Micron Technology, Inc. (MU) Earnings Call Transcript & Summary

Good morning. I'd like to welcome everyone to this session with Micron Technology. It's my distinct pleasure to introduce Sanjay Mehrotra, the Chief Executive Officer of Micron; and Dave Zinsner, the Chief Financial Officer of Micron. We've got about 30 minutes in this forum to have a fireside chat. If you have any questions, feel free to e-mail me, and I will try to incorporate them into my prepared questions. But first, I want to thank both Sanjay and David for taking time out to support the conference. I wish we were in person and not virtual, but there's always next year.

So Sanjay, I usually open up the fireside with kind of an open-ended question to kind of give you the opportunity to help set the discussion. But you stole a little bit of my thunder this morning with your positive pre-announcement pre the market open. I'm wondering if you want to make some just comments around kind of the strength that you saw in the November quarter and the press release that you put out this morning.

Thank you, John, for having us here in this conference. Good to be back with you all. I want to start off by highlighting that during the course of the discussion, I may be making certain forward-looking statements, so please do refer to the risk factors in our SEC filings that Micron makes from time to time. And also some of the numbers that I may be referring to here would be non-GAAP-based numbers. So John, yes, you highlighted that we did pre-announced the quarter earlier this morning. We are very pleased with the execution that Micron has delivered during the quarter. We saw strength versus our prior expectations in the earnings call last time with respect to the guidance we had provided. We saw strength pretty much across the board. We saw strength in mobile, in auto, in industrial, in PC. As we had expected, cloud was healthy for us. Of course, enterprise continue to be weak, just like we had stated in our last earnings call that enterprise environment is overall weak. I'm pleased with how Micron executed with strength both coming from DRAM and NAND, strength coming in terms of volume in DRAM and NAND as well as with respect to pricing. So we saw strength in both volume as well as pricing versus our prior expectations. So we have revised the guidance for our F Q1 to revenue of $5.7 billion to $5.75 billion. Gross margins, we have increased it to 30%, plus/minus 50 basis points. And EPS, also, we have increased it accordingly, $0.69 to $0.73 in this revised range that we issued today. This sets us up well for F Q2 as well. Keep in mind, F Q2 does fall in our seasonally weak quarter. And when I look at DRAM beyond the seasonally weak calendar first quarter, I expect that strength will continue to build up in DRAM through the course of the year with all the industry supply considerations, but more importantly, all the demand drivers that continue to build up nicely for DRAM. Whether it is in the cloud or it is on the edge, DRAM will do well for us during the course of 2021. NAND side, while we saw the strength in volume and pricing better compared to our previous expectations of the guidance we had provided for the November quarter, NAND industry, as we have shared before, continues to be challenging. And of course, we have taken steps to manage our supply bit growth, as we have said, in calendar year '21 to be below the industry demand. I do believe that, overall, the NAND industry needs to continue to take actions to better manage the supply. Otherwise, in calendar year 2021, there may be some risk of supply exceeding demand for NAND. However, on the DRAM side, we continue to see strong fundamentals. And I'm very proud also of Micron's execution on the technology front. We announced during the quarter our 176-layer NAND technology. We leapfrog the competition with respect to this technology and as we transition from floating gate to replacement gate. And of course, Scott DeBoer, just yesterday in his technology webinar, which if you have not watched, I really, highly encourage you to watch that webinar, very informative in terms of how Micron is approaching its technology road map and execution, Scott discussed our 1-alpha node as well for DRAM. Both of these nodes position us very well for future cost reductions. They both will be -- our 176-layer NAND as well as our 1-alpha node will be very strong nodes with respect to cost reduction capability for Micron. And Scott provided a lot of details, I'm not going to repeat them here. Maybe later on in the discussion, we can go through that. But I would like to highlight that both of those nodes will be providing us -- starting to provide benefit in the second half of our fiscal year because the 1-alpha node will begin production sometime in calendar first half for DRAM. And the 176-layer node, which has begun production, will gradually transition over to our breadth of portfolio over the course of the year. So these will provide us ability for bit growth as well as for cost reductions in the second half of our fiscal year. And of course, mantra at Micron is to manage our supply growth carefully, keeping our supply growth CAGR to be in line with demand growth CAGR. So we are very pleased with our strong technology execution with these leadership nodes of 1-alpha DRAM and 176-layer RG NAND. But we'll, of course, be ramping them gradually over the course of time, continuing to focus on supply bit growth, carefully managing it and making sure that the supply bit growth CAGR is in line with the demand CAGR that we estimate as well. So overall, I'm very pleased with Micron's execution. I'm pleased with how the market opportunities continue to shape up. And I think during calendar year '21, as the global economies get past the pandemic and as global economies grow, the need for more memory and storage will continue to broaden as well. And I believe that even enterprise, which today is weak, once we get post pandemic, I believe that during calendar year '21, enterprise sector as well will start improving. So I think long-term trends continue to be strong for memory and storage. Micron is extremely well positioned, and we are focused on executing our strategy.

