|Seagate CEO Luczo On Drives, Zettabytes, Flash And His Tattoo |
Right at the end of March, I sat down in the cafeteria at Seagate headquarters in Cupertino, California with CEO Steve Luczo to talk about the state of the disk drive industry. There was a lot to discuss: the aftermath of last year’s highly disruptive floods in Thailand, the consolidation of the industry to just a few players, the stock market’s consistent disinterest in disk-drive company stocks, the opportunity for Seagate in the cloud, the role Flash memory will play in data storage going forward and the fact that he has a tattoo on his ring finger and a red string around his wrist.
Luczo has become the drive industry’s most outspoken leader. He makes no bones about his disdain for the investment community’s lack of respect for disk-drive industry shares; he thinks anyone worrying about the threat from flash storage is deluded; and he sees vast opportunity ahead and annual consumption soars into the zettabytes.
Here’s an edited version of the interview. It’s long, but stick with it; good stuff lies within:
Q: Steve, you guys have had quite an eventful year. How’s the recovery from the floods in Thailand going?
A: Well, it’s still happening. There are still a lot of issues, and the industry’s not being rebuilt the way it was. People just don’t understand that when you have that kind of structural damage to a supply chain it doesn’t get rebuilt the way it was. It will be rebuilt for the future, not for the past. And a lot of capital has to flow back in to rebuild all the assets that were destroyed. There’s a long lead time on a lot of that equipment. So it is still recovering. It is getting better. Certain market segments this quarter felt like they were coming into balance. Others are still massively short and will be through the summer. Overall, the recovery is still pegging pretty close to what we said initially which is kind of remarkable, [with industry-wide production of] 120 million units for the December quarter, 140 million for March, 160 million for June, 180 million for September. It seems like we are going to be within spitting distance of all of those.
Q: Pre-flood, the market run rate was what?
A: 180 million units. The TAM [total addressable market] is 170-180 million, something like that. 180 is where we were pre-flood, and our estimates then were 180 for December, 180 for March, 180 for June and probably 200 for September. I still think that is probably the right unconstrained TAM. Whether or not September goes to 200 or we get there in December is kind of a function of when does stuff take off for Windows 8, but it is close enough. The most important thing is that the obsession about units is ridiculous.
Q: Why do you say that?
A: Because it isn’t about units, it is about petabytes. For us, however big the shortfall is in units, the shortfall in petabytes is bigger.
Q: Why is that true?
A: Because people are mixing down in order to ship units.
Q: So some of the units have lower capacity.
A: That’s mostly happening on PC drives. If manufacturers are limited on motors, or heads, or disk, then you try to get as many units as you can so customers can sell computers, because you need a disk drive to sell a computer. So if that means I am building a 320 GB drive versus a 500 GB drive, or a 750 GB drive, that is what I’m going to do.
Q: Ah, right.
A: For the March quarter, most of the industry analysts have said they think it is going to be about a 140 million units shipped. But there’s been a big draw-down in inventories again, though. Inventories were drawn really heavily in the December quarter in order to get to 120 million, and they were drawn down again. Inventories are going to be at rock-bottom levels, and at some point those have to be replenished. We can’t run the business on the level of inventory that we have, and WIP [work in progress] is drained.
Q: What’s happening with the drive supply chain?
A: A lot of the critical suppliers are still recovering. Seagate has done a lot of work with our suppliers in terms of assisting them, whether in terms of capital, or clean room access, engineering talent, getting them machines that they need, all sorts of things. When all is said and done, there is probably going to be a reinvestment of at least a couple of billion dollars back into the industry just to get to pre-flood levels.
Q: Will it all be reinvested back into Thailand?
A: It kind of depends. The big U.S. multinationals seem to be reinvesting almost dollar for dollar in Thailand. Some of the big Japanese suppliers are into China or the Philippines; overall, I’d say there is a bit of a re-balancing away from Thailand, and in Thailand the industry will bee less heavily concentrated in the Bangkok area and the flood plain.
