Micron Technology's Management Presents at Raymond James 35th Annual Institutional Investors Conference (Transcript)

Mar. 3.14 | About: Micron Technology (MU)

Micron Technology, Inc (NASDAQ:MU)

Raymond James 35th Annual Institutional Investors Conference Call

March 3, 2014 10:25 AM ET

Executives

Kipp Bedard - VP, IR

Analysts

Hans Mosesmann - Raymond James

Hans Mosesmann - Raymond James

Okay, we’ll go ahead and get started. Good morning everybody. I am Hans. I cover Semi for Ray-Jay, and Kip Bedard, and most of you know who Kip is over at Micron. He is the VP of Investor Relations, with employee number?

Kip Bedard

790.

Hans Mosesmann - Raymond James

790. Okay.

Kip Bedard

We’re over 30,000 now.

Hans Mosesmann - Raymond James

Okay well, early days. So Kip, thank you for coming. And before we start, we just published our new semiconductor reference guide, there in the back. This is great if you have young kids. But anyway it is good a cheat sheet if you like to have that. So Kip let’s go ahead and get started. So you just had a Analyst Day in Scottsdale, couple of weeks back, and a lot of talk about return on investment, a lot of debt reduction, and a lot of things that if you go back 10-15 years ago, what have been kind of surreal to kind of even see what?

Kip Bedard

All new words for us to use, yes.

Hans Mosesmann - Raymond James

That’s right. So can you give us an update on just the market as you see it in 2014 just at a high level and then we can open up for questions.

Kip Bedard

Sure thanks, as I always for inviting us down my wife and I always enjoy the warm weather coming out of Idaho. So I appreciate a weekend here. But quick update on supply and demand dynamics, starting with DRAM, we think again another pretty light year in terms of supply, well actually be fewer wafers produced in the DRAM sector this year than last year by probably 3% or 4%. Of course we do spend CapEx to lower the line, geometries that we use, we do generate additional bits, all that tied together probably means we will increase supply this year somewhere in the low 20s, in terms of DRAM.

If you go to the different demand dynamics, still led this year by I think the same themes, that we saw last year which is anything, Internet and mobile in particular. I just came back from Barcelona which you’re probably familiar is kind of the largest mobile conference in the world today. And I think there was some disappointment that the higher end phones didn’t increase content as much although I will suggest you there were at least three handsets that went from 2 gigabytes to 3 gigabytes of content.

But what is critical for the DRAM space on the demand side is that this mid to lower end continues to move up. And in that regard we saw a tremendous increase in content. So what all that means when you wrap it around is mobile just like last year will lead the highest bit growth demand this year as well which will be in that 60% to 70% year-over-year consumption number. Mobile today consumes about 35% of all the DRAM bits, so it’s a pretty good start to the demand story.

Behind that, because of all these mobile devices and new devices that are getting connected to the Internet and they are driving more and more information, they’re creating data packets, so they’re sending it up to the cloud and back down. That’s driving the second highest driver on demand, is the server area. That’s growing at about 40ish% per year. And that’s about 20% of the market. So you got over 50% of the market just in those two segments that are growing well above the supply possibilities.

Then you go into networking. And the recent discussion around the build out of LTE in China is a huge demand driver and we’re feeling that real time. Networking represents about 10% to 12% of the consumption, and it’s going to grow at over 30% to 35% this year in terms of bit demand. So now you’re up to about 65% to 70% of the market that is growing well above our ability to supply it.

The one area that still feels just mediocre at best is PCs, they will probably be flat at best may be a slight growth in terms of units, and maybe the content is up 5% or so. So that will be one area where probably underperforms in terms of demand dynamics. Similar to last year and the year before, but still the rest of the market looks pretty darn healthy to us. So you got supply coming in that low 20s to mid-20s, you’ve got a demand situation that’s looking 30%, probably maybe a little bit higher than that this year. So demand looks pretty good to us in DRAM, the demand supply relationship.

In NAND, I think it’s going to be a little bit more choppy, there are some very nice markets that are very elastic with pricing Client SSDs would be one of them. Last year we probably had an attach rate, meaning that of the notebooks shipped in the PC space, probably only 10%, roughly chose to insert an SSD instead of a typical hard disk drive. This year today we are already running about 15% to 25% attach rate, so kind of a nice increase there. But it does require some ASP encouragement for consumers to move there. A little bit more volatile, I think in the dynamics around NAND, but good long-term growth rate.