I've got one quick follow-up from the press release this morning. And I know that typically, you like to save the detail until you actually report earnings, which I believe is going to be January 7. But when you originally guided the November quarter, you guided gross margins below, where I think most investors had thought -- and one of the headwinds that you had in the quarter was your ramp of new graphics, DRAM and the yields there. Did that get better throughout the quarter? And how do we think about the trajectory of kind of that graphic DRAM yield curve as we go from November into February?

So you're right to point out that we provide more details, and we will be providing more details on our business results in our January -- early January earnings call. Overall, as we had noted last time that one of the factors is that some of the new products, such as, yes, that graphics, GDDR6X that we are very proud. It's a differentiated product offering that we have worked with NVIDIA over the course of multiple years to introduce that. Absolutely leading edge, fastest DRAM product in the market. It's very common, John, that when you ramp up these new products -- and not just GDDR6X, we have begun shipments of HBM2E, the high bandwidth DRAM as well and LPDDR5 DRAMs as well. As you ramp up these new products, it's very common that there is a ramp-up of yield that occurs. And in the early phases, the cost is not most effective, and that was all baked into our guidance that we had provided at the time. Overall, I would say that we continue to execute well to our plans with respect to our new product ramps that are occurring. And our guidance, of course, had baked in our assumptions at the time with respect to overall industry pricing assumptions. And keep in mind that our 1Z-nanometer ramp is pretty much behind us. We executed it successfully. So that ramp has reached 1Z nanometer, and DRAM has reached as maximum stage. Now in terms of the supply bit growth, it will be fairly flattish for us until we bring the 1-alpha node as well, which will then provide cost benefits and supply benefits in the second half of the fiscal year as estimated. So those were the factors that were baked into the guidance at that time. But as you can see, we have produced meaningfully stronger results, again, driven by volume as well as pricing and some of the mix factors and Micron's continued execution on the cost front as well.

That's helpful, Sanjay. I want to get into Scott's sort of technology workshop that he hosted yesterday. But before I do that, I'm going to ask you just kind of a bigger picture question. I mean there are things that you can control and there are things that you can't. But what I thought was kind of interesting as I was preparing for this year's conference, as I started running some numbers -- and what I realized is that Apple for the first time ever topped the $2 trillion market valuation this year. And don't get me wrong, Apple is a great company. But what I find amazing is when I aggregate the entire global semiconductor ecosystem, so EDA, equipment, fabless, foundry, IDM, memory, the combined market cap of that interior ecosystem is only $2 trillion. And it strikes me that it's pretty easy to argue that semis have been more consequential to society than Apple, and yet the market doesn't seem to be valuing them any differently. I'd be curious from your perspective how do you think about value capture as the CEO of Micron? What's the value proposition that you're trying to put forth to the investment community?

