growth with strength across all regions and growing interest in applying the high bandwidth capabilities of GDDR5 and GDDR5-X to the networking segment. Finally within the client segment, we saw market demand which exceeded our expectation as well as a continued transition to DDR4 which for the first time represented a majority of client shipments. The mobile business unit delivered fiscal Q4 revenue of $671 million, up 20% sequentially, driven by a strong quarter in our eMCP products. The non-GAAP operating loss was $45 million or 7% of revenue and was partially impacted by the consumption of higher cost early production inventory accumulated over the last two quarters. We saw significant growth in the recently qualified higher density MCPs and as smartphone OEMS are positioning their products based on higher memory density specifications, which is helping to accelerate adoption of smartphones with richer memory content. The embedded business unit delivered fiscal Q4 revenue of $513 million, up 5% sequentially. Non-GAAP operating income was $133 million or 26% of revenue. The results were primarily driven by record automotive and increasing consumer business combined with some recovery in our industrial multi-market business. The automotive business delivered solid results with revenue up 6% sequentially driven by strong and increasing demand for both DRAM and eMMC solutions for infotainment, instrument cluster and advanced driver assistance systems applications. European, Korean and Japanese customers continue to fuel this growth and our portfolio of leading edge automotive grade solutions is continuing to enable platform design wins. The industrial and multi-market business increased 16% sequentially, primarily driven by the ramp of our NAND solutions into the Japanese amusement market. In addition, we continue to see strong demand for our NOR and NAND based MCPs used in machine-to-machine wireless communication modules. Consumer and connected home revenue was up 7% sequentially with solid demand for MCPs used in action cameras and home automation applications. We continue to see this demand pattern continue in the current fiscal quarter, which typically sees strong seasonal demand. Our 20-nanometer DDR4 products continue to ramp into set-top boxes and IMM applications. The storage business delivered fiscal Q4 revenue of $758 million, up 5% sequentially. The non-GAAP operating loss was $69 million or 9% of revenue. During the quarter, we entered production and OEM qualification of TLC 3D NAND-based SSDs in the clients and consumer segments, providing customers with a refreshed 3D-based portfolio as storage products expand the demand spectrum in these markets. In the enterprise SSD segment, consecutive quarter bits sold were up 45% based on higher sales of our planar MLC based cloud drive. We are deploying 3D TLC across the enterprise nm cloud portfolio with several product launches over the next two quarters. Moving on to overall company results, revenue for the fourth fiscal quarter was $3.2 billion at the top end of our guided range and up 11% sequentially. Fairly significant increases in volume shipments for DRAM were offset by moderating ASP declines, while trade NAND shipments increased as a result of early success of our 3D ramp and we experienced generally stable ASPs. Gross margin for the quarter was 18% also at the top end of our guided range. The non-GAAP net loss for the quarter was $56 million or $0.05 per share, significantly better than the guidance. For the full fiscal year, we achieved non-GAAP profitability despite our technology transition and memory market pricing pressure achieving non-GAAP net income of $66 million or $0.06 per share. As a reminder, Micron included both amortization of acquisition-related intangibles and stock compensation expense in our fiscal Q4 non-GAAP results. Taken together, these two items represent $0.04 a share in the fourth fiscal quarter and $0.20 per share for the full 2016 fiscal year. Turning to results by product line, DRAM revenue increased 13% compared to the prior quarter, as a result of a 20% increase in bit shipments, partially offset by a 6% decrease in ASPs. DRAM gross margins for the fourth quarter increased two percentage points to 20%, primarily driven by strong cost reductions as a result of the continued 20-nanometer ramp. Our non-volatile trade revenue increased 12% compared to the prior quarter, reflecting a 13% increase in bit shipments, partially offset by a 1% decrease in ASPs. Gross margin decreased a couple of percentage points to 16% as cost per bit was up related to the 3D ramp and build out of higher cost storage and mobile solutions to support future growth. Non-GAAP operating expenses for the quarter were $559 million. This was below our guidance, primarily due to continued expense control and the timing of prequalification expenses for the fourth quarter, some of which will now occur in fiscal Q1. The Company generated operating cash flow of $896 million, which included strong quarter for collection activity and we ended the quarter with cash and marketable investments of approximately $4.8 billion. In the fourth fiscal quarter, capital expenditures net of partner contributes were approximately $1.7 billion. Before I move on to our guidance for the upcoming quarter, I want to share some changes to our reporting. The first of these changes relates to the depreciation schedule for our DRAM capital equipment. Previously, this equipment had been depreciated over five years. However, given the longer intervals between technology transitions, we have changed the depreciable life of our DRAM capital equipment from five years to seven years. This change will reduce depreciation by approximately 100 million per quarter on a going forward basis. The depreciation schedule for NAND related equipment remains unchanged. Additionally, and to be consistent with the majority of semiconductor companies who report non-GAAP results, Micron will exclude stock-based compensation and the amortization of acquisition related intangibles from our non-GAAP reporting. These expenses on average are approximately $50 million per quarter. Moving on now to our guidance for the first fiscal quarter, which we developed in the context of the market environment that Mark described earlier, on a non-GAAP basis, we expect the following: Consolidated revenue in the range of $3.55 billion to $3.85 billion; gross margin in the range of 23% to 25.5%; operating expenses between $600 million and $650 million due to the resumption of variable compensation expense and the higher pre-qual expenses that we spoke of earlier. Taken together, these items represent a sequential increase of approximately $80 million of which $25 million resulted from the timing of pre-qual between fiscal Q4 and fiscal Q1. Operating income in the range of $245 million to $330 million, and EPS ranging between $0.13 and $0.21 per share based on 1,046 million diluted shares. Relative to our capital spending plans for fiscal ‘17, as we shared last month, we expect to spend between $4.8 billion and $5.2 billion net of partner contributions. Between 40% and 50% of the total will be allocated to DRAM; between 30% and 40% to non-volatile memory; and between 15% and 25% to technology and product enablement. Consistent with our public comments, this level of CapEx allows us to appropriately fund our technology investments while achieving neutral to positive free cash flow generation in fiscal 2017. Operationally, we are on track to deliver the bit growth and cost per bit reduction targets that we previously shared as we complete the 20-nanometer DRAM conversion, reach crossover of first generation 3D NAND this quarter and commence the ramp of 1X nanometer DRAM and second generation 3D NAND in 2017. For FY17 specifically, we expect DRAM and NAND cost per bit reductions between 20% and 25%. Finally relative to the Inotera transaction, Micron and Inotera have concluded that sufficient progress has been made in Inotera addressing the issues that caused the delay in closing. An Inotera board meeting has been scheduled for October 11th and the directors of Inotera are expected to set a share swap record date at that meeting. We anticipate that the share swap record date will be set for the first half of December of 2016. With that I’ll turn it back to Mark.