Overall supply probably increases 35% to 40% this year, and I think demand is going to be just in that 40% to 45% range probably encouraged little bit by ASP in some of these segments. Again biggest drivers are probably SSDs in general, both client and enterprise. It’s on its early hub, it’s early ramp in terms of penetration in the enterprise segments, again driven by this need to have data supplied in the cloud and readily available to us on handsets and mobile devices, and then us pushing it back up to cloud to store it and keep it long-term. That’s driven by more DRAM as a cache and then right behind that is a first storage level of NAND and then your big hard drive kind of behind that, so, again supply demand dynamics look pretty healthy, but I don’t think it’ll be quite as smooth as in the DRAM space because we do have to add capacity in NAND to meet the 40%, which means we’re going to be adding wafers this year. We probably won’t do it perfectly as an industry but the mid to long-term I still think looks pretty healthy.

Hans Mosesmann - Raymond James

Okay. And this thing on NAND, what are pricing trends here entering the spring or as we exit the winter, what does it look like?

Kip Bedard

Yes, pretty similar to what we guided on the conference call, we’re expecting kind of mid to high-teens in terms of pricing reductions. I think we’re pretty much seeing that today in the markets. Lot of it -- we’re probably a little bit heavier on this price decrease because we do have a product coming out of our new Singapore transition that doesn’t, isn’t qualified up into strong OEM enterprise environment, so that’s a little more of our product is floating around the spot market today. But all in all it looks like the market’s pretty much like we expected.

Hans Mosesmann - Raymond James

Okay, so any questions from the audience? Okay, so let's move onto 3D NAND. It was a topic with SanDisk earlier. They presented -- they won't have a 3D NAND product or their ramp won't be until really 2016 or maybe September in 2015. What’s Micron's position there and how is that coming along from the planar to 3D transition?

Kip Bedard

Sure, we’re probably going to be in the market with samples a little bit sooner we’ve talked about over the next few months getting 3D sampled into various clients and -- but I think we’re similar to SanDisk and Toshiba’s thinking and even Hynix’s thinking that the real ramp is probably latter part of 2015. If you think about NAND in general, it just has a much longer qualification period and certainly a new technology like 3D, we as an industry have to develop this infrastructure of controller technology and firmware and then those solutions have to be qualified by customers, so it’s quite a lengthy process to get that rolling, and even then on top of that when you layer the initial 3D applications, or the initial 3D products out there, won’t necessarily be more cost efficient relative to the planar technology that we’re deploying today. So we have got some economic reasons that probably it’s going to be later rather than sooner when 3D really starts to take a decent share in the market.

Hans Mosesmann - Raymond James

In terms of 3D in your architecture, I guess how is it different from say Samsung or Hynix's or Toshiba or SanDisk's if you know?

Kip Bedard

Yes, it’s a good question, not one that our R&D guys will let me share with you yet, other than to say everybody’s 3D architecture is different. And so we will have different strengths and different weaknesses and we’ll have different price points where cost points, where it starts to compare favorably to our existing NAND planar technology. So those are a few things that will be different but literally all the cell architectures are different, different approaches from each one of the non-volatile suppliers.

Hans Mosesmann - Raymond James

And how many layers do you guys -- have you disclosed? I think Samsung is at 22, 24?

Kip Bedard

Samsung is 24 we have not disclosed and I don’t believe the other competitors have disclosed yet either.

Hans Mosesmann - Raymond James

But it's the industry's view or the consensus view that you need to be around 30, 35, or something like that.

Kip Bedard

To make it cost competitive, yes.

Hans Mosesmann - Raymond James

Okay. Shifting gears back to DRAM. And I'll ask the pricing question there or perhaps I should ask it. Are contract negotiations -- do they continue to extend in terms of timeframe, not two weeks but now every month or two months or…?

Kip Bedard

Haven’t seen a lot of change in the length so far, PCs again if you’re not that familiar we renegotiated with our tier 1 OEMs about every month, typically servers are about twice that length of time, mobile can be anywhere from a quarter to three quarters out in terms of pricing negotiations and/or trends that you’re looking at. So haven’t seen a lot of change in those negotiated contract lengths, I think there’s always an interesting dynamic when pricing is stronger customers certainly want to lengthen the negotiations and of course when their prices are falling they want them pretty short, and the industry approach really and Micron’s as well has been over the years we’ve just established, however often you want to renegotiate it’s going to be the same no matter where we are in the cycle. So if you want to be 30 days, great, we’ll negotiate with you in 30 days. If you want to be 90 days, great, we’ll do that. But we’re not going to change back and forth -- so not a lot of change.