So I think as you have seen that over the course of last 3.5 years or so, we have focused on technology execution and product strength, and you have seen the examples that I just discussed with respect to our 1-alpha DRAM node as well as 176-layer NAND. We are now technology leaders in terms of having the technology capability, and we focused on that technology acceleration in the past going forward. Now that we have caught up on the technology front, the cadence of new technology nodes that we'll be introducing will be fairly competitive, will be in line with the industry's advancement of the new technology nodes. But this positions us well to build value for our customers as well as to build value for our investors. Aside from strong technology positioning, both in DRAM and NAND, what is important for us is to continue to strengthen our product positioning. So for example, in NAND, our focus has been high value solutions, and we actually achieved that earlier than our original targets. We have 80% of our NAND now in high value solutions. We are proud of this achievement. We have done very well in mobile with discrete as well as multichip packages with NAND. We have done well on the SSD front, on the consumer side and continuing to expand our portfolio for client and data center applications as well. With respect to product strength, also focused on -- in DRAM. We just discussed bringing leadership products such as GDDR6, lowest power; our 1-alpha node that actually gives us a 40% bits per wafer advantage versus our 1Z node at mature yields in 176-layer NAND; our die is 30% smaller than the leading die by some of the competitors. So we have positioned well with respect to technology and product, and this is how we build value for customers as well as for our investors. And what I would like to highlight here is that when you look at Micron through the cycle and how Micron has delivered profitability even through the cycle. This is a very different Micron from Micron of the past. I mean we have stayed profitable through the troughs and producing strong results, not only through the cycle, but in the face of once in 100-year pandemic as well. And this is how we are focused on building strength, building value for our customers and for our investors. And when you look at this in the backdrop of how memory and storage is actually becoming increasingly more valuable in the ecosystem today, then I think -- I do believe that we will continue to close the gap in terms of our valuation with respect to some of the other players, particularly in the semiconductor ecosystem. Just want to point out, John, that memory and storage is growing faster than the semiconductor industry average. In 2000, memory and storage used to represent about 10% of the total semiconductor industry. Now it is nearly 30% of the total semiconductor industry. And with all the applications of AI with 5G, autonomous, IoT, from consumer to business, from data center to edge, you are seeing more and more applications where more data is being created, more data is being used, more intelligence is being derived from data, more processing is being done across all verticals, across all industries, which just points to more need for data solutions, which is memory and storage, which is where Micron is a pioneer. So when you look at the backdrop of this, secular trend of growth in memory and storage and how Micron has transformed itself with respect to technology and product leader and applying the discipline of growing our bits in line with the market demand, we believe we are well poised for continuing to drive financial strength through the cycles of our industry. And that's where I feel very excited about the opportunities to build greater value for all stakeholders for Micron.

Sanjay, I've argued in the past that the core philosophical change that you brought to Micron was this idea that it's hard to succeed being a fast follower on an exponential curve. Now what I mean by that is the Micron of old was always good at optimizing for lowest entry cost but not necessarily lowest cost of ownership, and I think one of the key things that you've really been focused on is trying to close the cost gap that Micron's historically had on both the DRAM and the NAND front vis-à-vis peers. Curious if you could give us something -- your grade on how you've done. But more importantly, relating back to Scott's webinar yesterday, what's the future look like? It feels to me like the 1-alpha node in DRAM and the second generation of replacement gates are not only better-than-normal cost downs for shrinks, but they're also the beginning of a sustainable path of kind of having good cost downs relative to peers going forward.