Thank you, Ernie. To summarize, we faced challenging market conditions in fiscal 2016 and envision to going through a transition period of our technology migrations. Despite these headwinds, we ended the year with positive non-GAAP EPS. Compared to prior cycles, this represents a material improvement in our performance and further reinforces our belief in structural industry improvements. We are now experiencing a more positive and improving environment, and Micron is committed to making even further improvements in our relative performance with enhanced growth and cost reductions and expanded capability to deliver value added products to enable our customers. Okay, operator, we are now ready to begin Q&A.

Thank you. [Operator Instructions] Our first question comes from C.J. Muse with Evercore. Your line is open.

Good afternoon. Thank you for taking my question. I guess first question Ernie for you on the pricing front. I know you don’t want to go into specific details but curious if you think about November, what you’re embedding in terms of I guess changes to mix and perhaps any flow-through of lower margin inventory, and any other kind of factors that we should be thinking about as it relates to DRAM ASPs?

This is Mark. Maybe I’ll take that one C.J. Certainly relative to the sell-through of inventory, there is an impact there, given some of the inventory we built up over the last couple of quarters relative to the qualifications we mentioned. Relative to market assumptions, we’re not going to get into the detail as always of the embedded margin because we don’t want to forecast the ASPs. But I think it’s fair to say that we’re seeing some early signs of significant price movements. And as we’ve done in the past, we want to wait before we get ahead of ourselves in projecting what may come as we move further out into the quarter but I do think that the trend is positive and the bias is positive.

It’s really helpful. I guess as my follow up, how do you think about channel inventory today? And I guess this is another DRAM question. And then as part of that how do you expect customers to act as you look into 2017 in an environment where bit supply is sub-20% and demand presumably higher than that? How do you expect the kind of changes in behavior there? Thank you.