Hans Mosesmann - Raymond James

Investors are worried about Hynix's Wuxi fab that had the fire and is coming back online here in the next several months. Any update there that you can see in terms of market implications?

Kip Bedard

Yes, I think we’re in the midst of seeing that product float around the spot, so I would suggest to your clients here to expect some volatility in the spot market here for the next 30 to 60 days primarily around PC and mobile which was the -- our largest percent of output that they had, product coming out of that fab, so that’ll have, those two markets will have the most bit growth coming in here over the next 30 to 60 days. I think what we’re seeing, and back to my original comments about where the demand pockets are I think PCs are again just kind of floating, they’re not getting any worse but there’s not a big growth rate there, so I wouldn’t be surprised if we give a little bit of pricing back there, on the mobile side however tremendous growth. In fact we’re seeing clients expedite requests for additional product real time, so I have a better feel for that but I think we’re going to run through mobile without much impact even as Wuxi comes back on, so again just to kind of repeat, maybe a little bit of volatility in pricing around PC, certainly in the spot market, certainly in the spot, maybe we give a little back. Maybe we don’t on the contract side, we will see. At mobile feels like there’s an up demand, to probably get through it.

Hans Mosesmann - Raymond James

Okay and even with the Wuxi coming on-board you still think that there’s no change in the overall year bits supply growth rates.

Kip Bedard

In fact that’s all factored into that load of mid-20 kind of supply numbers, those guys are coming back on.

Hans Mosesmann - Raymond James

Any questions from the investors -- okay here we go. Once you ask and I will repeat it.

Question-and-Answer Session

Unidentified Analyst

[Indiscernible]

Hans Mosesmann - Raymond James

Memory equipment orders are being pushed out -- increased.

Kip Bedard

Increased, okay yea a couple of thoughts around that; number one, Samsung and Micron are in the very beginnings of a 3D sample ramp, and that equipment for the same relative wafer output is 1.5 to 2 times more expensive for us. So you are going to see just to try and get to 5,000 wafers of capacity a month. It’s a pretty big capital spend in 3D. Also for Micron specifically we’re spending quite a bit of money today in transitioning Elpida and Rexchip, our new acquisition along with Micron facilities into the next generations of DRAM. So we ourselves have picked up CapEx. For example last year we spent about 1.3 billion-1.4 billion. This year we have targeted about 2.6 to 3.2; it doesn’t add any wafers.

And then my third comment would be I think all of us as DRAM suppliers are on record that every additional shrink even within its category, whether it is NAND going from 20 to 19 to 16, or whether it is DRAM 30 to 25, to 20; it’s getting more capital intensive, and we are losing square footage. So we have to spend about 25% to 50% more to keep the same wafer output with each shrink because of the DRAM in particular because the process steps increased so much so to get the same wafer output you have to spend more money.

I think that’s what is confusing around some of the capital equipment guys talking about increased orders and me sitting here telling you that we have no DRAM wafers coming on, in fact we will have fewer in the marketplace this year. NAND we’re adding wafer capacity. We as a fab, of course are moving DRAM to NAND and then we have got a little bit of growth out of Samsung on 3D and we will see a little bit out of Hynix this year as well. So yes we’re adding some wafers in NAND which is I think all capsulated in the capital equipment spending and then these other factors are increased. It shows up for them as increased orders. But for us it’s not capacity increases.

Hans Mosesmann - Raymond James

More questions? Ted?

Unidentified Analyst

[Indiscernible]

Hans Mosesmann - Raymond James

So the question is, the complexity of controllers for 3D NAND versus planar?

Kip Bedard

I don’t know how to give you a comparative number. All I can tell you is that controlling 3D NAND is exponentially more difficult and so between the hardware reconfiguration of controllers and in particular the firmware impact, it’s pretty significant. I don’t have a number for you yet because we just quite frankly haven’t seen enough applications and build outs of those yet for me to have a good solid number for you, but absolutely no, it’s going up.

Hans Mosesmann - Raymond James

And just as a follow-up. Is it possible that 3D NAND is a dead-end; I mean that just won’t work?