Q: How goes the integration of your acquisition of Samsung’s drive business?
A: Really good. We were done late in December. We actually shipped 700,000 units or something like that in the last week or so of Samsung product. We keep the operations separate from a sales and pricing perspective.
Q: Is that for regulatory reasons?
A: Yes, what MofCom (China’s Ministry of Communications) asked us to do was to keep pricing and sales separate, but other than that we are able to integrate. The design center is up and running, they are doing great, I was just over there, they’re really enthusiastic, we’ve had really successful hiring over there.
Q: How much does it increase your capacity?
A: They were doing about 12-15 million drives a quarter. When we get fully ramped, they will be at that level or higher, depending on market needs.
Q: That’s on top of your production of around 50 million – another 30% or so. The other big deal, meanwhile, is that your chief rival Western Digital bought Hitachi GST. How would you contrast the two deals?
A: Hitachi was a separate company, not an integrated company inside the parent like Samsung. So it was already kind of stand-alone, and much bigger – a.t least twice as big. They also play in different segments. Hitachi’s stronger in enterprise and notebook; Samsung is really notebook, and channel, so like desktop drives. Samsung’s notebook business was pretty concentrated in terms of customer base, whereas Hitachi’s was broader. But Hitachi was also vertically integrated, with their own head business, and Samsung basically sourced all their technology from TDK and external media guys. So the deals were different on a few levels.
Q: Had you looked at the possibility of buying Hitachi?
A: Not at all. WD needed to do the deal because they needed to get the enterprise business, and they weren’t able to do that on their own, despite trying for 10 years. They are hard products to build, it takes a lot of R&D, the customer set is different, servicing the customer is different, it is really hard. And I’d say the two biggest strengths of Hitachi are their enterprise business and their core technology. They are really good in heads, which is also probably a relative weakness at WD. For us, we have a leading head company and we have the leading enterprise company, so the value for us was different than the value for WD.
Q: Steve, this is an industry that has been consolidating for 30 years. A whole bunch of once-public companies that gone, Maxtor, Micropolis, Miniscribe. We’re now down to three, and one of those, Toshiba, is relatively minor. Is this process now done? And if so, do we now see an industry where pricing is more stable? Are the boom-and-bust days over for the drive business?
A: At the drive level you’re done. And you are probably done for the most part throughout the supply chain. As much consolidation that has occurred in drives, upstream it has occurred too. There used to like 9 head companies, and now there’s just one, TDK. In media there were 25 companies, and now there’s 3. Or 2. There has been a lot of consolidation, and the companies that had the strongest technology or the lowest costs have tended to survive, that’s how it is supposed to work.
Q: So will the economics of the business improve?
A: If you look back 5-6 years, there have not actually been a lot of boom and bust cycles other than the macro economy. Supply and demand has been getting more and more in balance, especially as the supply chain has consolidated. You just don’t have as much potential for excess capital as it consolidates. People are more cautious about over-investing. Also, the technology has gotten so much harder -unless you’re vertically integrated, it is really difficult to make next generation products.
Q: And what about pricing?
A: You have a situation where supply and demand probably more often than not is more closely aligned and therefore you don’t have massive under-supplies or massive over-supplies. In the last few years, you’ve either had over-supply because there’s been a shock to the economic system, or under-supply because there’s a shock to the economic system. We had the recession in 2008, where all of a sudden, boom, no one bought drives for four months, so there was this over-supply. But then all of a sudden everybody realized, oh my God, they still needs drives, and everyone had cut back their capital, so there was this huge rebound, and the industry couldn’t respond, so margins went from 15 points to 32 points. And you had the whole European slowdown, which caught the industry and everyone off guard.
Q: But the point is, the drive industry was not the victim of its own bad decision making.
A: Right. It was not so about bad allocation of capital, which is what it was before. The industry has done a pretty good job for the last 4-5 years, and will continue to. But you do have to remember that it is fundamental to our business that you have to keep lowering the price of technology. You have to go from 1.7 billion people using the Internet to 2.5 billion. To get there, chip costs have to come down, drive costs have to come down, bandwidth has to come down. Otherwise you can’t address the developing markets. There’s no doubt that even with three players, it is going to be an aggressive business – but I think some of the volatility gets reduced.
