Micron Technology's (MU) Management Presents at Needham Growth Conference (Transcript)

| About: Micron Technology (MU)
This article is now exclusive for PRO subscribers.

Micron Technology, Inc. (NASDAQ:MU) 19th Annual Needham Growth Conference January 10, 2017 12:50 PM ET

Executives

Ernie Maddock - CFO

Analysts

Rajvindra Gill - Needham and Company

Rajvindra Gill

Everybody, thank you for attending Needham Growth Conference. My name is Rajiv Gill. I'm the Semiconductor and Memory Analyst here at Needham and Company. We're very pleased to have Micron presenting. With us today is Ernie Maddock who is the CFO, as well as Ivan Donaldson, Director of Investor Relations.

This is going to be a Q&A fire-side chat. We don't necessarily have an individual microphone tagged to them, so it's unfortunate, but I'll kick it off with maybe a question and then –we can go to the podium and take it from there and then we'll open up to the audience.

Question-and-Answer Session

Q - Rajvindra Gill

Again, Ernie, my first question is really around NAND supply, tightness -- there has been a lot of questions and comments about that NAND supply is pretty tight in the first half and there's a lot of capacity constraints. Wanted to get your thoughts on that.

Conversely, there's also a sentiment out there in the second half that more NAND supply will start coming online, particularly from the 3D NAND Flash vendors and that will increase the NAND supply in the second half?

So, I wanted to get your thoughts on both the supply in the first half and the second half and what the potential impact would be on pricing?

Ernie Maddock

Sure. So, the company's belief is that this year you're going to have roughly 40% to 45% bit growth in both supply and demand and we don't really spend a significant amount of time trying to figure out what quarter that's going to come to pass, because at the end of the day the reality is if we are balanced for the year, you're going to see the industry mature and project from a cost point of view in the way that it needs to.

So, I would absolutely agree that there's likely to be more supply in the second half of the year. If you look at, for example, Micron, we've given two year bit growth targets, last year which was our primary year of conversion, we under grew the market in order to hit our bit growth targets, we will outgrow the markets this year.

And likewise other suppliers to the space who are in the process of doing conversions, which really is nearly everyone will have a bit growth lower than expectations during the time they're doing that conversion and then when that capacity comes online, they will in fact likely outgrow for whatever small period of time that is over the course of the year.

So, the quarterly dynamics of who is bringing on supply, when, really is a function of so many things that again trying to stay focused on that is probably not the most significant thing to do relative to Micron as a manufacturer, we're very focused on how we think we're going to progress throughout the course of 2017.

Rajvindra Gill

Moving to questions in the audience. Maybe we can start and I'll continue.

Unidentified Analyst

You already referred to a transition in -- I guess it’s Flash that's underway.

Ernie Maddock

That's right.

Unidentified Analyst

And there's probably some transitions in DRAM that you can tell us about and just finally let me know where the 3D cross points that Intel has and I think you're involved with two. [Indiscernible]

Ernie Maddock

So, the question for those of you on the webcast is relates to the growth in NAND and what's going on there as well as DRAM as well as 3D cross points.

Our review of the DRAM business is that there will be somewhere between 15% and 20% bit supply from Micron and all the other participants in the industry. And then from a demand point of view, we think demand is going to be somewhere a little bit north of 20%, so somewhere between 20% and 25%. Micron will be going through its transition to the 1X node. We talked on our last call about having a meaningful number of bits out by the end of our fiscal year. We completed our 20 nanometer transition over the course of our fiscal 2016. So, we will participate meaningfully in the bit growth from an industry point of view that we're projecting this year.

And then relative to 3D cross point, 3D cross point we've said is a very de minimis amount of revenue in -- fiscal 2017. We will ship for revenue, but it's actually a fairly small amount and then we've set the expectation for somewhere around 5% of company revenues in 2018.

Unidentified Analyst

So then it would be a big rolling thing in the next five years, perhaps.

Ernie Maddock

Well, it's a very immature market right now, so I think that the appropriate approach is to take it with a measure of caution as the industry develops and we see what the market opportunities are for the first new memory technology in, gosh 20 years. No one can map that out particularly precisely, but we want to make sure that we maintain the flexibility so that we have the productive capacity available to us as we see the market develop.

We have a question in the back.