Kip Bedard

I think for certain applications I think we are over that risk. It may prove to be more difficult to make an effective economical product, so it might take longer for us to penetrate. But there are some distinct advantages of 3D NAND. It has much better re-write ability, which just means we can write to it much more than we ever have before, which has a lot of implications. One is, stability for the person using it or the enterprise using it. But also we don’t have to over provision the NAND near as much. And today we get the rewrites that they want because we, instead of 128 gig it’s what we’re really giving you 120 with the other 8 as extra that you never can use, but we use it if the bit goes bad. So you can start to do some things like that. So 3D has some advantages there. But the early material in 3D will have some disadvantages. It will have worse retention, which means we have to rewrite the data every so often or else we will lose it. And so there are trade-offs even within 3D that it’s probably not the perfect solution for everything in the future. But I think there are applications where it will be quite good. Enterprise is a great example. They want that reliability. They want that additional speed to write to it and read from it, and those are things that are enhanced with 3D technology.

Hans Mosesmann - Raymond James

I don’t -- go ahead.

Unidentified Analyst

[Indiscernible]

Hans Mosesmann - Raymond James

Yes so the question is debt reduction, and where do you see in 2 to 3 years, your capital structure?

Kip Bedard

We want the debt to cap ratio to be in that 20% to 25% range, when we’re all done with this. Yes in terms of order, kind of capital allocation, we want to continue to be focused on reducing the dilutive effect to the converts that we have out there. Yes, there are selling at a pretty high premium and we need to get those out before we start looking more seriously at some stock buybacks, because it’s pretty expensive for us to keep pushing the stock price up. Secondly we want to get into this 20% to 25% debt to cap. And then thirdly I think we would start looking at return on cash, if the market really is, and if we were in a new world order here and things continue to look pretty good then that will be kind of the way we get there.

Unidentified Analyst

[Indiscernible]

Hans Mosesmann - Raymond James

Yes the question was, is that the goal to get rid of all of them?

Kip Bedard

Probably not all of them, no, but I think we will probably issue more high yield debt, and will certainly continue in the capital equipment markets, the capital equipment leasing markets, pretty heavily. But now I think some of those converts, I think they’re pretty good long-term debt rates for us and so but we do want to take out these ones that are so heavily in the money, because they’re very diluted very expensive for us to go get them as the stock price continues to walk up.

Unidentified Analyst

[Indiscernible]

Kip Bedard

We will use a combination, the question is would we use cash for it, yes, I think we will continue to be like this. Last call where we do, we’ll probably do some debt financing, but we start using of higher and higher percentage of cash.

Hans Mosesmann - Raymond James

So when are you going to provide guidance, earnings call?

Kip Bedard

I keep telling it’s only all the lazy people who want us to do that. It’s a big advantage for people who want to do to work to get the guidance we have and make a model of it. I understand, it’s a very complex model and every quarter we do try and get better at how we provide data, and if you have suggestions, believe me we are very open to it. There’s going to be a point in time Hans to your point, where the guidance is going to change. Yes I think we feel, again if the market continues to develop like we hope and think it’s going to, we will feel better about putting out longer term kind of revenue and gross margin type guidance. But keep in mind we just got profitable for the first time last May, so we’re less than a year into this new world, a bit of ahead of it so yes.

Hans Mosesmann - Raymond James

Any other questions in the audience, Ted?

Unidentified Analyst

[Indiscernible]

Hans Mosesmann - Raymond James

Yes the question was we announced the $250 million prepay from a major OEM last quarter and is that having any impact on anybody else?

Kip Bedard

I think that today I would characterize the market as yes. We could do more of that if we saw a benefit to it. That the appetite is there on the customer side, in particular those who are in a faster growth arena and worried about their ability to get bits and so yes, there are discussions around that. If we saw benefit to it then we’d probably execute a few more.

Hans Mosesmann - Raymond James

So what’s after 3D NAND, and where are you guys comparatively, whatever the answer maybe to that?

Kip Bedard

Sure, there is four to five technologies around memory that have potential to be the follow-on to 3D NAND. I think our opinion is whether it is MRAM or whatever happens to be, but I think our opinion is probably none of them today look like they are a major market share component but they look like they have attributes that are in slices of a pie. They can be probably the next technology. So nothing really is a follow-on to 3D NAND that I think can be a major market share play, but there are these technologies that I think will fit pretty nicely and just still a certain specific applications.

Hans Mosesmann - Raymond James

Anybody? So there’s a lot of discussion, I was at an event called the Open Compute Summit, a few weeks back in Silicon Valley, and it’s one of these Open Compute Project efforts that Facebook started a few years back, and big push to have everything opened in the next-generation Hyper-scale datacenters and to the degree that you’re participating in that, what are those dynamics due to memory players and specifically Micron if anything or as such?