Certainly if we think about inventory in the marketplace, certainly channel inventory I think is very, very tight, maybe one to two weeks or less than a week. At the major customers, we have less precision relative to that but I’d say, it’s certainly down and there is a sense of urgency at the large OEM customers as well. In some cases, we may be looking at allocation of certain products in certain markets. And certainly the customers are beginning to ask for longer-term price commitments, which is in my mind always a good thing. And we’ll look at those on a case-by-case basis and potentially take customers up on that where appropriate but obviously always keeping the long-term interest of shareholders in mind.

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

Yes. I just want to come back to the guidance because it’s something that people are trying to sort of gauge given that last quarter at least relative to how you originally guided, you guys came in a lot better. And I know at the time that there was a lot of moving pieces and you guys wanted to give yourself cushion. So, could you talk a little bit about visibility and at least on the pricing front if we could get a sense just directionally how you are thinking about ASPs in the November period? And then I have a follow-up.

Romit, sure. This is Ernie. So, I think it’s important to acknowledge that although we beat our EPS guidance, we actually were well in the ranges of both our revenue and our gross margin guidance. And so, we did think about the opportunity that was ahead of ourselves as we were giving the guidance for the last quarter and we’ve done the same thing in essence this quarter as well. So, I think you can look at the range of our guidance and it generally contemplates the outcomes that we see. And obviously if things continue on a strong and favorable momentum, it’s reasonable to think that you’d end up at the upper end of both pieces of the range which would generate a result similar to this quarter. And I think consistent with Mark’s prior comments, we are continuing to see positive momentum. I do think it’s important to understand that that momentum has existed and has started to strengthen and so as a result of its actual impact in our fiscal quarter with part of it already completed and we’re in negotiations for others, I think the trap we want to avoid is presuming that whatever the latest pricing news is gets retroactively applied across the entire quarter because that’s certainly not the environment that we’re living in. But I do think we were pretty thoughtful about the revenue range. And obviously we’re going to work hard and try to do as well as we can but we’ve put a lot of thought behind the numbers we have just shared with you.

Yes, that’s helpful. And then, Mark, I have a strategic question for you which is you talk about one of the priorities being to just ramp advanced products in non-volatile memory but at the same time sort of implicit in the fiscal 2017 free cash flow guidance is that you -- at least the way I’m interpreting it is that you’ll continue to burn cash in NAND. And so I guess my question is at what point do you consider a partnership to lessen the period expense and the CapEx burden associated with running this business?

We’re always looking for ways to improve our relative competitive position. And we’ve used partnerships and strategic relationships extensively in the past. We’ll continue to evaluate that and again always with the long-term in mind. We are -- there is an embedded assumption in there. There are lots of opportunities out there today. And I think I would not try to dissuade you of that view of life. I would just caution you that we are going to be pretty careful and deliberate about making sure we take a long-term view relative to the relationships and the partnerships we engage in and we are doing something that really drives a long-term strategic benefit for the Company as opposed to a short-term fix, more supply in the marketplace.

Thank you. Our next question comes from Mark Delaney with Goldman Sachs. Your line is open.

First question is I mean you could update us on the Inotera transaction, and if that were to close, if you could help us to understand the potential accretion or dilution from that transaction? And then related to it, to what extent is Inotera closing already embedded in of the financial guidance that you gave for fiscal ‘17?

So, as we had in our prepared remarks, we do expect the transaction to close at this point, sometime in the middle of December. We did contemplate that timing in our guidance, so it doesn’t really have a dramatic impact on our FQ1 guide at all. And certainly, the transaction will continue to be accretive to gross margins and continues to be accretive to free cash flow. And essentially the EPS accretion is really dependant on how we choose to finance that last billion dollar component which we don’t have to make a decision on until about 30 days or so prior to closing. So, as we said before, we have the option of using equity or convert or cash or a combination of those. And so the specific EPS accretion will really be dependent upon what we finally decide for those things.

And then, a clarification on your comments that Micron will be above the CAGR that is provided for DRAM and bit growth in fiscal ‘17. I just want to clarify you would expect to be above the high end of your guidance for both the DRAM and NAND, so I mean I think it was over 30% in DRAM for fiscal ‘17 and over 40% in NAND for fiscal ‘17?