Unidentified Analyst

Can you comment about DRAM [Indiscernible] PCs and mobile and how do you see that unfolding maybe one to two quarters?

Ernie Maddock

Sure. So, the question is about the relative mix, and again, I don't really pay attention quarters. We can't run our business on a quarter-by-quarter-by quarter basis relative to how we allocate our production mix.

PCs, in general, are somewhere less than 20% of aggregate industry demand. Mobile is probably double that amount. So as you might expect we are very focused on the mobile market as a piece of the business that will likely outgrow PCs; we’re seeing some small unit growth decline in PCs, but you are seeing some content increase. So, in terms of opportunity, the mobile segment of the business is much more significant relative to the company's opportunity in a normal market, it tends to be something that fits very well into our desire to want to be differentiated as opposed to the PC space, which tends to be slightly more commodity driven. So, we continue to focus on that mobile space.

Unidentified Analyst

Can you talk about your competitor [Indiscernible] and how rational [indiscernible]?

Ernie Maddock

Well, I don't necessarily like to comment on other companies just as I don't like them commenting on Micron. I think that you can -- I think their comments need to stand on their own and their comment seems to suggest a rational approach to addressing the supply/demand constraints of the DRAM market. But beyond that, really these are questions better addressed for them.

Unidentified Analyst

But maybe compared to [Indiscernible]?

Ernie Maddock

Again, the world is very different today than it was a few years ago. So, trying to -- trying to pick two isolated years and compare and contrast, I don't necessarily believe that their behavior was irrational then, there was a shortage of bits being supplied into the market and one of the suppliers stepped in, but again, I'm not here to talk about what my competitors are doing, I'm here to talk about what Micron is doing.

Unidentified Analyst

Can you talk about the server [Indiscernible] trends have been and how has your share been in that market over the last few years [indiscernible]?

Ernie Maddock

So, Micron, in general, had less bit growth in the industry over the course of the last couple of years, predominately as a result of the acquisition of Elpida and the resulting consolidation that Micron did at 20 nanometer roadmap level. So, were we were later to transition to 20 nanometer, because we were consolidating a roadmap at Rexchip or roadmap at Elpida and a roadmap at Micron.

And given that share is a function of how fast you're growing bits versus how fast your competitors are growing bits, the reality is the company broadly participated less across all segments. With our transition to 20 in our subsequent transition to the 1X node we believe the opportunity is there for us to participate at levels similar to what we enjoyed prior to having that delay in the implementation and you know if you look at the segmentation of our DRAM business in particular, it tends to be nicely distributed over each of the end market.

So, relative to a couple years ago I think you're going to see us be at a higher representative bit share across any number of segments as a result of slightly outgrowing on a relative basis the industry vis-à-vis bits because of that 20 nanometer transition.

Rajvindra Gill

Just on the NAND side, you mentioned in the past that the transition from first generation 3D NAND to second-generation should reduce cost by about 30% and then double the bits per wafer. Could you give us a timeline on when you think this may occur in and what do you think some of the puts and takes will be for this transition?

Ernie Maddock

So, what we've said is that we will have meaningful output on that second-generation NAND which is 64-layer by the end of our fiscal year. So, it's ramping now and will continue to ramp over the course of our fiscal year and thus far at least from a yield performance and perspective, things are going pretty well, so we really aren’t experiencing some of the old challenges that are rumored to exist throughout the industry.

Rajvindra Gill

And on the DRAM side, the progress to 1X nanometer on DRAM side, that looks -- can you talk about where are you going to see future demands from some of these products coming from this particular transition will be for mobile, PCs, can you maybe define where do you think the demand will come from on the 1X nanometer transition on DRAM.

Ernie Maddock

Well, we think the demand is actually going to come from the broad market. So, I would expect that you will see PCs transition to 1X output DRAM. We think that there's opportunity in mobile, in the embedded segment and the server segment. So, really we wouldn’t foresee a bifurcation of the market relative to that DRAM content at the 1X node. Yes?

Unidentified Analyst

Just going forward you want to say a little bit about usage of cash and as [Indiscernible]?

Ernie Maddock

Yes, so what I've said is first I want to generate the cash. We were slightly free cash flow negative last quarter. We have the opportunity to generate significant cash over and above what we expected at the beginning of the year based on market conditions and the priority would be towards debt repayment. So, that's the immediate view of how we would use free cash flow in excess of our expectations.