Q: So, Steve, will that give the Street more respect for what you do? Drive companies historically trade at some of the lowest P/Es in the entire stock market.
A: Investors don’t fundamentally understand what we do. They tend to think, if things are bad, that’s how they’re going to stay, and if things are good, they’re going to get bad. That’s really been the philosophy. The glass is either half-empty, or its empty. Do I think that’s going to change? I don’t know. There are other big successful technology companies where you scratch your head and say why are they trading at an 8 P/E. Why is Microsoft trading at an 8 P/E? Why is Intel trading at an 8 P/E? I think a lot of it is a different discussion entirely, which is do our capital markets fairly value companies anymore fundamentally, and I’ll tell you my answer is no.
Q: Why not?
A: Because they’re not being price on fundamentals, they’re being priced by the large investment banks for volume. And volume requires volatility. So that’s why all the big firms are in a different camp than the smaller research firms on their perspective of the drive industry. Is it because these are the smart guys, and those aren’t? That doesn’t make any sense. It is because the big banks are motivated by volatility and the boutique firms aren’t. And therefore the research reflects that. So you can create an environment that always creates doubt, with billion dollar market cap swings in a week. It’s insane. Why does that work? Because you have traders who love to make that work. So do I think that changes? I don’t know that I see that changing for our equity markets in general, which is a terrible thing for the efficient allocation of capital. Does it change at the margin for the drive industry, if there is less volatility? Sure, I think it does. It takes time though. It’s going to take time.
Q: Let’s talk about the impact of flash memory on the drive industry.
A: Any analyst who is worrying about that fundamentally doesn’t understand the industry. Flash is a complimentary technology, it’s not a competitive technology.
Q: Intel is out pushing ultrabooks with all their might. Some have drives, and some of which don’t.
A: If you want to store anything on it, it will have a drive, or you are going to pay a lot of money for flash. I mean, yeah, if a bunch of rich people in Atherton want to buy a PC for $1,000 that has 128 gigs, then, sure. I just don’t think that is much of the world, and oh by the way, if they have that machine, they have some other machine somewhere that’s got 5 terabytes on it, because I can’t do much with 128 gigs. So those products that are built for speed and mobility, those still need support of mass storage somewhere. And oh by the way, most ultrabooks, or thin and light, which is a different term – ultrabook is an Intel marketing term, so hopefully you’ve received your check this week for using that term – those have certain requirements on performance, and certain budgets on price. And the right answer for that is probably going to be a hybrid drive, because that is the only thing that can get you the performance they’re asking for at the budget they’re asking for if you want any amount of storage capacity.
Q: What about the idea that tablets have begun to cannibalize laptop sales?
A: Maybe they have. But then they’ve driven storage sales somewhere else. If it’s an Internet access device, which is what I think it is, then people are using it to watch YouTube and share videos. So where is all that stuff?
Q: Sitting someplace on a drive. So, demand for drive capacity will continue to grow.
A: Our industry shipped 100 exabytes of data five years ago, 400 exabytes in 2011, and we’ll probably ship a zettabyte sometime between 2015 and 2016. A zettabyte is equal to all the data that’s been digitized from 1957 through 2010. Everything, however you want to think of it, cards, tapes, PCs, mainframes, client/server, minicomputers – one zettabyte. And we’re going to ship that in one year. So whatever the architecture is, pads, phones, notebooks, ultrabooks, real notebooks, PCs, servers, clouds, one year, a zettabyte – that’s all going to be on rotating mass storage.
Q: And demand will keep ratcheting up from there.