What we said, Mark, was that we’ve given these two-year CAGRs and we were slightly below for both NAND and DRAM in fiscal ‘16. So, it’s reasonable to think that would be slightly above the range that we provided in fiscal ‘17, so that we over the two-year CAGR fulfill the commitment that we made around those bit growth CAGRs. So yes that’s a reasonable thing to assume.

That’s very helpful. Thank you.

Thank you. Our next question comes from Timothy Arcuri with Cowen. Your line is open.

Thank you. I had two. Ernie, I am still little confused on the guidance. If I try to adjust for the changes in the reporting, it sounds like you are excluding about $150 million per quarter now. So on an apples to apples basis, the guidance is actually like $0.13 or something like that lower than what the headline number is or more like $0.04 at the midpoint; am I not thinking about that right?

Yes. That’s a reasonable way to think about it. Sure, Tim.

Okay. And then, I guess that leads me to the question about OpEx because it seems like some of that’s related to OpEx. So, I guess if you add the stock comp back and you assume that’s mostly OpEx, you are guiding to like 650 to 700, but it sounds like maybe 80 million of that is really one-time in nature; so, maybe the baseline is more like 600 million going into next quarter, is that -- sorry going into the fiscal Q2, is that the right run rate headed into the January quarter?

So, let me sort of try to frame that for you a little bit. We had a couple of things that are impacting this quarter’s OpEx. The first was that about 25 million, give or take, are pre-qual expenses that we’ve planned to spend in FQ4 are appearing in FQ1. And then secondly, we have -- we had even prior to this rollover, the highest point of pre-qual expenses for the entire fiscal year was actually occurring in Q1, plus we have the resumption of variable compensation given the new pricing environment. And so what we’ve tried to frame for folks is sequentially between the variable comp being reinstated, as well as the pre-qual expenses, that’s a sequential increase of about $80 million from FQ4 to FQ1, so that predominately explains the change to the OpEx. And then certainly we’re hopeful that the variable compensation piece continues, because we’re hopeful that the business environment is favorable. And we would expect pre-qual expenses to moderate such that we should be in that range of 600 plus or minus, so probably midpoint around 600, a few quarters, maybe couple of quarters higher, couple of quarters a little bit lower under the new framework for reporting OpEx. So, those numbers would still exclude stock-based comp, but you would expect to see a pretty meaningful moderation of OpEx as we go forward throughout the year as a result of this pre-qual situation being resolved from the rollover.

Our next question comes from Ian Ing with MKM Partners. Your line is open.

Yes, thank you. Congrats on the guidance. DRAM question here, you’ve got favorable contract pricing; you didn’t quite get the bit output acceleration in Q4. Just wanted to understand that a bit and are there any shortages in terms of meeting customer demand at this point?

I think part of the explanation of Q4 bit growth being slightly below, I think what we previously talked about was in fact that as we watched the market evolve we made some choices around where we’d like that output to be. And the result is that our Q1 -- fiscal Q1 bit output is now going to be higher than what we had expected it to be. So, obviously that did play into our thinking as we saw market evolve. Relative to specific shortages, I’m not aware of any. I don’t know Mark if you are. But certainly, all of our customers are being very attentive to making sure that their supply needs are taking care of. And we are seeing certain segments and certain customers where the possibility of allocation is present.

I don’t think we want to call this out today on this call, but there is certainly that risk.

Okay, thanks. And for my follow-up here, server as a percent of DRAM that is down sequentially and August is -- what’s happening here, perhaps enterprise is offsetting the cloud? And shouldn’t this be a pretty attractive market to serve? I think pricing and server typically tracks PCs with a little bit of lag.

Yes. I think it will reaccelerate relatively quickly. It has been a little bit more of a lumpy business here over the last year or so, but we like the trends there.

Thank you. Our next question comes from Rajvindra Gill with Needham & Company. Your line is open.

Yes. Thanks and congrats on good execution. Just a question on the inventory increase on a year-over-year basis relative to the revenue increase year-over-year based on your guidance. So, it seems like that the inventory is increasing at a faster clip than revenue. So, just wanted to get a sense in terms of any concerns you see there? So, is inventory outgrowing revenue on a year-over-year basis?