So in terms of Micron's achievements over the course of last few years, as I highlighted earlier, I'm very proud of the team at Micron and its execution. The team really has a tremendous set of capabilities, and that shows in our ability to now introduce industry-leading 1-alpha node in DRAM and 176-layer node in NAND. We are, of course, also focused on building a portfolio of high value solutions. Over the course of last few years, we have delivered in DRAM, leadership with LPDRAM, lowest power DRAM solutions. And of course, in NAND, we have delivered leadership with QLC Solutions, which is an exciting, growing opportunity for us in the future as well. And aside from technology, we have strengthened our product portfolio as well with strong position in mobile now, not only just with LPDRAM, but with solutions that combine LPDRAM and NAND as well. So I feel very good about Micron's ability to continue to execute well on our technology road map and the product road map and overall continue to execute to our strategy overall. Going forward, we have an exciting road map for technology and products as I mentioned earlier and as Scott discussed in his webinar as well, that we have the road map laid out through 1-alpha node, 1-beta, 1-gamma. And all of these nodes in DRAM will continue to use multi-patterning. And we'll, of course, introduce EUV when we see that it is suitable for us in terms of delivering the cost objectives. But I want to highlight that our road map with 1-alpha, 1-beta and 1-gamma node will be well positioned with respect to cost, but also with respect to performance of our products. And certainly, quality is a very important factor. And Micron has tremendous experience over the years with multi-patterning, with lithography, with process technologies, with materials to really deliver the kind of quality and performance and cost blend that we achieve with our road map. So I feel very confident with our team's capabilities to deliver there. And then NAND, we are well positioned. I mean if you look at Micron from day 1 of our 3D NAND, we have at CMOS under the Array, right? We are now in the fourth generation of dual stacking. We have been leaders in that. Others are now starting to do that as well with respect to 176-layer NAND. So I feel very good and particularly good about how Micron successfully transition. And I know a lot of people were concerned about floating gate to replacement gate transition, but Micron executed very well. I think all this speaks to the capabilities here. We will continue to execute well. I feel good about our technology and product road map. And with respect to products in NAND, as I said, we have made a lot of progress, but more progress to be had with respect to over the next 12 to 18 months, continue to strengthen our client SSD portfolio, continue to bring out our internally-developed controllers and fully leverage the vertical integration capabilities, and then next will come more data center SSD solutions beyond. They're strengthening our position in client SSDs as well. So we have a strategy of continuing to stay at par with the industry with respect to technology nodes, continue to manage our bit growth prudently and then manage our CapEx investments carefully and continue to strengthen our product portfolio and leverage our position with our customers with product leadership, as we have demonstrated with several announcements over the course of last few months.

Now Scott had some impressive details, especially on the second-generation replacement gate versus floating gate, which I thought was fairly impressive. Sanjay, you mentioned earlier in this conversation but also on the last conference call, which was a little bit of a change, your view that the NAND industry still needs to consolidate further. And I'm wondering if you can elaborate on that. I mean we've seen sort of the Intel-Hynix announcement. Is that enough? How do we -- how do you think about consolidation? And Micron has never been shy about being consolidated. Or do you see yourself driving that process? Or do you think others in the industry need to?

So you're right, John. I have always said that the NAND industry can benefit from consolidation, and the Intel and Hynix combination that you mentioned that the deal that got announced is one step toward that consolidation. With respect to Micron itself, obviously, we don't speculate on any consolidation or any M&A matters. But what I'll tell you is that we are focused on executing to our strategy in NAND. With respect to -- as we just have discussed in detail, with respect to our technology and product positioning and with respect to our position with our customers having both DRAM and NAND, I believe that Micron is well positioned. Of course, NAND, as an industry, needs to work toward higher levels of profitability. But Micron is absolutely focused on continuing to deliver its strategy with respect to NAND execution, and that's what we remain focused on.

Unfortunately, over the last couple of years, semiconductors have become synonymous with China-U.S. geopolitical tensions. I know that this is somewhat in the realm of speculation, but I'd be curious to get your thoughts on that dynamic because there's clearly some pluses and some minuses. The minus is being your ability to ship to customers like Huawei. The plus is being that when you think about new entrants in the memory space, they're mostly domestic Chinese. How do you think about that sort of push/pull? And what ultimately you think the end game here?

So with respect to the customer ecosystem in China, Micron is very well engaged. In fact, I would like to point out that in the last earnings call, we had talked about inability to shift to Huawei due to commerce department's ruling beyond September 14. And we had said that we would be mitigating the effect of Huawei by the end of the fiscal second quarter for us. And I'm very pleased that, overall, our team, in terms of our applications engineering team, our sales team that works with the customers, our operations team, our quality teams did a great job in terms of our ability to shift some of our production from Huawei to other mobile customers. And this is where our diversified engagement in the mobile ecosystem, including a diversified engagement with respect to our customers in China, really played to our strength. So we had said that we will have the effect of Huawei mitigated by the end of fiscal second quarter, so we are on a trajectory that is actually better than that, and that certainly helped us with respect to our mobile strength in F Q1 update that we just provided today. But as I noted, we saw broad-based strength except for the enterprise space with respect to our execution and business in F Q1. So with respect to China, we continue to stay engaged certainly with customer ecosystem, and that is important for us. With respect to competitors in China that I've talked about and that have been working on both on the NAND and the DRAM side, we, of course, remain extremely focused and paranoid about all competition all the time. We do not underestimate any competition. Micron actually over its 40-plus-year history has really been in a neighborhood with very many competitors, but you can see that we have had the tenacity to execute well with respect to our technology and product objectives. So I would like to point out with respect to those competitors in China is that we will continue to monitor it. We don't underestimate it. We do always account for a certain amount of bit growth coming from those suppliers. However, it seems to keep getting pushed out on a year-over-year basis. And just remember that in technology and product, it's just not enough to demonstrate a working die. It really requires products. It really requires deep understanding of technology, reliability, customer applications, how to enhance quality, how to ramp up production, how to deliver it in cost. There is a lot of work there, a lot of heavy lifting that needs to be done. And of course, we, in the meantime, are not sitting idle leader. We stay totally focused on deepening our customer engagements across the globe and continuing to move fast with our technology and product road map. So that's what -- as I keep repeating, that's what we stay focus on as we look ahead at addressing the opportunities globally, including the market opportunities in China.