A: By 2020, that number is somewhere between 7 and 35 zettabytes, depending on who you’re talking to – Seagate, which says 7, or EMC, which says 35. There is no amount of flash that can even address one tenth of one percent of that. People get locked in to this view at a device level. Yes, you could have some number of units that are serviced by flash. Let’s hope so. In fact, my bigger concern is that the flash guys can’t figure out how to keep delivering the performance and costs that they’ve been able to as they get to sub-21 nanometers, than it is that somehow they’re going to replace HDDs. Not without literally $500 billion of investment in fabs they’re not. And even then they’d only be scraping the surface.
Q: Someone said to me if you tried to replace all world’s drives with flash, today – your 400 exabytes – that there’s not even enough flash capacity in the world…
A: It would be like 25% of that. And that would mean not building any phones or iPads or anything else. And what’s the revenue opportunity for that? 25% of the drive market is $10 billion. So you’re going to turn over every fab in the world for $10 billion, even though they’re generating probably $150 billion doing the other stuff? Think of it the other way though – if I am trying to address 400 exabytes of data, and ok we admit the whole world’s not going to go that way – like we admit its only 20% of the notebook market – so the notebook market is 125 exabytes, and 20% of that is 25 exabytes; that means I need four of the most recent Samsung Fab 16s – the world’s largest memory fab – four. That’s $50 billion. And by the way – it has to be up and running today.
Q: What’s the total global capacity of magnetic media now?
A: Again, 400 exabytes, which is why units don’t matter. You can increase units, but they you’re taking a big exabyte hit. Here’s the real question – if you’re trying to grow exabytes, what’s the capital investment to grow exabytes. For flash, it is $10 billion at a chunk, which is what it takes to build a new chip fab. For drives, it’s a fraction of that. If the industry had to grow from 400 to 500 exabytes, we’d do that on $1 billion of capital.
Q: To get to a zettabyte, how much additional capital is that?
A: About $1 billion a year for us, and then about $1 billion a year for WD. Versus $50 billion. Here’s the more interesting question. I don’t think we can get to 10-20 zettabytes. I think around 2016, we start running into a sustained storage on meeting exabyte demand unless somehow people start getting aggressive on capital. And the only way people are going to get aggressive on capital is if there is some stability on margins. People look at drive capacity, but our industry isn’t about drive capacity. That just means final assembly and test. There’s also media, heads, chips, a lot of other stuff that has to happen before it all gets stuck into a disk drive. I
Q: That could be a factor in the ongoing recovery from the floods in Thailand.
A: That’s true. Someone might say, theoretically the drive guys at today’s mix could do 180 million units by September. I’m not sure all the sub-suppliers line up to that number. You may not have something, a motor base plate, a head, something else. There are 200 parts per drive. That’s another thing people have lost focus on. When we’re doing 50 million drives a quarter, we’re taking in 10 billion parts a quarter. And we’re configuring that not into 50 million units, but thousands of configurations and products. The supply chain management of what we do is huge. It’s this head, matched with this media, which is matched with this lube, which is matched with this motor. These are really integrated machines.
Q: How many SKUs?
A: Thousands. We have one customer that we sell 500 SKUs. I wish it was more of a commodity, because my manufacturing would be a lot easier. It would be way better for us. The problem we fight is, you start to building a bunch of drives and the customer says, oh, wait a second, I didn’t really mean I wanted 1,000 oranges, I wanted 1,000 lemons. But we didn’t build lemons, we built oranges. And then you have to figure out how to re-jigger everything.
Q: As demand grows into the zettabytes, what happens technologically? Are there ways to see leaps in performance or capacity?
A: That’s the heart of the issue. Storage demand growth right now is over 50% in the cloud, in other businesses its 25%. Overall call it 40%. Areal density growth is right now under 25%. So you’re going to say, well, wait a second. If this is growing at 40, and this is growing at 25, how are you going to fill that gap. The only way we can do it is more heads and disks. Or more units. Which means more capital. And not just more capital from the drive guy that’s doing final assembly and test, that’s actually relatively cheap capital. It’s more capital upstream in heads and disks which is relatively expensive capital, and it is long lead time capital.
Q: That sounds like trouble.