I think you have a couple of things. First, as we’ve talked about in the last quarter or so, as we’re ramping on the new technologies and predominantly because of these delayed mobile calls, we had some inventory built up that is now flowing through. However, we absolutely still continue to get more-and-more output from the factory and it’s going to be a couple of quarters before that inventory actually starts to decline, but we have a very clear view of an ability to decrease inventory meaningfully over the course of fiscal ‘17. I’d also remind you that we are also building supply chains for SSDs and MCPs which take longer in terms of their aggregate end-to-end cycle time than selling components. So, we have that effect as well. That will cap out here in the next quarter or so and then you’ll see normalization and a flow through of that as well. So, we’re not -- we’re cognizant of the inventory position, we’d like to think that we anticipated that and share that with you. And we’re equally confident that the inventory will decline nearly every quarter in fiscal ‘17.

Yes. To be honest, we feel pretty good about where we sit, given the trends in the marketplace as well. And we don’t feel quite as much urgency as we might under different market conditions.

And just last question from me on the cash flow, and obviously you can’t talk little bit about fiscal year ‘17. But if you kind of take your CapEx guidance and your depreciation add back, it does seem like it still will be kind of difficult to generate a meaningful free cash flow next year, assuming a fairly big ramp in gross margins and revenue, gross margins maybe to continue to increase above and beyond what you’re guiding for November. So, I just wanted to get a sense of how you’re looking at free cash flow. And Inotera agreement, how much free cash flow does that add to the Company, add that back on?

On average, the Inotera transaction is going to add somewhere between $300 million and $600 million of cash flow on a free cash flow basis, but I’d really trust that that’s an average amount which would be sort of the free cash flow net of the incremental depreciation -- I’m sorry, the incremental CapEx. But I think that the best way I could -- the guys are thinking there is that we’ve prepared pretty conservative plan and we have a free cash flow neutral position in the context of that conservative plan. Obviously Q4 is a little bit stronger; it’s fair to say that Q1 was a little bit stronger as we look at it now versus our conservative plan. And so, we would expect to be able to accumulate cash over the course of 2017 if in fact this current market environment continues. Anytime we use words like significant, it gets hard to determine what’s significant to you or significant to me, but trust me every dollar is significant to us. And so, we’re being very careful and monitoring that very carefully.

Thank you. Our next question comes from Harlan Sur with JP Morgan. Your line is open.

NAND cost per bit was actually up slightly in Q4. Was this mix related? Maybe you could just tell us what cost per bit declines were on a like for like basis? And should we anticipate an acceleration of cost per bit starting here in the first quarter especially I think as you mentioned you get 3D bit crossover over 2D? But it seems like to me as gross margins for NAND should be inflecting up here in Q1 but I’m trying to get a sense for that.

A lot of moving pieces, Harlan; I think just to get to the crux of your question, yes, you should anticipate cost per bit start coming down nicely with the 3D ramp as we start flushing those wafers through the backend and out in the marketplace. Costs are coming down nicely this quarter and we continue roadmap ahead for that as we move through the year.

Yes. And we’ve provided a cost per bit reduction forecast for the year to you. And I everything Mark said is absolutely true and counterbalancing that just a bit is this supply chain build up that I talked about because as MCPs and those SSDs have a higher COGS component, theoretically, they’re also generating more margin but if we’re only looking at the cost perspective, that will have a tendency to dampen the cost for bit reduction because the mix has changed.

Great, thanks for the insights there. And then, on the 1X DRAM ramp that you highlighted in your slides, when does that ramp actually commence; is it end of this year, beginning of next calendar year? And you’ve mentioned also that you expect significant output by middle of next year and maybe you can just be more specific on when you expect bit cost over 1X versus 20-nanometer. Thank you.

So, we have -- we actually have wafers in manufacturing fabs in both Hiroshima and Taiwan that are yielding. The output today is insignificant. We’ll continue to run engineering pilot line lots. And by the time we get to the middle of next year, you should see the significant terms of our overall output. I think probably got to wait another quarter before we give you little bit more precision on what the -- when the bit crossover might be.

Great, thank you.

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

Just two questions from me, just following on that prior question regarding NAND cost per bits and all the mix issues. Can you just take that one step forward and talk about maybe how the storage business unit starts to recover in terms of profitability; is there a path to turning profitable this fiscal year? And then secondly, can you just talk about seasonality this year maybe versus last year in terms of what kind of unit shipments you’re seeing versus content shipments from some of your core markets and whether it’s stronger or weaker than a year ago? Thanks.