Unidentified Analyst

Can you talk about your viewpoint on EUV introduction in DRAM manufacturing?

Ernie Maddock

So, currently we have no plans to introduce EUV in our DRAM roadmap. I know that that might be slightly different from what others have talked you about, but our roadmap right now does not contemplate the use of EUV.

Unidentified Analyst

Can you just elaborate why you don’t need it and others do?

Ernie Maddock

Well, again, I can't tell you about their process nodes, because I'm not a technologist and at the end of the day, how you move forward down the technology roadmap is really a function of where you are and what you think you'd like to achieve with the next several nodes.

So, I can speak to the fact that our technologists believe we can scale down to the onesie node without EUV. I can't explain why my competitor might choose to introduce EUV.

Rajvindra Gill

If we could look at -- switching gears to the specialty DRAM side of your business, there's been a significant amount of growth, primarily from networking, graphics, automotive, wondering if you could maybe narrow in on automotive segment and the breakdown that you're seeing from infotainment and ADAS applications, driving up memory content, DRAM content or NAND content really tied to ADAS proliferation in the automotive sector is that something you see as a big trend going forward?

Ernie Maddock

Absolutely, it's probably from a percentage point of view is one of the highest growth opportunities within DRAM. As we project forward and you think about realizing the vision of autonomous vehicles you are talking likely about four to five times increase in the current memory content on average per vehicle and then you start thinking about some of just the outright storage capabilities of having real-time maps and keeping them updated so that autonomous vehicle is equipped with the latest and greatest and you could be talking about up to a terabyte of storage in a car. So, it’s certainly over time represents one of the most significant growth opportunities from a percentage point of view and you obviously, there are fewer cars sold than there are phones or PC.

So, that high growth rate is applied over a bit of a smaller base, but is a really promising area and I think it plays very nicely into the company's strength because automotive customers are customers with whom you have to develop a much longer term relationship. They have to have assurance of supply; the entire supply chain of any company has to be fully qualified to automotive standards, which generally tend to be higher levels of scrutiny and care around the entirety of the supply chain.

So, having that ability, we currently enjoy double the share of our next largest competitor and with that learning and with those relationships comes a confidence that we will enjoy. Hopefully, our fair share -- or our greater than fair share of the growth that’s present in that business.

Rajvindra Gill

And is the memory content in automotive for ADAS primarily going to be NAND or DRAM or is there going to be more alternative memory technologies such as MRAM that have combined lower power, but high write-cycle speed, because the memory requirement is going to change based on ADAS applications?

Ernie Maddock

It potentially could, but one of the things that's different about the automotive design cycle is we are designing and [calling] [ph] today for 2018 and 2019 cars so by the time those other technologies become more prevalent we’re probably out from a model year perspective out into the mid-2020 sometime than I think that anybody who has the courage to predict exactly what's going to be happening a decade from now in the memory business is you know is a very bold person.

So, I would tell you that it’s more likely that certainly over a meaningful time horizon for the company with the technology that we know that it will be generally more familiar rather than less familiar technologies that will get us there.

Unidentified Analyst

When you get beyond the February quarter could you talk a little bit about the biggest [Indiscernible] biggest factors on your incremental operating margins?

Ernie Maddock

It would obviously be the pricing environment right. We've set forth for you what we do from a cost perspective and although we haven't been very specific again over the quarterly cadence of that, it's reasonable to think that we're going to be making some progress each and every quarter from a cost side, we've kind of given you the roadmap of our operating expenses and how you know how that's going to look over the balance of our physical 2017. So, really is a perspective about the pricing environments and that's an area where we have long ago you we do a business plan for sure but were not in the business of forecasting industry pricing.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

While the impact of those ramps is baked into the cost reduction right, so when we talk about being able to -- for example, generate a nominal 20% to 25% cost reduction in DRAM presumes a reasonable progress of the yield curve with that 1X node. If we do faster, we'll probably be at the higher end of the range if we encounter some difficulties it may cause us to be at the low end of the range, but there's nothing that we that we can anticipate it hasn't been baked into the perspective that we provided for sure. Yes.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

Well, I think the idea of a fully fungible manufacturing environment for technologies as different as 3D NAND and planer DRAM becomes a really interesting and frankly not terribly relevant discussion you don't move equipment back and forth with relative ease company is developing continues to develop with the acquisition of Inotera, sort of a DRAM center of manufacturing volume in Taiwan with NAND that's down in Singapore and I think it would take a very clear point of view on our point about the relative features of both technologies are either before you think about is changing that trajectory.

You build enough flexibility in your manufacturing infrastructure, if you wanted to add a small of wafers or those sorts of things, those are typically available within the geographies that are available to you on either technology whether it be NAND or DRAM, but in terms of this idea of -- today I'm going to decide to make DRAM in a month from now, I'm going to make NAND and then a month later, I might make DRAM. That's really -- in reality, it's just not the way things work. Yes.

Unidentified Analyst

Yes, so the biggest driver of bits demand growth in NAND is this belief that with the advent of 3D and TLC and QLC and ultimately 64 layer that you are going to hit price points for SSDs where the long-awaited full replacement of ACDs across a broad spectrum not 100% of the spectrum of storage will take place and you know you do have some growth coming in the mobile space, but the big headline story for the NAND space is that dramatic growth in SSD and then sort of the tip over, if you will, throughout the entirety of the client, data center and in some cases, hyper scale still environment.

So, I think the flexibility that we have is that -- what I want it's important to remember that's a very price elastic market. So, the only way you're going to realize that vision is if SSDs meet the price points that allow them to economically replace HDDs and should that not occur for whatever reason, whether its challenges around yield for the industry because I've just said that Micron isn’t experiencing those whether it's some sort of economic disturbance that creates less of an opportunity since we are getting and our competitors are typically getting a lot of that supply from both conversions and new capacity additions, you have the ability to meter those new capacity additions, again when you become convinced that that demand environment is going to be less robust that you think. So that's the way we would address those provocations [ph].

Unidentified Analyst

[Indiscernible]

Ernie Maddock

I think that's a broad general statement, but the nuances of the competitive landscape. So, if you say that maybe a true statement for the injury, but some suppliers don't have yield issues and therefore they are meeting their cost curves and as a result of meeting the cost curves, they are able to price accordingly.

If you are a supplier who is having yield issues, you have some difficult choices, right. You either don't anticipate or you participate painfully because you have to match that that pricing environments I think it's a true statement is as it applies to the broad industry, but I don't think it's reasonable to think the broad industry would encounter those challenges.

I think the real question would be does enough of the industry encounter those challenges you don't down the price curve and that appears to be less likely the case rather than more likely the case right now.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

Well, I mean we read the same industry publications. I mean we don't try to develop a significantly independent forecast of what that dynamic could look like. So, we read the same things you do. We talk to our customers and they're all very anxious to have the benefit of SSDs has provide if they can do it in a way that makes economic sense for them. So, 20 of those things were if you hit the price curve, you're likely going to see the tip over.

Rajvindra Gill

If we could just maybe touch on the cycle, kind of your general thoughts. So, if you look at the 2014 cycle which was a good point in the overall process, there was consolidation that was happening, there was more pricing discipline as result of that consolidation and demand was pretty good, but it didn't necessarily pan out into some structural change in the industry because a couple of years later, you have a drop off demand in PCs and that they created a similar situation where pricing fell and you entered into a period losses.

So, so it does seem as we going to be sure that were this year we're approaching maybe back a similar type of cycle we had in 2014, but perhaps a bit better because the demand for at least for DRAM is more diversified away from PCs and there's other drivers. So, I mentioned automotive, I mentioned network graphics et cetera.

So, I guess my question is what's your view of this cycle versus the 2014 cycle? And can you maybe talk about what happened in 2014 to 2016 and is there ways to prevent that or re duce it -- the risk?

Ernie Maddock

It's interesting because my definition of the cycle is actually not what happened in 2014 versus today, its everything that happened in the between, right, because the cycle is about starting at a point wherever you bottom out and then wherever you peak. And we still have no data that suggests we're at the peak, right, so I can only talk about what happened in the peak quarter, whenever that was in 2014 to today.

And I think there's a very strong case we make for the factored industry consolidation benefit. The industry gets 700 base points more profitability over the course of that period of time than they did the last time.

The semiconductor industry went through a "cycle". The cycle was shorter. It was less deep. Micron, although we lost money in a couple of quarters, we made money for the fiscal year so I think there's really significant data that suggests that industry consolidation has delivered benefits relative to historical performance.