Sanjay, you gave us a lot of great commentary about some of the end demand dynamics going on in your business. I'm wondering if we could dive into that a little bit more. One question on DRAM, one question on NAND. On the DRAM side, clearly, the incremental story has been the emergence of data center as a driver of bit growth. And when you go back a decade ago and the server market was probably 10% to 15% of DRAM demand. Today, it's 30%, and it's not hard to imagine the trajectory of kind of 50 or greater over time. I'm kind of curious, though, as memory becomes a bigger percent of the bill of material inside of the data center, these are really smart customers that are paid to try to get every inch of ROI out of their infrastructure, are you worried that there are things that they can do to optimize memory that actually changes the growth trajectory of DRAM with inside the data center?

Certainly, data center customers do look always for software optimization opportunities in terms of how they use the memory. That's nothing new. I mean that is being done over the course of the years, yet the need for memory and storage is increasing fast, as you noted. I mean all the trends of AI, certainly, in terms of data center, is driving greater need for memory and storage. Look at new processor architectures that are being announced, they have more channels. They have more ability to attach to higher memory. They have higher performance capabilities and requirements. All of that is definitely driving increased need for memory and storage. So we think that this trend of more memory and more storage in servers and in data center applications, particularly in the cloud applications, is secular. It's here to stay. And of course, software optimizations will be made from time to time, yet there is -- we have barely seen the tip of the iceberg when it comes to AI. And with respect to DRAM, I would just like to point out, John, that it's not only about data center. Look at mobile, 5G, I mean next year, the calendar year 2021, about 0.5 billion smartphones are expected to be sold that should have 5G. 5G drives more DRAM and more NAND content as well. So our end markets are well diversified, and we see longer-term growth trends in all of our end markets.

So Sanjay, maybe you can comment because I think one of the things that's unique about Micron is the fact that you not only provide NAND and DRAM, but you also have efforts in XPoint. Where does XPoint fit into all of this?

You're right. I mean we are the only company in the world that have NAND, DRAM and 3D XPoint. We are -- we introduced the world's fastest SSD, 2x faster than the closest competitor with 3D XPoint technology. So again, goes to show Micron's focus on product leadership and building stronger product portfolio. 3D XPoint, in terms of ecosystem, in terms of architectural changes and software stack changes that are required to advance this opportunity, will take a while. We, of course, in the meantime, continue to develop product both memory semantic as well as storage semantic with 3D XPoint. Over the course of next few years, we definitely expect 3D XPoint to become an increasing opportunity for us, and that will further play out to our strength to be able to offer to our customers the full portfolio of NAND, DRAM and 3D XPoint and combination of these technologies in product solutions and ultimately let the customer pick in terms of their application, what delivers the greatest value. So I'm excited about the opportunities with 3D XPoint. But as we have always said before, these new transformative technologies do take multiple years before they become a bigger portion of the commercialization opportunity, partially given that DRAM and NAND are large market opportunities today.

Perfect. With that, Sanjay, we've come to the end of this session, but I really wanted to thank you for spending time with us this morning. And I want to pass along my wishes that you, your immediate family and the broader Micron family continue to stay safe and healthy in what's been a very challenging 2020.

Thank you, John, and that's what I would say to all our listeners here as well that please continue to be safe, be well, and thank you for your support.

Thank you.