So, let me address the question about the storage business. We are certainly hopeful that we can exit the year at least with a profitable note for the storage business. I think it’s dependent on a few things. Obviously we’re in the middle of a very significant portfolio sort of change out and we have to execute on those. So, far our record has been good with respect to the client and consumer segments, but we have some big product launches with respect to both data center and enterprise coming up here toward the end of this calendar year and in early calendar ‘17. And those would be really pivotal to us in completing that transition. I think that again track record has been great, the SSDs have been well-reviewed and we’re very optimistic that our subsequent product launches will be equally successful. I also think it’s important to keep in mind that we also are seeing the benefit of 3D in the memory -- I’m sorry, in the mobile business unit as well. We saw that -- we had the mobile percentage of our NAND business this quarter move from the low teen essentially to the high teens as a result of more and more calls [ph] there and the increasing importance of the MCP portfolio. So, we are seeing the benefit of that really across multiple business units.

Let me just add to that on the seasonality question. Q4 obviously typically is a very strong quarter for NAND. Hard to know for sure but our sense is that the demand picture probably has more staying power than just the typical seasonality that you see, given the strong growth in the end applications.

Thank you. Our next question comes from Kevin Cassidy with Stifel. Your line is open.

Your guidance for first quarter is significant increase over fourth quarter. Can you give us a ranking of the end markets or the end products that are driving the increase?

I’ll take that. So, obviously as has been the case, the sort of specialty markets, the embedded markets whether it’s automotive, industrial, medical probably top of the list within the computing and networking segment, graphics has been quite strong, server probably next and then followed up with mobile and client.

Okay. And are you seeing a significant increase with 3D NAND versus the planar NAND?

In terms of demand?

No, growth. I am sorry.

It’s a very quickly moving dynamic right now. So, yes, we’re seeing that happen sort of real time as we sell through the early production and move more and more of the products into the TLC format.

And certainly more of that opportunity is ahead of us than it is behind us given that we’re just going to hit bit crossover for 3D in this quarter. And as Mark said, we’re continuing to transition, which adds cost because it suppresses output. But with more significant bit growth ahead, we expect the cost reduction opportunity to accelerate a little bit here throughout fiscal 2017.

Okay, great. Congratulations on good results.

Thank you. Our next question comes from John Pitzer with Credit Suisse. Your line is open.

I guess I want to go back to the cost side of the equation, guys. I guess just given the change in the depreciation schedule for the November quarter, I would have thought your ability to get gross margin to kind of the high 20% range especially in this pricing environment would have been fairly easy. So, I’m just kind of curious as to what the offsets are in the November quarter. Is this all about the calls on mobile and if it is, what do you adjust to make that to be a hit to gross margins and when might that be done? And then, Ernie, I know you gave full year cost targets on a per bit basis, could you give the fiscal first quarter? If you did I missed them.

I think, John, actually it’s really a little bit more about the lag that you typically see as pricing starts to turnaround as some of these things. Yes, there is an inventory effect relative to some of these products that we’ve built up in inventory that we’re early product on either 20-nanometer or 3D NAND etcetera as we sell through that’s a little bit of a headwind until that’s flushed out. But it’s really more around as ASPs turn around, they will turn around instantaneously and they didn’t turn around until end of the quarter last quarter. So remember, we had a pretty significant feed relative to the midpoint of our guidance last quarter, which gives you an indication that pricing was starting to accelerate then. It takes little while to play out and to start to [ph] do as we see these things happen. So, the trends are all in place. I think we are heading where you think where we ought to be. But it can take a little while to play out through the financials.

Yes, and I didn’t provide a cost specific cost per bit target for FQ1.

Okay. And I guess as my follow-up, Mark, just on the Inotera acquisition, if you go back and look at what you guys have tried to accomplish over the last kind of seven, eight years, it’s actually been to try to mitigate the cyclicality of the industry within your financials by setting up some of these JVs. Buying Inotera kind of reverses that. I am kind of curious, is this purely a financial decision on your part? You mentioned in your prepared comments that this downturn troughed at much higher levels than past downturns. So, does that give you the confidence or is there actually a fundamental reason to own the asset; will this allow you to kind of ramp 1X, or 1Y or 1Z more quickly on Inotera? If you just give a justification that would be great.

Yes. There is a lot of components to it. I think you put your finger on a piece of it which is the volatility in this business I think we can conclude is going to be less going forward. This certainly seems to have been born out as we move through this last cycle. It doesn’t mean that cycles are gone, it just means that the volatility is less. So that is less of a defensive driver, so to speak. An important thing for us in this whole picture is operational flexibility and control. As we think about our ability to drive new technologies, either in the manufacturing or to transition technologies with the Inotera assets, how the ownership of that asset and the ability to mix and match different technologies as well as potentially more value add products and capture that value is significant. Notwithstanding the fact we have a good operating relationship with Inotera, we believe that there is more value we can bring as sole owners than as board members and operational partners. And at the end of the day, it gives us the ability as well to take cash flow that’s going to be generated with Inotera and to put it across our network where we find that most useful. So number of different factors play into it, all of which I think point in direction that we are moving.

Thank you. Our next question comes from Mehdi Hosseini with SIG. Your line is open.

Ernie, just going back to your CapEx for the next fiscal year, it sounds like you are incrementally more confident with the Inotera deal. And in that context, how should we think about the incremental CapEx requirement for Inotera? In the context of your comment that in Inotera would help with free cash accretion, I’m just trying to reconcile everything. And I have a follow-up.

So, when we provided that CapEx guidance of essentially midpoint of 5 billion net of partner contributions, we did say it included contemplated investments from Inotera. So, that’s an all-in number. And as you know, we don’t break out specific CapEx by fab. So, we don’t plan to start doing that now. But that 5 billion midpoint with a little range around that is inclusive of anticipated investments at Inotera.

Got it. Thanks for clarification. And then, with the 3D NAND and the crossover with 3D NAND, should I assume that all the 3D NAND capacity coming on line is TLC or is this clearly MLC left in the mix?

No. The first out 3D was MLC but we very quickly introduced TLC. And as we said in our prepared remarks, most of our client consumer SSDs are now out on TLC and we’re qualifying the TLC into some mobile applications as well and we will be majority TLC by the middle of fiscal ‘17 here as we go forward. So that’s part of the engine of the significant bit growth that we’ve contemplated and forecast.

Thank you. Our next question comes from Joe Moore from Morgan Stanley. Your line is open.

Great. Thank you. I wonder if you could talk about the decision to move to seven-year depreciable life on DRAM. I guess that strikes me as a long time. And I look at your CapEx ratio, PP&E sort of a third. What was the thinking my seven years and can you remind us the depreciation life of the NAND equipment?

So, the depreciation life of the NAND equipment is about five years. And really, we conducted quite an extensive study around how long the technology transitions existed or took in the DRAM world, how long they were contemplated to take on a going forward basis, we looked at the practices of others, both competitors and partners. But really the substance of the decision was related to what we anticipate and what we’ve been saying for some time is a sort of slowing of the technology transitions and therefore the longer useability of that equipment as the technology transition times change. And the reality is there is a range of answers you come up with when you do a study like that. And much like the midpoint of guided range in seven years felt to us like it was not overly aggressive, but not overly conservative either. So, you could assume that we could have gone a year or two on either side of that and we chose a position that we thought was reasonable given what we know and understand about the pace of technology transition.

Okay, great. Thank you for that. And then as a follow-up, have you given or can you give an absolute number for depreciation for either the quarter or for the fiscal year?

We expect depreciation to be somewhere in the range of $4 billion for the year give or take.

Great. Thank you very much.

And operator, I think we have time for one more question.

Thank you. Our last question comes from David Wong with Wells Fargo. Your line is open.

Thanks very much. Ernie, are you noting that you might be able to generate cash over the next few quarters? Can you give us any idea what you would do with any net cash generated; would you be able to pay down debt or other uses of cash?

Certainly, first thing we’re going to be focused on is actually accumulating the cash. So, that’s my first priority because I have to have some of it to be able to decide what to do with it. And certainly at that point in time, we would certainly look at the opportunities that were available to delver and we would look at those in the context of everything else that’s going around. And it’s really impossible to say specifically what we would decide to do in the context of a future that’s not yet here. But I can tell you that delevering is an important priority. And depending on how much cash we generate, we may choose to deploy a 100 million of that or so back into CapEx, if that’s the right decision, but delevering is certainly nearest and dearest to our hearts at the moment.

Thank you. This concludes today’s Micron Technology fourth quarter 2016 financial release conference call. You may now disconnect.