I would be the last person to say this isn't a cyclical business because the one thing that suppliers will never be able to control his the demand environment and as you know we're in a very capital-intensive business.

So, once you have the capacity there almost no circumstances will you choose not to produce because you have a very, very high fixed cost structure and so the industry responded appropriately to the weakening PC demand environment and first nobody added capacity, and secondly, we all shifted our output to those more lucrative segments even mobile or automotive.

So, I am not of the school that tries to predict peak profitability to peak profitability because the end of the day I don't run my business for that we run our business for managing throughout the course of whatever the ups and downs of the business will be in delivering consistent profitability over the course of that. Yes.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

They are permutations of essentially specialty DRAM. They tend to be high performance DRAM high performance MCPs or multichip packages and again given the very long design cycles, so we're sitting here at the very early 2017 and we're starting to work on 2019 cars. The 2018 calls are well in the existing qualified on today's technology and in fact, if you go into the average automotive automobile that's produced in 2018, you're going to find older NAND and older DRAM because once they qualify those products, they don't like to change because changing costs a lot of money and it has to be qualified to automotive standards which are remarkably different than what you might expect for PC or even or even a mobile device.

So, I don't think at least the next two to three years and possibly a little bit longer that are going to look dramatically different other than permutations of the products that are available to the industry over the course of next couple years put together with controllers and you are you might have a high representation of high performance memory. But again that would look more like today's graphics products and not some dramatically new thing that doesn't exist.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

There's a little bit more on a relative basis, it's not so much and it's DRAM and high performance DRAM and NAND. So, I mean each one of those products has unique capabilities for the automotive industry in addition to just the general type of product that is more ubiquitous across the supply chain.

But I don't think you’re going to see -- you might see a higher representation of the specialty or high performance product, but again it's not going to look dramatically different from the kinds of high performance products that you're seeing today in the industry.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

I think as a general statement, no. But again if you're talking about Tesla, as opposed to a Prius, obviously, there's a difference. But they tend to be such a small percentage of the aggregate, kind of the aggregate automotive business that they tend to get lost in large numbers.

Rajvindra Gill

We have time for maybe couple more questions. Any questions in the audience, if not, I'll ask one question. Go ahead.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

Certainly, not over the next couple of years and you know less likely even over five or six years because I think if in a nice way referring to China and I think there's been a very clear statement on their part that their interest lies more with NAND than with DRAM.

So, I think if there's likelihood of anything which I would still assess as relatively small, it would be more along the lines of NAND and DRAM.

Unidentified Analyst

So, on NAND than what do you kind -- how do you view the [Indiscernible]?

Ernie Maddock

Well, in some measure what we think about is how much will supply expand whether it be China, whether it be, Samsung, whether it be us or our competitors. Clearly if there is a successful introduction of technology and I think that's a really important point because there is no track record of technology ownership which is very different than having capital to deploy.

An introduction of supply in excess of our currently modelled view would have a dampening effect on the pricing environment so it's no different than us or the industries that we have a lot of experience with like DRAM.

So, we don't pay as much attention in those out years as to whose supply it is. We pay attention to how much supply we think is going to come into the market and right now we believe that the market will support given from a demand perspective. There is going to need to be more supply, so whether it be China or someone else, pretty sure that's going to be supplied.

Unidentified Analyst

[Indiscernible]

Ernie Maddock

It's kind of hard to think about more consolidation on the DRAM side. You have three suppliers and -- I'm sorry.

Unidentified Analyst

More on the NAND side?

Ernie Maddock

On the NAND side, you have four holders of IP, right. So, you have Micron Intel partnership, there's Toshiba WD partnership, and then Samsung. So, I mean I think it's important to remember in DRAM we started out with 25 players in we've gotten down to three and were starting out essentially four holders of IP and so whether or not -- and by the way, DRAM is 30 odd years old and NAND technology is roughly half that age.

So, whether or not there needs to be more consolidation you can make an argument either case I think it's reasonable to think that any such consolidation is a way it's coming simply because the industry starting from a very different place than the DRAM industry started from.

Rajvindra Gill

All right. We'll have to wrap it up. But thank you so much. I appreciate it. Thanks. My pleasure.

Ernie Maddock

Thank you.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!