Kip Bedard

Yes, we’re actually pretty excited about it because it allows individualized solutions. So if you look at a Facebook, as an example or an Amazon or a Microsoft or a Google, or YouTube; they have significantly different ways that if there was an open architecture that they could take advantage of processing power and memory and they would do it differently. So it allows us to go in and provide different solution options for each one of those buckets as an example. So we’re pretty excited about it. We’re working with all of those types of customers, and I think it’s going to be kind of interesting in terms of how much memory you are going to see consumed there.

And then secondly I’d put a little plug for our Automata folks. As we continue to create these massive amounts of unstructured data and we’re trying to shift them all the time, it’s a huge opportunity for the Automata type top parallel processing architecture that we’re promoting there. I look at, if you think about all those companies we just talked about, their ways to increase revenue and profitability, is to know what types of ads to push to you. So they are going to continue on a real time basis understanding how they can process these tons of additional bits of information that are coming real time and real time feed that as one example into your advertising that gets fed back to you. So that there’s a massive change here going on, an opportunity for processing power, processing solutions all tied with memory and the way that we go and look at this data and figure out how to best utilize it, both for advertising and also for content for you.

Hans Mosesmann - Raymond James

We tend not to talk too much about NOR flash like we do obviously NAND and DRAM. What's the update there in embedded markets and strategically how does that position you as a broader-based memory provider to the world?

Kipp Bedard

Sure, it’s actually a product that we -- it’s actually pretty strategic for us, it’s not all that big in terms of percent of revenues by itself but what it does is in these embedded applications it drags with us, with that product increased margin and content for DRAM and NAND. And so it’s a tremendous way to get into markets that are very sticky, very solid gross margins, very little overhead for us to be in that business and it’s a huge cash generator for us. You just don’t need much CapEx in that space, and so it uses trailing edge equipment so it allows us to depreciate the equipment over a longer period of time, generates a lot of cash flow and it fits so nicely with other DRAM and other types of memories we can put around that enter these embedded systems that it’s absolutely worth being in it, you don’t want to necessarily be in the mobile part of that, because that’s shrinking pretty quickly with handset use. But these embedded applications like automotive, medical and industrial actually have a nice 8% to 10% growth rate per year and again very stable profitability in it.

Hans Mosesmann - Raymond James

I think embedded markets tie into this whole relatively new theme of Internet of Things, the IOT and big opportunity, kind of tough to play from an investment perspective. What's the Micron angle there, if any, and how do you play?

Kipp Bedard

I think NOR is part of our angle but it’s also having the right subsystem put together in MCP form or embedded form that has a combination of DRAM and NAND, so it’s more volatile and non-volatile type memory, so I think we feel pretty good that we’ve got the gamut covered with all three technologies and I think within those we can find the right solution for kind of where, where all we want to play, so I think we feel pretty good about where we are.

Hans Mosesmann - Raymond James

Okay, any questions out there? We have time for maybe one more or two.

Kipp Bedard

Can always make my answer really long.

Hans Mosesmann - Raymond James

So SanDisk was here as I said earlier, and they have made some acquisitions that is going to beef up maybe on the enterprise side of things and some other areas. What acquisitions can we expect, if any, from Micron? And what do you need or is it all organic going forward?

Kipp Bedard

Yes I probably won’t be really direct with the answer in terms of what we’re going to grow organically and what we’re going to inorganically grow but at our Analyst Day we mentioned we’re probably going to do a combination of both. We’ve got holes in our NAND solutions offering that we want to go fill. Like we don’t even feel good about how we’ve executed in NAND really to-date, we look at -- we compare ourselves and benchmark ourselves against the best and you just named one of them. We have this delta in gross margins that inherently just shouldn’t be there. And it’s our goal is to close up that margin, some of it is just simply blocking and tackling and some of it is adding the right technologies and adding the right solutions and getting them to the client at the right time and not being late to qualify. So we have got a lot of opportunity there, I kid with Mark and Mark and say I’m glad everything’s not working perfect today because everyone would ask me what you’re going to do for me next quarter so I’m okay with that and obvious way that we can still get better and return more than to you as a shareholder, so, I’m okay with that so far.

Hans Mosesmann - Raymond James

Okay, great. Last chance. Okay. Kipp, thank you very much.

Kipp Bedard

Yes thanks Hans I appreciate it.

Hans Mosesmann - Raymond James

We’ll go to the Cordoba XYZ room downstairs for the breakout. Thanks.

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!