A: One way the world has changed since 2008 is that people aren’t going to make capital decisions leaning into the wind. They’re going to wait, they’re going to say, I’m not making that commitment until I know that when that factory opens, it’s full. When money was easier, and everything was growing, and you felt like you weren’t taking any big macro risk, you’d say, ok, I’m going to plan in advance. If I’m here today, and here’s where I think the market is going, and I know right here I’m going to have to have a new factory, and the lead time on that factory is 12 months, then I’m going to start investing then. No way are people doing that now.
Q: Too much risk.
A: People are waiting until they are way within lead time to make sure they’re really down that curve, because they don’t want to have all of sudden a slowdown, and then think oh my God, what am I going to do with all this excess capacity. That’s the change that people don’t get. Now, I think as investors actually see that playing out, then they get more confidence and maybe P/E’s change, right. So where the problem lies for the drive industry, is that with areal density growth rates for the next couple of years being sub-25%, data growth rates over 40%, then what’s going happen is it is going to be more heads and disks. And how does that play out?
A: It probably plays out that on the client you start by going to more a single disk solution. Right now we’re doing a single disk 500 GB drive. Do people need 750s and one terrabytes, yes, but 500 is a lot, and then it go to 640, and then 750. So you probably start seeing a reduction in heads and disks for clients, but then you start seeing an increase for cloud. But overall somehow you’ve got to solve this problem. When you look out at 2016-2017, then you have a next generation technology called HAMR – heat-assisted magnetic recording – which accelerates the areal density curve, and you can start to catch up a little bit then.
Q: How much with HAMR help?
A: It probably gets you back to areal density growth rates back in the 40%-50% range. And there are some technologies in between that are media-related, or software-related that maybe boosts it a little bit. But my point is, I think you are in a perpetual shortage. So the only way you can solve the whole exabyte storage is with more heads and disks, which means you’ve got to put more capital to work, and no one is doing that. And they are not going to do it until they’re sure. The next shortage will be in media. Sometime in the next two years, there’s going to be a media shortage, because there’s just not enough capital being put to work.
Q: How many players are there in magnetic media?
A: There’s still a few. Fuji, Showa, Seagate, WD; Hoya is big in substrates, they make them and sell them to other guys who make media, like us, or Showa or Fuji. But it’s consolidated, that’s the point.
Q: Could we get new players?
A: The technology is too hard, on all of it. No.
Q: What does that mean for investors in drive stocks?
A: If you were a really big, smart technology company that saw all the trends in cloud and mobility; all that says is storage, storage, storage – you’d probably see a reason why owning these critical technologies is important. But you’d run into so much flack from someone saying, oh, you should do software and services only.
Q: So are you saying something could try to buy you or WD?
A: I think if someone were thinking about it from a pure technology perspective, they’d be all over it.
Q: But who would it be?
A: Any big technology company, you can fill in the blank. IBM, EMC, HP. Anyone that’s big. But they won’t, because that’s not what the market wants them to do. The market does not want them to be in the hardware business. Although Todd Bradley [who runs the PC business at HP] just wrote a great article on why hardware is important. And the end of the day, all this stuff has to be delivered on hardware. If I’m in the business of selling in the cloud, and all the cloud is is a disk drive with some bandwidth, I need access to drive technology. At the end of the day, all I’m doing is taking disk-drives and integrating them. It depends how big is my cloud business going to be.
Q: I’m curious when you think about all of these companies that want you to store your stuff in the cloud, DropBox, or Carbonite, there’s a bunch of them. Is there an opportunity for you to play in that market?
A: We have a service provider business. It’s small, $100 million. For small- and medium businesses. You’re on the right point. One question is what happens to the market (DAS is an acronym for direct attached storage – basically local storage rather networked storage.) Let’s just take the Bay Area market. It goes to NAS (network attached storage, which allow you to access information over the network.) At the end of the day, why would you have a DAS device if you could have a NAS device in your home that you could access with your phone or your notebook, that you could access when you’re traveling. It makes no sense to have multiple DAS devices.
Q: But that’s like a half step.
A: That’s a half step. Then the question is do I also want stuff in the cloud or not. And it depends how much data I have. And then there’s a security issue. Do I want my kids pictures in the cloud? Personally, I don’t. [Steve has two small kids and a third on the way.] I don’t care what anyone tells me about security, I’m not going to wake up one day and find out , oh , sorry, our thing got broken into, like what’s happened to really amazing companies like Sony, and find my home movies on the Internet.
Q: So, are you going to get into the cloud storage market, Steve?
A: Would we offer something like that? Maybe, it depends, if we can do it cheaply – and we have some ideas on why we can do it cheaply – but right now what we do is sell to all those other guys who do that. And that’s fine, too. If they’re successful, then they’re going to be need more disk drives. The investment community is obsessed on software and services, but the barriers to entry on software and services are actually a lot lower than then are on the hardware side.
Q: But what if when you went and bought a back-up drive, you turned it on, and there was an offer to give you a bunch of cloud storage for $10 a month, or something like that. Seem like a no-brainer.
A: It is a no-brainer, as long as you can make it easy. Do we think there’s a business like that for the consumer? In the developed markets, I think there is. And then the question is can we built that architecture cheaply, because at the end of the day’s it’s not like we’re going to rent Amazon’s infrastructure. And we think the answer to that is yeah.
Q: But not yet.
Q: Another business opportunity is data recovery for crashed drives.
A: Seagate Recovery Services is a really good business we have that we haven’t reall probably run right or marketed right. You can send us a crashed drive and 9.5 times out of 10 we can recover your data. It could be a $100 million business for us easy.
Q: Does it scale?
A: We’re really good at it. Imagine you had a window that popped up when you bought a new drive that said you could pay $2.50 a month for the right to send a drive into Seagate to have it recovered, would you do that? If you got into a room with 1,000 people, and said how many would say yes for $2.50, you’d probably get 900 people raising their hands.
Q: Yeah, because when it does happen, you’d pay $1,000 to recover your stuff.
A: That’s the thing. Which gets into the question of how do we price it when a person does call. And I think we make that mistake, of over-pricing. Not to be gougers, but because you don’t have the scale that you should have, you tend to charge more on average than you generally need to. I’ve had people calling me in tears, is there anything you can do to save my data. And I’m particularly vulnerable to grandmas calling. All my kids pictures are on here, blah, blah, blah. I’m like, I’ll do it for free. How much would she pay? Well maybe she’d pay $1,000 if she had it. But how much would we charge? This is their life. There is a balance there somewhere. It should be a great business.
Q: How much of the consumer population has a NAS drive right now?
A: Small, but growing. The DAS population is like 20%. So I would say somewhere between 2% and 5%. Again, what market are you looking at – in Palo Alto it might be 20%. I have friends who have more sophisticated networks than we have.
Q: So what’s the driver to get the penetration rate into the double digits?
A: Better marketing, better solutions, more integrated solutions. What are lessons we learn from Apple: one is integrating in hardware and software seems like a pretty good play if you do it well. I would say that for most DAS and NAS solutions to date, no one has really integrated hardware and software, they’ve tended to kludge something together that is not a great user experience. The industry just hasn’t invested a lot of money in that. I think for us, we’re on the front end of doing a fair amount of reorganization to position the company better to execute into some of these opportunities: cloud, consumer, mobility, and those can be hardware plays, software plays and systems plays. The answer is you need better solutions, and you need better marketing support for them.
Q Two last questions. One, what’s with the red string around your wrist.?
A: This is a Buddhist thing. A friend of mine who’s a monk gave it to me. It means long life. You wear them as long as you can. I always wear it.
Q: And then, you have a tattoo on your left ring finger. What’s with that?
A: It’s my wife’s name. I got it because I don’t like things on my hands. Her name’s Agatha. She’s Croatian, and her real name’s Agata, without the H, and my hand’s so small that that’s all they could get on here anyhow. So it worked out really well.
Q: Thanks, Steve.
This article is available online at: